{
  "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": 0.04592296,
      "auxiliary_loss_mlp": 0.0257779,
      "balance_loss_clip": 2.47145319,
      "balance_loss_mlp": 2.09008121,
      "epoch": 0.00012024289063909097,
      "flos": 24932483919360.0,
      "grad_norm": 40.29739113711146,
      "language_loss": 2.5798173,
      "learning_rate": 0.0,
      "loss": 1.90189219,
      "num_input_tokens_seen": 20375,
      "step": 1,
      "time_per_iteration": 14.374552011489868
    },
    {
      "auxiliary_loss_clip": 0.03096418,
      "auxiliary_loss_mlp": 0.01644569,
      "balance_loss_clip": 1.65448713,
      "balance_loss_mlp": 1.32070136,
      "epoch": 0.00024048578127818193,
      "flos": 30664624377600.0,
      "grad_norm": 55.6801668970107,
      "language_loss": 1.89169598,
      "learning_rate": 5.021476677069823e-07,
      "loss": 1.93910587,
      "num_input_tokens_seen": 39035,
      "step": 2,
      "time_per_iteration": 2.6549880504608154
    },
    {
      "auxiliary_loss_clip": 0.03069452,
      "auxiliary_loss_mlp": 0.01668741,
      "balance_loss_clip": 1.65277958,
      "balance_loss_mlp": 1.34735262,
      "epoch": 0.0003607286719172729,
      "flos": 19026227969280.0,
      "grad_norm": 42.95286757526908,
      "language_loss": 1.61725688,
      "learning_rate": 7.958852231401551e-07,
      "loss": 1.66463888,
      "num_input_tokens_seen": 57600,
      "step": 3,
      "time_per_iteration": 2.4245288372039795
    },
    {
      "auxiliary_loss_clip": 0.03077413,
      "auxiliary_loss_mlp": 0.01737639,
      "balance_loss_clip": 1.65165389,
      "balance_loss_mlp": 1.41434348,
      "epoch": 0.00048097156255636386,
      "flos": 19316314206720.0,
      "grad_norm": 38.50878970440572,
      "language_loss": 1.64756238,
      "learning_rate": 1.0042953354139647e-06,
      "loss": 1.6957128,
      "num_input_tokens_seen": 76465,
      "step": 4,
      "time_per_iteration": 2.5279381275177
    },
    {
      "auxiliary_loss_clip": 0.03073745,
      "auxiliary_loss_mlp": 0.01668212,
      "balance_loss_clip": 1.65347123,
      "balance_loss_mlp": 1.35121095,
      "epoch": 0.0006012144531954548,
      "flos": 13991264893440.0,
      "grad_norm": 55.12000627210488,
      "language_loss": 1.93631136,
      "learning_rate": 1.1659507774310057e-06,
      "loss": 1.98373091,
      "num_input_tokens_seen": 94350,
      "step": 5,
      "time_per_iteration": 2.7271769046783447
    },
    {
      "auxiliary_loss_clip": 0.03084172,
      "auxiliary_loss_mlp": 0.01682219,
      "balance_loss_clip": 1.65829468,
      "balance_loss_mlp": 1.36063993,
      "epoch": 0.0007214573438345458,
      "flos": 23148988225920.0,
      "grad_norm": 45.197790089278854,
      "language_loss": 1.60767651,
      "learning_rate": 1.2980328908471373e-06,
      "loss": 1.65534043,
      "num_input_tokens_seen": 114595,
      "step": 6,
      "time_per_iteration": 2.738478660583496
    },
    {
      "auxiliary_loss_clip": 0.03139845,
      "auxiliary_loss_mlp": 0.01602956,
      "balance_loss_clip": 1.79568696,
      "balance_loss_mlp": 1.40154052,
      "epoch": 0.0008417002344736367,
      "flos": 67663246170240.0,
      "grad_norm": 4.592432557004812,
      "language_loss": 0.81522697,
      "learning_rate": 1.4097067265369432e-06,
      "loss": 0.86265498,
      "num_input_tokens_seen": 179590,
      "step": 7,
      "time_per_iteration": 3.174860954284668
    },
    {
      "auxiliary_loss_clip": 0.03050276,
      "auxiliary_loss_mlp": 0.01708411,
      "balance_loss_clip": 1.64231014,
      "balance_loss_mlp": 1.39560556,
      "epoch": 0.0009619431251127277,
      "flos": 21281381504640.0,
      "grad_norm": 41.958804353189905,
      "language_loss": 1.58336723,
      "learning_rate": 1.506443003120947e-06,
      "loss": 1.63095415,
      "num_input_tokens_seen": 195090,
      "step": 8,
      "time_per_iteration": 2.7307381629943848
    },
    {
      "auxiliary_loss_clip": 0.03055876,
      "auxiliary_loss_mlp": 0.01696683,
      "balance_loss_clip": 1.64993596,
      "balance_loss_mlp": 1.3783462,
      "epoch": 0.0010821860157518186,
      "flos": 23331342597120.0,
      "grad_norm": 17.533532447129726,
      "language_loss": 1.47741866,
      "learning_rate": 1.5917704462803102e-06,
      "loss": 1.52494419,
      "num_input_tokens_seen": 211635,
      "step": 9,
      "time_per_iteration": 2.7709126472473145
    },
    {
      "auxiliary_loss_clip": 0.03042157,
      "auxiliary_loss_mlp": 0.01655804,
      "balance_loss_clip": 1.64871716,
      "balance_loss_mlp": 1.33918428,
      "epoch": 0.0012024289063909096,
      "flos": 17010166337280.0,
      "grad_norm": 13.340148248737915,
      "language_loss": 1.53029895,
      "learning_rate": 1.6680984451379884e-06,
      "loss": 1.57727861,
      "num_input_tokens_seen": 224705,
      "step": 10,
      "time_per_iteration": 2.7056832313537598
    },
    {
      "auxiliary_loss_clip": 0.03050718,
      "auxiliary_loss_mlp": 0.01681793,
      "balance_loss_clip": 1.64492559,
      "balance_loss_mlp": 1.37738001,
      "epoch": 0.0013226717970300007,
      "flos": 21288133261440.0,
      "grad_norm": 20.247523118864816,
      "language_loss": 1.32566643,
      "learning_rate": 1.7371455188905097e-06,
      "loss": 1.37299156,
      "num_input_tokens_seen": 244635,
      "step": 11,
      "time_per_iteration": 2.753889322280884
    },
    {
      "auxiliary_loss_clip": 0.03064137,
      "auxiliary_loss_mlp": 0.01704492,
      "balance_loss_clip": 1.64953983,
      "balance_loss_mlp": 1.37661862,
      "epoch": 0.0014429146876690916,
      "flos": 27237884935680.0,
      "grad_norm": 10.611284717823835,
      "language_loss": 1.25264049,
      "learning_rate": 1.8001805585541196e-06,
      "loss": 1.30032682,
      "num_input_tokens_seen": 265765,
      "step": 12,
      "time_per_iteration": 2.8265466690063477
    },
    {
      "auxiliary_loss_clip": 0.03044582,
      "auxiliary_loss_mlp": 0.01666323,
      "balance_loss_clip": 1.64179265,
      "balance_loss_mlp": 1.35885882,
      "epoch": 0.0015631575783081825,
      "flos": 19062174504960.0,
      "grad_norm": 6.641111339435591,
      "language_loss": 1.29169655,
      "learning_rate": 1.8581671739548328e-06,
      "loss": 1.33880568,
      "num_input_tokens_seen": 283500,
      "step": 13,
      "time_per_iteration": 2.7457244396209717
    },
    {
      "auxiliary_loss_clip": 0.03039553,
      "auxiliary_loss_mlp": 0.01619907,
      "balance_loss_clip": 1.63789761,
      "balance_loss_mlp": 1.30767441,
      "epoch": 0.0016834004689472734,
      "flos": 48139473985920.0,
      "grad_norm": 7.628868581592182,
      "language_loss": 1.1354475,
      "learning_rate": 1.9118543942439254e-06,
      "loss": 1.18204224,
      "num_input_tokens_seen": 305685,
      "step": 14,
      "time_per_iteration": 2.956005334854126
    },
    {
      "auxiliary_loss_clip": 0.03015445,
      "auxiliary_loss_mlp": 0.01677622,
      "balance_loss_clip": 1.63282371,
      "balance_loss_mlp": 1.36119246,
      "epoch": 0.0018036433595863645,
      "flos": 34970026314240.0,
      "grad_norm": 6.172883116310707,
      "language_loss": 1.12811732,
      "learning_rate": 1.961836000571161e-06,
      "loss": 1.17504787,
      "num_input_tokens_seen": 327340,
      "step": 15,
      "time_per_iteration": 2.9613378047943115
    },
    {
      "auxiliary_loss_clip": 0.03034383,
      "auxiliary_loss_mlp": 0.01455365,
      "balance_loss_clip": 1.77332854,
      "balance_loss_mlp": 1.26615572,
      "epoch": 0.0019238862502254555,
      "flos": 59768284440960.0,
      "grad_norm": 3.8093761427120203,
      "language_loss": 0.64661431,
      "learning_rate": 2.0085906708279293e-06,
      "loss": 0.69151175,
      "num_input_tokens_seen": 382710,
      "step": 16,
      "time_per_iteration": 6.035049676895142
    },
    {
      "auxiliary_loss_clip": 0.03002347,
      "auxiliary_loss_mlp": 0.01637859,
      "balance_loss_clip": 1.63741493,
      "balance_loss_mlp": 1.32657981,
      "epoch": 0.0020441291408645466,
      "flos": 20814543417600.0,
      "grad_norm": 4.416208460274287,
      "language_loss": 1.16158068,
      "learning_rate": 2.0525099325728135e-06,
      "loss": 1.20798278,
      "num_input_tokens_seen": 400890,
      "step": 17,
      "time_per_iteration": 2.8714346885681152
    },
    {
      "auxiliary_loss_clip": 0.02993946,
      "auxiliary_loss_mlp": 0.01405907,
      "balance_loss_clip": 1.76401675,
      "balance_loss_mlp": 1.21974993,
      "epoch": 0.0021643720315036373,
      "flos": 63857001582720.0,
      "grad_norm": 3.5322948103033753,
      "language_loss": 0.72193605,
      "learning_rate": 2.0939181139872922e-06,
      "loss": 0.76593459,
      "num_input_tokens_seen": 462605,
      "step": 18,
      "time_per_iteration": 3.241466760635376
    },
    {
      "auxiliary_loss_clip": 0.02971844,
      "auxiliary_loss_mlp": 0.01584843,
      "balance_loss_clip": 1.63056493,
      "balance_loss_mlp": 1.28424525,
      "epoch": 0.0022846149221427284,
      "flos": 31284981192960.0,
      "grad_norm": 6.772871332273532,
      "language_loss": 1.01690805,
      "learning_rate": 2.1330868934640175e-06,
      "loss": 1.06247497,
      "num_input_tokens_seen": 483280,
      "step": 19,
      "time_per_iteration": 2.8688836097717285
    },
    {
      "auxiliary_loss_clip": 0.02946278,
      "auxiliary_loss_mlp": 0.01373815,
      "balance_loss_clip": 1.75152946,
      "balance_loss_mlp": 1.19147265,
      "epoch": 0.002404857812781819,
      "flos": 51083648161920.0,
      "grad_norm": 3.577193987194342,
      "language_loss": 0.76498199,
      "learning_rate": 2.170246112844971e-06,
      "loss": 0.80818295,
      "num_input_tokens_seen": 537620,
      "step": 20,
      "time_per_iteration": 2.971428632736206
    },
    {
      "auxiliary_loss_clip": 0.0291929,
      "auxiliary_loss_mlp": 0.0154075,
      "balance_loss_clip": 1.61947107,
      "balance_loss_mlp": 1.24148679,
      "epoch": 0.0025251007034209102,
      "flos": 15815347309440.0,
      "grad_norm": 3.8845593076364615,
      "language_loss": 1.01823497,
      "learning_rate": 2.2055919496770983e-06,
      "loss": 1.06283522,
      "num_input_tokens_seen": 555760,
      "step": 21,
      "time_per_iteration": 2.7682454586029053
    },
    {
      "auxiliary_loss_clip": 0.02904595,
      "auxiliary_loss_mlp": 0.01521818,
      "balance_loss_clip": 1.61508119,
      "balance_loss_mlp": 1.22369981,
      "epoch": 0.0026453435940600014,
      "flos": 37851857458560.0,
      "grad_norm": 6.784273055481209,
      "language_loss": 0.89552367,
      "learning_rate": 2.2392931865974923e-06,
      "loss": 0.93978781,
      "num_input_tokens_seen": 578450,
      "step": 22,
      "time_per_iteration": 2.9105112552642822
    },
    {
      "auxiliary_loss_clip": 0.02865225,
      "auxiliary_loss_mlp": 0.01506225,
      "balance_loss_clip": 1.60591066,
      "balance_loss_mlp": 1.21001339,
      "epoch": 0.002765586484699092,
      "flos": 21141976821120.0,
      "grad_norm": 4.614920141222138,
      "language_loss": 1.0207814,
      "learning_rate": 2.271496085962064e-06,
      "loss": 1.06449592,
      "num_input_tokens_seen": 596145,
      "step": 23,
      "time_per_iteration": 2.733872413635254
    },
    {
      "auxiliary_loss_clip": 0.0283711,
      "auxiliary_loss_mlp": 0.01484282,
      "balance_loss_clip": 1.59264302,
      "balance_loss_mlp": 1.19131327,
      "epoch": 0.002885829375338183,
      "flos": 20667381396480.0,
      "grad_norm": 3.262086141534061,
      "language_loss": 1.0294981,
      "learning_rate": 2.3023282262611022e-06,
      "loss": 1.07271206,
      "num_input_tokens_seen": 614920,
      "step": 24,
      "time_per_iteration": 2.764143705368042
    },
    {
      "auxiliary_loss_clip": 0.02846659,
      "auxiliary_loss_mlp": 0.01484036,
      "balance_loss_clip": 1.60060585,
      "balance_loss_mlp": 1.19888711,
      "epoch": 0.003006072265977274,
      "flos": 34823869873920.0,
      "grad_norm": 2.853033677958846,
      "language_loss": 0.92435491,
      "learning_rate": 2.3319015548620114e-06,
      "loss": 0.96766192,
      "num_input_tokens_seen": 636060,
      "step": 25,
      "time_per_iteration": 2.8501956462860107
    },
    {
      "auxiliary_loss_clip": 0.02804191,
      "auxiliary_loss_mlp": 0.01453565,
      "balance_loss_clip": 1.58819497,
      "balance_loss_mlp": 1.17699957,
      "epoch": 0.003126315156616365,
      "flos": 24422021118720.0,
      "grad_norm": 2.245137244810618,
      "language_loss": 0.93046165,
      "learning_rate": 2.3603148416618152e-06,
      "loss": 0.97303921,
      "num_input_tokens_seen": 655575,
      "step": 26,
      "time_per_iteration": 2.817383289337158
    },
    {
      "auxiliary_loss_clip": 0.0281109,
      "auxiliary_loss_mlp": 0.01434244,
      "balance_loss_clip": 1.58906984,
      "balance_loss_mlp": 1.16168427,
      "epoch": 0.003246558047255456,
      "flos": 23622326674560.0,
      "grad_norm": 2.361945591359085,
      "language_loss": 1.0101757,
      "learning_rate": 2.3876556694204647e-06,
      "loss": 1.05262899,
      "num_input_tokens_seen": 675730,
      "step": 27,
      "time_per_iteration": 2.7356748580932617
    },
    {
      "auxiliary_loss_clip": 0.02771328,
      "auxiliary_loss_mlp": 0.01438783,
      "balance_loss_clip": 1.58208597,
      "balance_loss_mlp": 1.1524899,
      "epoch": 0.003366800937894547,
      "flos": 17820275725440.0,
      "grad_norm": 2.9818497678423515,
      "language_loss": 0.90620762,
      "learning_rate": 2.414002061950908e-06,
      "loss": 0.94830871,
      "num_input_tokens_seen": 694605,
      "step": 28,
      "time_per_iteration": 2.7878856658935547
    },
    {
      "auxiliary_loss_clip": 0.02754835,
      "auxiliary_loss_mlp": 0.01410376,
      "balance_loss_clip": 1.57468307,
      "balance_loss_mlp": 1.14201212,
      "epoch": 0.003487043828533638,
      "flos": 24426115269120.0,
      "grad_norm": 2.216056068974684,
      "language_loss": 0.99936974,
      "learning_rate": 2.4394238264681557e-06,
      "loss": 1.04102194,
      "num_input_tokens_seen": 714340,
      "step": 29,
      "time_per_iteration": 2.764742136001587
    },
    {
      "auxiliary_loss_clip": 0.02729587,
      "auxiliary_loss_mlp": 0.01408997,
      "balance_loss_clip": 1.56779027,
      "balance_loss_mlp": 1.13529277,
      "epoch": 0.003607286719172729,
      "flos": 26140311002880.0,
      "grad_norm": 2.6510685202549302,
      "language_loss": 0.9945575,
      "learning_rate": 2.4639836682781433e-06,
      "loss": 1.03594339,
      "num_input_tokens_seen": 734470,
      "step": 30,
      "time_per_iteration": 2.797646999359131
    },
    {
      "auxiliary_loss_clip": 0.02743581,
      "auxiliary_loss_mlp": 0.01397551,
      "balance_loss_clip": 1.58156228,
      "balance_loss_mlp": 1.11698043,
      "epoch": 0.00372752960981182,
      "flos": 20593082113920.0,
      "grad_norm": 2.4885327003664335,
      "language_loss": 1.00256562,
      "learning_rate": 2.487738122623307e-06,
      "loss": 1.04397702,
      "num_input_tokens_seen": 753380,
      "step": 31,
      "time_per_iteration": 2.800851583480835
    },
    {
      "auxiliary_loss_clip": 0.02704224,
      "auxiliary_loss_mlp": 0.01376455,
      "balance_loss_clip": 1.56545734,
      "balance_loss_mlp": 1.10408592,
      "epoch": 0.003847772500450911,
      "flos": 22674608282880.0,
      "grad_norm": 2.5339218763242557,
      "language_loss": 0.98898959,
      "learning_rate": 2.510738338534912e-06,
      "loss": 1.02979636,
      "num_input_tokens_seen": 772105,
      "step": 32,
      "time_per_iteration": 2.7398064136505127
    },
    {
      "auxiliary_loss_clip": 0.02566986,
      "auxiliary_loss_mlp": 0.01358765,
      "balance_loss_clip": 1.52260804,
      "balance_loss_mlp": 1.09135485,
      "epoch": 0.003968015391090002,
      "flos": 17967796882560.0,
      "grad_norm": 2.348056889898114,
      "language_loss": 1.02350163,
      "learning_rate": 2.5330307420306648e-06,
      "loss": 1.06275904,
      "num_input_tokens_seen": 788955,
      "step": 33,
      "time_per_iteration": 2.781377077102661
    },
    {
      "auxiliary_loss_clip": 0.02524188,
      "auxiliary_loss_mlp": 0.01344382,
      "balance_loss_clip": 1.51670671,
      "balance_loss_mlp": 1.09985983,
      "epoch": 0.004088258281729093,
      "flos": 27304103658240.0,
      "grad_norm": 2.2115796200789326,
      "language_loss": 0.88163292,
      "learning_rate": 2.554657600279796e-06,
      "loss": 0.92031866,
      "num_input_tokens_seen": 810230,
      "step": 34,
      "time_per_iteration": 2.870767593383789
    },
    {
      "auxiliary_loss_clip": 0.02503936,
      "auxiliary_loss_mlp": 0.01325464,
      "balance_loss_clip": 1.50777054,
      "balance_loss_mlp": 1.07312143,
      "epoch": 0.004208501172368184,
      "flos": 23258587599360.0,
      "grad_norm": 2.116867061559311,
      "language_loss": 1.03360844,
      "learning_rate": 2.5756575039679493e-06,
      "loss": 1.07190251,
      "num_input_tokens_seen": 829780,
      "step": 35,
      "time_per_iteration": 2.9073262214660645
    },
    {
      "auxiliary_loss_clip": 0.02465737,
      "auxiliary_loss_mlp": 0.01350687,
      "balance_loss_clip": 1.49645627,
      "balance_loss_mlp": 1.09910738,
      "epoch": 0.0043287440630072746,
      "flos": 17312104062720.0,
      "grad_norm": 2.001889173623935,
      "language_loss": 0.94864929,
      "learning_rate": 2.5960657816942747e-06,
      "loss": 0.98681355,
      "num_input_tokens_seen": 848695,
      "step": 36,
      "time_per_iteration": 2.8230321407318115
    },
    {
      "auxiliary_loss_clip": 0.02309714,
      "auxiliary_loss_mlp": 0.01395315,
      "balance_loss_clip": 1.57972264,
      "balance_loss_mlp": 1.25188243,
      "epoch": 0.004448986953646365,
      "flos": 53092491160320.0,
      "grad_norm": 1.3853112564289791,
      "language_loss": 0.60955733,
      "learning_rate": 2.6159148575788668e-06,
      "loss": 0.64660764,
      "num_input_tokens_seen": 906730,
      "step": 37,
      "time_per_iteration": 3.3263089656829834
    },
    {
      "auxiliary_loss_clip": 0.02414137,
      "auxiliary_loss_mlp": 0.01360482,
      "balance_loss_clip": 1.4858557,
      "balance_loss_mlp": 1.11061943,
      "epoch": 0.004569229844285457,
      "flos": 13444165866240.0,
      "grad_norm": 2.6112171088493428,
      "language_loss": 0.98623204,
      "learning_rate": 2.635234561171e-06,
      "loss": 1.02397823,
      "num_input_tokens_seen": 925125,
      "step": 38,
      "time_per_iteration": 2.9241151809692383
    },
    {
      "auxiliary_loss_clip": 0.02389896,
      "auxiliary_loss_mlp": 0.01325395,
      "balance_loss_clip": 1.47801161,
      "balance_loss_mlp": 1.09098196,
      "epoch": 0.0046894727349245475,
      "flos": 16209609966720.0,
      "grad_norm": 2.2277901119292878,
      "language_loss": 0.93951499,
      "learning_rate": 2.6540523970949877e-06,
      "loss": 0.97666794,
      "num_input_tokens_seen": 939970,
      "step": 39,
      "time_per_iteration": 2.7007386684417725
    },
    {
      "auxiliary_loss_clip": 0.02360213,
      "auxiliary_loss_mlp": 0.01336131,
      "balance_loss_clip": 1.47734702,
      "balance_loss_mlp": 1.09942937,
      "epoch": 0.004809715625563638,
      "flos": 23914244505600.0,
      "grad_norm": 2.874301601127075,
      "language_loss": 0.9243241,
      "learning_rate": 2.6723937805519533e-06,
      "loss": 0.96128762,
      "num_input_tokens_seen": 957470,
      "step": 40,
      "time_per_iteration": 2.8220551013946533
    },
    {
      "auxiliary_loss_clip": 0.02347358,
      "auxiliary_loss_mlp": 0.01303545,
      "balance_loss_clip": 1.46529198,
      "balance_loss_mlp": 1.07809639,
      "epoch": 0.00492995851620273,
      "flos": 20773030273920.0,
      "grad_norm": 2.0944599060714157,
      "language_loss": 0.92880535,
      "learning_rate": 2.690282243737839e-06,
      "loss": 0.96531445,
      "num_input_tokens_seen": 976405,
      "step": 41,
      "time_per_iteration": 2.776846408843994
    },
    {
      "auxiliary_loss_clip": 0.02311401,
      "auxiliary_loss_mlp": 0.01324455,
      "balance_loss_clip": 1.45360422,
      "balance_loss_mlp": 1.09442842,
      "epoch": 0.0050502014068418205,
      "flos": 20338655103360.0,
      "grad_norm": 2.973442025194003,
      "language_loss": 0.99373257,
      "learning_rate": 2.7077396173840807e-06,
      "loss": 1.03009105,
      "num_input_tokens_seen": 994690,
      "step": 42,
      "time_per_iteration": 3.7124176025390625
    },
    {
      "auxiliary_loss_clip": 0.0228694,
      "auxiliary_loss_mlp": 0.01315059,
      "balance_loss_clip": 1.44621408,
      "balance_loss_mlp": 1.09476042,
      "epoch": 0.005170444297480911,
      "flos": 25994872834560.0,
      "grad_norm": 2.2046987059557805,
      "language_loss": 0.92603016,
      "learning_rate": 2.7247861909342594e-06,
      "loss": 0.9620502,
      "num_input_tokens_seen": 1015615,
      "step": 43,
      "time_per_iteration": 3.8323168754577637
    },
    {
      "auxiliary_loss_clip": 0.02283406,
      "auxiliary_loss_mlp": 0.01312388,
      "balance_loss_clip": 1.44547439,
      "balance_loss_mlp": 1.09571314,
      "epoch": 0.005290687188120003,
      "flos": 20954055841920.0,
      "grad_norm": 2.339493103770965,
      "language_loss": 0.82909518,
      "learning_rate": 2.7414408543044743e-06,
      "loss": 0.86505306,
      "num_input_tokens_seen": 1031255,
      "step": 44,
      "time_per_iteration": 2.725626230239868
    },
    {
      "auxiliary_loss_clip": 0.02231763,
      "auxiliary_loss_mlp": 0.01333082,
      "balance_loss_clip": 1.4303937,
      "balance_loss_mlp": 1.1125927,
      "epoch": 0.005410930078759093,
      "flos": 15851401585920.0,
      "grad_norm": 13.295874996940988,
      "language_loss": 0.79185271,
      "learning_rate": 2.7577212237113157e-06,
      "loss": 0.82750118,
      "num_input_tokens_seen": 1048295,
      "step": 45,
      "time_per_iteration": 2.7430593967437744
    },
    {
      "auxiliary_loss_clip": 0.02218978,
      "auxiliary_loss_mlp": 0.0131081,
      "balance_loss_clip": 1.42451239,
      "balance_loss_mlp": 1.09604239,
      "epoch": 0.005531172969398184,
      "flos": 21104988791040.0,
      "grad_norm": 1.9554625033166702,
      "language_loss": 1.04178882,
      "learning_rate": 2.7736437536690466e-06,
      "loss": 1.07708669,
      "num_input_tokens_seen": 1067925,
      "step": 46,
      "time_per_iteration": 2.7796523571014404
    },
    {
      "auxiliary_loss_clip": 0.02207859,
      "auxiliary_loss_mlp": 0.01275808,
      "balance_loss_clip": 1.42461169,
      "balance_loss_mlp": 1.06657219,
      "epoch": 0.005651415860037276,
      "flos": 20844887431680.0,
      "grad_norm": 2.554091690826338,
      "language_loss": 1.07793164,
      "learning_rate": 2.789223836941131e-06,
      "loss": 1.11276841,
      "num_input_tokens_seen": 1088060,
      "step": 47,
      "time_per_iteration": 2.7990071773529053
    },
    {
      "auxiliary_loss_clip": 0.02172731,
      "auxiliary_loss_mlp": 0.01287001,
      "balance_loss_clip": 1.41169548,
      "balance_loss_mlp": 1.08186591,
      "epoch": 0.005771658750676366,
      "flos": 13260195383040.0,
      "grad_norm": 2.4392611130646276,
      "language_loss": 1.08378434,
      "learning_rate": 2.8044758939680847e-06,
      "loss": 1.11838174,
      "num_input_tokens_seen": 1104130,
      "step": 48,
      "time_per_iteration": 2.7202019691467285
    },
    {
      "auxiliary_loss_clip": 0.02147062,
      "auxiliary_loss_mlp": 0.01282087,
      "balance_loss_clip": 1.41173136,
      "balance_loss_mlp": 1.07561612,
      "epoch": 0.005891901641315457,
      "flos": 24425396997120.0,
      "grad_norm": 3.2176352129614005,
      "language_loss": 1.01912093,
      "learning_rate": 2.8194134530738863e-06,
      "loss": 1.05341244,
      "num_input_tokens_seen": 1122900,
      "step": 49,
      "time_per_iteration": 2.7268407344818115
    },
    {
      "auxiliary_loss_clip": 0.0213871,
      "auxiliary_loss_mlp": 0.01289959,
      "balance_loss_clip": 1.40585709,
      "balance_loss_mlp": 1.09369338,
      "epoch": 0.006012144531954548,
      "flos": 23076197314560.0,
      "grad_norm": 3.5029839325988377,
      "language_loss": 0.90221584,
      "learning_rate": 2.834049222568994e-06,
      "loss": 0.93650252,
      "num_input_tokens_seen": 1140250,
      "step": 50,
      "time_per_iteration": 2.761090040206909
    },
    {
      "auxiliary_loss_clip": 0.02134155,
      "auxiliary_loss_mlp": 0.01256345,
      "balance_loss_clip": 1.40252638,
      "balance_loss_mlp": 1.06503749,
      "epoch": 0.006132387422593639,
      "flos": 22528775064960.0,
      "grad_norm": 9.8036769161449,
      "language_loss": 0.9260208,
      "learning_rate": 2.848395155712969e-06,
      "loss": 0.95992577,
      "num_input_tokens_seen": 1160470,
      "step": 51,
      "time_per_iteration": 2.7344744205474854
    },
    {
      "auxiliary_loss_clip": 0.02118259,
      "auxiliary_loss_mlp": 0.01293493,
      "balance_loss_clip": 1.40352023,
      "balance_loss_mlp": 1.09827614,
      "epoch": 0.00625263031323273,
      "flos": 27628340751360.0,
      "grad_norm": 2.1406021930393107,
      "language_loss": 0.97542429,
      "learning_rate": 2.8624625093687977e-06,
      "loss": 1.00954175,
      "num_input_tokens_seen": 1177605,
      "step": 52,
      "time_per_iteration": 2.7855544090270996
    },
    {
      "auxiliary_loss_clip": 0.02100484,
      "auxiliary_loss_mlp": 0.01261326,
      "balance_loss_clip": 1.39333069,
      "balance_loss_mlp": 1.07736182,
      "epoch": 0.006372873203871821,
      "flos": 23110671392640.0,
      "grad_norm": 2.110693769211949,
      "language_loss": 0.88798487,
      "learning_rate": 2.876261897070029e-06,
      "loss": 0.92160296,
      "num_input_tokens_seen": 1197735,
      "step": 53,
      "time_per_iteration": 2.7590248584747314
    },
    {
      "auxiliary_loss_clip": 0.02099513,
      "auxiliary_loss_mlp": 0.01280276,
      "balance_loss_clip": 1.39740086,
      "balance_loss_mlp": 1.09364152,
      "epoch": 0.006493116094510912,
      "flos": 22856028900480.0,
      "grad_norm": 10.880884874730942,
      "language_loss": 0.92517531,
      "learning_rate": 2.889803337127447e-06,
      "loss": 0.95897317,
      "num_input_tokens_seen": 1216335,
      "step": 54,
      "time_per_iteration": 2.7762508392333984
    },
    {
      "auxiliary_loss_clip": 0.02069932,
      "auxiliary_loss_mlp": 0.0130026,
      "balance_loss_clip": 1.38596392,
      "balance_loss_mlp": 1.10323107,
      "epoch": 0.006613358985150003,
      "flos": 23071708114560.0,
      "grad_norm": 15.491268485458184,
      "language_loss": 0.84593523,
      "learning_rate": 2.903096296321516e-06,
      "loss": 0.87963712,
      "num_input_tokens_seen": 1234480,
      "step": 55,
      "time_per_iteration": 2.7236132621765137
    },
    {
      "auxiliary_loss_clip": 0.02071536,
      "auxiliary_loss_mlp": 0.01251565,
      "balance_loss_clip": 1.38789153,
      "balance_loss_mlp": 1.07503986,
      "epoch": 0.006733601875789094,
      "flos": 26537662229760.0,
      "grad_norm": 1.935920326197683,
      "language_loss": 0.917858,
      "learning_rate": 2.9161497296578907e-06,
      "loss": 0.95108896,
      "num_input_tokens_seen": 1253870,
      "step": 56,
      "time_per_iteration": 2.827345371246338
    },
    {
      "auxiliary_loss_clip": 0.02053292,
      "auxiliary_loss_mlp": 0.012576,
      "balance_loss_clip": 1.38166904,
      "balance_loss_mlp": 1.07859492,
      "epoch": 0.006853844766428185,
      "flos": 15523178083200.0,
      "grad_norm": 2.274470708084189,
      "language_loss": 0.86057323,
      "learning_rate": 2.928972116604173e-06,
      "loss": 0.89368212,
      "num_input_tokens_seen": 1270145,
      "step": 57,
      "time_per_iteration": 2.673583745956421
    },
    {
      "auxiliary_loss_clip": 0.02025722,
      "auxiliary_loss_mlp": 0.01235108,
      "balance_loss_clip": 1.37306416,
      "balance_loss_mlp": 1.06678486,
      "epoch": 0.006974087657067276,
      "flos": 24243760897920.0,
      "grad_norm": 2.1835296283127246,
      "language_loss": 1.0203737,
      "learning_rate": 2.9415714941751377e-06,
      "loss": 1.05298209,
      "num_input_tokens_seen": 1291365,
      "step": 58,
      "time_per_iteration": 2.835153818130493
    },
    {
      "auxiliary_loss_clip": 0.02042916,
      "auxiliary_loss_mlp": 0.01261387,
      "balance_loss_clip": 1.37508285,
      "balance_loss_mlp": 1.09201431,
      "epoch": 0.007094330547706367,
      "flos": 25772513690880.0,
      "grad_norm": 2.078204932381026,
      "language_loss": 0.93671525,
      "learning_rate": 2.9539554871897396e-06,
      "loss": 0.96975827,
      "num_input_tokens_seen": 1311535,
      "step": 59,
      "time_per_iteration": 2.8793962001800537
    },
    {
      "auxiliary_loss_clip": 0.02011843,
      "auxiliary_loss_mlp": 0.012433,
      "balance_loss_clip": 1.36773133,
      "balance_loss_mlp": 1.07840967,
      "epoch": 0.007214573438345458,
      "flos": 21319015979520.0,
      "grad_norm": 2.062275396970176,
      "language_loss": 0.97577798,
      "learning_rate": 2.9661313359851253e-06,
      "loss": 1.00832939,
      "num_input_tokens_seen": 1329420,
      "step": 60,
      "time_per_iteration": 2.786170482635498
    },
    {
      "auxiliary_loss_clip": 0.01988771,
      "auxiliary_loss_mlp": 0.01238239,
      "balance_loss_clip": 1.36445343,
      "balance_loss_mlp": 1.07764041,
      "epoch": 0.007334816328984549,
      "flos": 24937088192640.0,
      "grad_norm": 1.960074337230255,
      "language_loss": 0.93954498,
      "learning_rate": 2.978105921839922e-06,
      "loss": 0.97181499,
      "num_input_tokens_seen": 1349965,
      "step": 61,
      "time_per_iteration": 2.8105406761169434
    },
    {
      "auxiliary_loss_clip": 0.01975841,
      "auxiliary_loss_mlp": 0.01251134,
      "balance_loss_clip": 1.36125982,
      "balance_loss_mlp": 1.08986783,
      "epoch": 0.00745505921962364,
      "flos": 18510586277760.0,
      "grad_norm": 3.096047863164156,
      "language_loss": 0.72212291,
      "learning_rate": 2.9898857903302893e-06,
      "loss": 0.75439268,
      "num_input_tokens_seen": 1368915,
      "step": 62,
      "time_per_iteration": 2.8966891765594482
    },
    {
      "auxiliary_loss_clip": 0.01978552,
      "auxiliary_loss_mlp": 0.01256822,
      "balance_loss_clip": 1.35914779,
      "balance_loss_mlp": 1.09135938,
      "epoch": 0.007575302110262731,
      "flos": 18477656484480.0,
      "grad_norm": 3.0685303188225173,
      "language_loss": 0.88111019,
      "learning_rate": 3.001477172817253e-06,
      "loss": 0.91346395,
      "num_input_tokens_seen": 1386805,
      "step": 63,
      "time_per_iteration": 2.7652645111083984
    },
    {
      "auxiliary_loss_clip": 0.01953508,
      "auxiliary_loss_mlp": 0.01227687,
      "balance_loss_clip": 1.3510437,
      "balance_loss_mlp": 1.07834172,
      "epoch": 0.007695545000901822,
      "flos": 24973178382720.0,
      "grad_norm": 2.437845911525819,
      "language_loss": 0.96231973,
      "learning_rate": 3.012886006241894e-06,
      "loss": 0.99413162,
      "num_input_tokens_seen": 1406190,
      "step": 64,
      "time_per_iteration": 2.8037002086639404
    },
    {
      "auxiliary_loss_clip": 0.01959035,
      "auxiliary_loss_mlp": 0.01231211,
      "balance_loss_clip": 1.35284066,
      "balance_loss_mlp": 1.07423592,
      "epoch": 0.007815787891540913,
      "flos": 21324223451520.0,
      "grad_norm": 2.0856738963629144,
      "language_loss": 0.88041037,
      "learning_rate": 3.0241179513858383e-06,
      "loss": 0.91231287,
      "num_input_tokens_seen": 1425500,
      "step": 65,
      "time_per_iteration": 2.777071714401245
    },
    {
      "auxiliary_loss_clip": 0.01939395,
      "auxiliary_loss_mlp": 0.01255336,
      "balance_loss_clip": 1.34210134,
      "balance_loss_mlp": 1.09178114,
      "epoch": 0.007936030782180003,
      "flos": 21575777374080.0,
      "grad_norm": 2.4639792510238756,
      "language_loss": 0.87789875,
      "learning_rate": 3.035178409737647e-06,
      "loss": 0.90984601,
      "num_input_tokens_seen": 1442950,
      "step": 66,
      "time_per_iteration": 2.750734567642212
    },
    {
      "auxiliary_loss_clip": 0.01918592,
      "auxiliary_loss_mlp": 0.01219663,
      "balance_loss_clip": 1.33552456,
      "balance_loss_mlp": 1.08376431,
      "epoch": 0.008056273672819095,
      "flos": 20120785159680.0,
      "grad_norm": 3.231532183327812,
      "language_loss": 0.8841995,
      "learning_rate": 3.046072539090907e-06,
      "loss": 0.915582,
      "num_input_tokens_seen": 1460915,
      "step": 67,
      "time_per_iteration": 2.7618565559387207
    },
    {
      "auxiliary_loss_clip": 0.01913889,
      "auxiliary_loss_mlp": 0.01216125,
      "balance_loss_clip": 1.33542013,
      "balance_loss_mlp": 1.07412243,
      "epoch": 0.008176516563458186,
      "flos": 18333116156160.0,
      "grad_norm": 2.3801800167248297,
      "language_loss": 1.04924071,
      "learning_rate": 3.056805267986779e-06,
      "loss": 1.0805409,
      "num_input_tokens_seen": 1478385,
      "step": 68,
      "time_per_iteration": 2.741159677505493
    },
    {
      "auxiliary_loss_clip": 0.01895641,
      "auxiliary_loss_mlp": 0.01221655,
      "balance_loss_clip": 1.32847881,
      "balance_loss_mlp": 1.08289504,
      "epoch": 0.008296759454097276,
      "flos": 21872076664320.0,
      "grad_norm": 2.3189305943746352,
      "language_loss": 0.95458227,
      "learning_rate": 3.0673813091022194e-06,
      "loss": 0.98575521,
      "num_input_tokens_seen": 1497605,
      "step": 69,
      "time_per_iteration": 3.7655205726623535
    },
    {
      "auxiliary_loss_clip": 0.01756409,
      "auxiliary_loss_mlp": 0.01182128,
      "balance_loss_clip": 1.35445309,
      "balance_loss_mlp": 1.11117446,
      "epoch": 0.008417002344736368,
      "flos": 63408228036480.0,
      "grad_norm": 1.304992641173884,
      "language_loss": 0.62112987,
      "learning_rate": 3.0778051716749317e-06,
      "loss": 0.6505152,
      "num_input_tokens_seen": 1561150,
      "step": 70,
      "time_per_iteration": 5.136059522628784
    },
    {
      "auxiliary_loss_clip": 0.01866457,
      "auxiliary_loss_mlp": 0.01211963,
      "balance_loss_clip": 1.31028092,
      "balance_loss_mlp": 1.07539678,
      "epoch": 0.008537245235375458,
      "flos": 22966454286720.0,
      "grad_norm": 2.0659865904691372,
      "language_loss": 0.90387511,
      "learning_rate": 3.0880811730470094e-06,
      "loss": 0.93465924,
      "num_input_tokens_seen": 1580605,
      "step": 71,
      "time_per_iteration": 2.769749641418457
    },
    {
      "auxiliary_loss_clip": 0.01727474,
      "auxiliary_loss_mlp": 0.0113653,
      "balance_loss_clip": 1.33663738,
      "balance_loss_mlp": 1.07091737,
      "epoch": 0.008657488126014549,
      "flos": 61984046712960.0,
      "grad_norm": 1.2001261533117524,
      "language_loss": 0.585756,
      "learning_rate": 3.098213449401257e-06,
      "loss": 0.61439604,
      "num_input_tokens_seen": 1647535,
      "step": 72,
      "time_per_iteration": 3.220968008041382
    },
    {
      "auxiliary_loss_clip": 0.01856456,
      "auxiliary_loss_mlp": 0.01209761,
      "balance_loss_clip": 1.30759025,
      "balance_loss_mlp": 1.0795846,
      "epoch": 0.00877773101665364,
      "flos": 30296791152000.0,
      "grad_norm": 2.10631929945268,
      "language_loss": 0.98958266,
      "learning_rate": 3.1082059657570015e-06,
      "loss": 1.02024484,
      "num_input_tokens_seen": 1666770,
      "step": 73,
      "time_per_iteration": 2.7994461059570312
    },
    {
      "auxiliary_loss_clip": 0.01827816,
      "auxiliary_loss_mlp": 0.01197105,
      "balance_loss_clip": 1.29851937,
      "balance_loss_mlp": 1.06444919,
      "epoch": 0.00889797390729273,
      "flos": 23514056104320.0,
      "grad_norm": 2.293994013425572,
      "language_loss": 0.96654904,
      "learning_rate": 3.1180625252858496e-06,
      "loss": 0.99679822,
      "num_input_tokens_seen": 1685200,
      "step": 74,
      "time_per_iteration": 2.8356196880340576
    },
    {
      "auxiliary_loss_clip": 0.01812057,
      "auxiliary_loss_mlp": 0.01206743,
      "balance_loss_clip": 1.28967273,
      "balance_loss_mlp": 1.08190703,
      "epoch": 0.009018216797931822,
      "flos": 23075838178560.0,
      "grad_norm": 2.856104036128846,
      "language_loss": 0.80339921,
      "learning_rate": 3.1277867780021663e-06,
      "loss": 0.83358717,
      "num_input_tokens_seen": 1701835,
      "step": 75,
      "time_per_iteration": 2.7326548099517822
    },
    {
      "auxiliary_loss_clip": 0.01790701,
      "auxiliary_loss_mlp": 0.01174466,
      "balance_loss_clip": 1.28236783,
      "balance_loss_mlp": 1.06069303,
      "epoch": 0.009138459688570914,
      "flos": 15918877284480.0,
      "grad_norm": 2.173902665009442,
      "language_loss": 0.95672882,
      "learning_rate": 3.1373822288779824e-06,
      "loss": 0.98638052,
      "num_input_tokens_seen": 1718415,
      "step": 76,
      "time_per_iteration": 2.7821381092071533
    },
    {
      "auxiliary_loss_clip": 0.01788383,
      "auxiliary_loss_mlp": 0.01209506,
      "balance_loss_clip": 1.28346872,
      "balance_loss_mlp": 1.08838964,
      "epoch": 0.009258702579210003,
      "flos": 27016531372800.0,
      "grad_norm": 2.055338045369414,
      "language_loss": 0.79598349,
      "learning_rate": 3.1468522454274533e-06,
      "loss": 0.82596231,
      "num_input_tokens_seen": 1738770,
      "step": 77,
      "time_per_iteration": 2.9289002418518066
    },
    {
      "auxiliary_loss_clip": 0.01776052,
      "auxiliary_loss_mlp": 0.01188571,
      "balance_loss_clip": 1.27662373,
      "balance_loss_mlp": 1.06993425,
      "epoch": 0.009378945469849095,
      "flos": 26903196984960.0,
      "grad_norm": 2.1475058952604544,
      "language_loss": 0.91680169,
      "learning_rate": 3.15620006480197e-06,
      "loss": 0.94644791,
      "num_input_tokens_seen": 1758040,
      "step": 78,
      "time_per_iteration": 2.800227403640747
    },
    {
      "auxiliary_loss_clip": 0.01772259,
      "auxiliary_loss_mlp": 0.01184862,
      "balance_loss_clip": 1.27350307,
      "balance_loss_mlp": 1.06679726,
      "epoch": 0.009499188360488187,
      "flos": 35694236327040.0,
      "grad_norm": 3.467447797058441,
      "language_loss": 0.74830413,
      "learning_rate": 3.1654288004333087e-06,
      "loss": 0.77787536,
      "num_input_tokens_seen": 1776705,
      "step": 79,
      "time_per_iteration": 2.8954670429229736
    },
    {
      "auxiliary_loss_clip": 0.01750833,
      "auxiliary_loss_mlp": 0.01170944,
      "balance_loss_clip": 1.26733017,
      "balance_loss_mlp": 1.06313097,
      "epoch": 0.009619431251127276,
      "flos": 21503201944320.0,
      "grad_norm": 2.9759605162559244,
      "language_loss": 0.76237071,
      "learning_rate": 3.1745414482589353e-06,
      "loss": 0.79158843,
      "num_input_tokens_seen": 1795915,
      "step": 80,
      "time_per_iteration": 2.876530170440674
    },
    {
      "auxiliary_loss_clip": 0.01739727,
      "auxiliary_loss_mlp": 0.01166875,
      "balance_loss_clip": 1.26259518,
      "balance_loss_mlp": 1.05758429,
      "epoch": 0.009739674141766368,
      "flos": 17421056991360.0,
      "grad_norm": 2.346682078673733,
      "language_loss": 0.86991739,
      "learning_rate": 3.1835408925606204e-06,
      "loss": 0.89898342,
      "num_input_tokens_seen": 1814055,
      "step": 81,
      "time_per_iteration": 2.7230310440063477
    },
    {
      "auxiliary_loss_clip": 0.01718669,
      "auxiliary_loss_mlp": 0.01181892,
      "balance_loss_clip": 1.25507689,
      "balance_loss_mlp": 1.0737927,
      "epoch": 0.00985991703240546,
      "flos": 27527109246720.0,
      "grad_norm": 2.2820651328969648,
      "language_loss": 0.89250356,
      "learning_rate": 3.1924299114448214e-06,
      "loss": 0.92150915,
      "num_input_tokens_seen": 1834535,
      "step": 82,
      "time_per_iteration": 2.7493081092834473
    },
    {
      "auxiliary_loss_clip": 0.01728417,
      "auxiliary_loss_mlp": 0.01182177,
      "balance_loss_clip": 1.26007283,
      "balance_loss_mlp": 1.07507896,
      "epoch": 0.00998015992304455,
      "flos": 13808084509440.0,
      "grad_norm": 2.2796700817491677,
      "language_loss": 0.83249819,
      "learning_rate": 3.2012111819909055e-06,
      "loss": 0.86160409,
      "num_input_tokens_seen": 1851865,
      "step": 83,
      "time_per_iteration": 2.730642080307007
    },
    {
      "auxiliary_loss_clip": 0.01712914,
      "auxiliary_loss_mlp": 0.01177135,
      "balance_loss_clip": 1.25123668,
      "balance_loss_mlp": 1.07285023,
      "epoch": 0.010100402813683641,
      "flos": 20191385341440.0,
      "grad_norm": 2.992778898784648,
      "language_loss": 0.9490639,
      "learning_rate": 3.2098872850910627e-06,
      "loss": 0.9779644,
      "num_input_tokens_seen": 1868540,
      "step": 84,
      "time_per_iteration": 2.692213535308838
    },
    {
      "auxiliary_loss_clip": 0.01710508,
      "auxiliary_loss_mlp": 0.0117688,
      "balance_loss_clip": 1.25284529,
      "balance_loss_mlp": 1.07645774,
      "epoch": 0.010220645704322733,
      "flos": 17201642762880.0,
      "grad_norm": 1.9291513336701338,
      "language_loss": 0.89222872,
      "learning_rate": 3.2184607100038194e-06,
      "loss": 0.92110258,
      "num_input_tokens_seen": 1887180,
      "step": 85,
      "time_per_iteration": 2.7914376258850098
    },
    {
      "auxiliary_loss_clip": 0.0170872,
      "auxiliary_loss_mlp": 0.01176829,
      "balance_loss_clip": 1.25333369,
      "balance_loss_mlp": 1.07812357,
      "epoch": 0.010340888594961822,
      "flos": 21470415805440.0,
      "grad_norm": 2.152205081445866,
      "language_loss": 0.93045104,
      "learning_rate": 3.2269338586412414e-06,
      "loss": 0.95930654,
      "num_input_tokens_seen": 1904765,
      "step": 86,
      "time_per_iteration": 2.8686442375183105
    },
    {
      "auxiliary_loss_clip": 0.01694194,
      "auxiliary_loss_mlp": 0.0116713,
      "balance_loss_clip": 1.24550986,
      "balance_loss_mlp": 1.07428944,
      "epoch": 0.010461131485600914,
      "flos": 23002831785600.0,
      "grad_norm": 3.6626671763645975,
      "language_loss": 0.96531081,
      "learning_rate": 3.2353090496083106e-06,
      "loss": 0.99392402,
      "num_input_tokens_seen": 1922600,
      "step": 87,
      "time_per_iteration": 2.7136776447296143
    },
    {
      "auxiliary_loss_clip": 0.01670913,
      "auxiliary_loss_mlp": 0.0116547,
      "balance_loss_clip": 1.23564577,
      "balance_loss_mlp": 1.07854223,
      "epoch": 0.010581374376240005,
      "flos": 33546850571520.0,
      "grad_norm": 2.050048570668282,
      "language_loss": 0.81170756,
      "learning_rate": 3.2435885220114572e-06,
      "loss": 0.84007138,
      "num_input_tokens_seen": 1943950,
      "step": 88,
      "time_per_iteration": 2.8408892154693604
    },
    {
      "auxiliary_loss_clip": 0.01678007,
      "auxiliary_loss_mlp": 0.01154193,
      "balance_loss_clip": 1.24108863,
      "balance_loss_mlp": 1.06364202,
      "epoch": 0.010701617266879095,
      "flos": 21763087822080.0,
      "grad_norm": 2.313678325068964,
      "language_loss": 0.93963873,
      "learning_rate": 3.2517744390519113e-06,
      "loss": 0.96796077,
      "num_input_tokens_seen": 1962815,
      "step": 89,
      "time_per_iteration": 2.752561092376709
    },
    {
      "auxiliary_loss_clip": 0.01663841,
      "auxiliary_loss_mlp": 0.0115339,
      "balance_loss_clip": 1.22724867,
      "balance_loss_mlp": 1.06789351,
      "epoch": 0.010821860157518187,
      "flos": 19060199256960.0,
      "grad_norm": 2.0189697171168133,
      "language_loss": 0.75111318,
      "learning_rate": 3.259868891418298e-06,
      "loss": 0.77928555,
      "num_input_tokens_seen": 1980580,
      "step": 90,
      "time_per_iteration": 2.787611246109009
    },
    {
      "auxiliary_loss_clip": 0.01671896,
      "auxiliary_loss_mlp": 0.01187044,
      "balance_loss_clip": 1.23656821,
      "balance_loss_mlp": 1.09806633,
      "epoch": 0.010942103048157278,
      "flos": 25447378757760.0,
      "grad_norm": 1.9612183601279998,
      "language_loss": 0.85066247,
      "learning_rate": 3.2678739004917757e-06,
      "loss": 0.8792519,
      "num_input_tokens_seen": 2000315,
      "step": 91,
      "time_per_iteration": 2.7469875812530518
    },
    {
      "auxiliary_loss_clip": 0.01654193,
      "auxiliary_loss_mlp": 0.01166394,
      "balance_loss_clip": 1.22933662,
      "balance_loss_mlp": 1.0830425,
      "epoch": 0.011062345938796368,
      "flos": 27493928058240.0,
      "grad_norm": 1.648286530282033,
      "language_loss": 0.91897559,
      "learning_rate": 3.275791421376029e-06,
      "loss": 0.94718146,
      "num_input_tokens_seen": 2023760,
      "step": 92,
      "time_per_iteration": 2.8423190116882324
    },
    {
      "auxiliary_loss_clip": 0.01642373,
      "auxiliary_loss_mlp": 0.01144976,
      "balance_loss_clip": 1.22116864,
      "balance_loss_mlp": 1.06858706,
      "epoch": 0.01118258882943546,
      "flos": 16071210864000.0,
      "grad_norm": 2.114541513379243,
      "language_loss": 0.96269846,
      "learning_rate": 3.2836233457634622e-06,
      "loss": 0.99057198,
      "num_input_tokens_seen": 2041895,
      "step": 93,
      "time_per_iteration": 2.703707218170166
    },
    {
      "auxiliary_loss_clip": 0.01639222,
      "auxiliary_loss_mlp": 0.01177589,
      "balance_loss_clip": 1.21974087,
      "balance_loss_mlp": 1.08660817,
      "epoch": 0.011302831720074551,
      "flos": 20668602458880.0,
      "grad_norm": 2.270941476267931,
      "language_loss": 0.85331905,
      "learning_rate": 3.2913715046481135e-06,
      "loss": 0.88148719,
      "num_input_tokens_seen": 2061640,
      "step": 94,
      "time_per_iteration": 2.7671830654144287
    },
    {
      "auxiliary_loss_clip": 0.01635502,
      "auxiliary_loss_mlp": 0.01156376,
      "balance_loss_clip": 1.21817446,
      "balance_loss_mlp": 1.07927144,
      "epoch": 0.011423074610713641,
      "flos": 13072238490240.0,
      "grad_norm": 2.03624700870222,
      "language_loss": 0.88865674,
      "learning_rate": 3.299037670895023e-06,
      "loss": 0.91657555,
      "num_input_tokens_seen": 2078255,
      "step": 95,
      "time_per_iteration": 2.7635302543640137
    },
    {
      "auxiliary_loss_clip": 0.01634719,
      "auxiliary_loss_mlp": 0.01142505,
      "balance_loss_clip": 1.22221267,
      "balance_loss_mlp": 1.0635885,
      "epoch": 0.011543317501352733,
      "flos": 30335646689280.0,
      "grad_norm": 1.783137457887902,
      "language_loss": 0.80133307,
      "learning_rate": 3.3066235616750667e-06,
      "loss": 0.82910538,
      "num_input_tokens_seen": 2099490,
      "step": 96,
      "time_per_iteration": 2.785303831100464
    },
    {
      "auxiliary_loss_clip": 0.01613959,
      "auxiliary_loss_mlp": 0.01138203,
      "balance_loss_clip": 1.20936966,
      "balance_loss_mlp": 1.06415057,
      "epoch": 0.011663560391991824,
      "flos": 15522962601600.0,
      "grad_norm": 2.077525289496164,
      "language_loss": 0.92428941,
      "learning_rate": 3.3141308407736276e-06,
      "loss": 0.95181102,
      "num_input_tokens_seen": 2116125,
      "step": 97,
      "time_per_iteration": 4.618737459182739
    },
    {
      "auxiliary_loss_clip": 0.0162001,
      "auxiliary_loss_mlp": 0.0114511,
      "balance_loss_clip": 1.20729864,
      "balance_loss_mlp": 1.07205856,
      "epoch": 0.011783803282630914,
      "flos": 19902125116800.0,
      "grad_norm": 2.6546458256323,
      "language_loss": 0.86714077,
      "learning_rate": 3.321561120780869e-06,
      "loss": 0.89479202,
      "num_input_tokens_seen": 2134835,
      "step": 98,
      "time_per_iteration": 2.7951748371124268
    },
    {
      "auxiliary_loss_clip": 0.01609689,
      "auxiliary_loss_mlp": 0.01137013,
      "balance_loss_clip": 1.20977092,
      "balance_loss_mlp": 1.07154322,
      "epoch": 0.011904046173270006,
      "flos": 22340674517760.0,
      "grad_norm": 2.463202910699752,
      "language_loss": 1.01441598,
      "learning_rate": 3.3289159651708192e-06,
      "loss": 1.04188299,
      "num_input_tokens_seen": 2152410,
      "step": 99,
      "time_per_iteration": 2.71770977973938
    },
    {
      "auxiliary_loss_clip": 0.01604558,
      "auxiliary_loss_mlp": 0.01136609,
      "balance_loss_clip": 1.20486593,
      "balance_loss_mlp": 1.06513131,
      "epoch": 0.012024289063909096,
      "flos": 19100060375040.0,
      "grad_norm": 2.2996490497581226,
      "language_loss": 0.97540843,
      "learning_rate": 3.3361968902759768e-06,
      "loss": 1.00282001,
      "num_input_tokens_seen": 2172090,
      "step": 100,
      "time_per_iteration": 2.7766711711883545
    },
    {
      "auxiliary_loss_clip": 0.01598694,
      "auxiliary_loss_mlp": 0.01125159,
      "balance_loss_clip": 1.20338929,
      "balance_loss_mlp": 1.0624547,
      "epoch": 0.012144531954548187,
      "flos": 15012205159680.0,
      "grad_norm": 8.111755759813908,
      "language_loss": 0.94011414,
      "learning_rate": 3.343405367163663e-06,
      "loss": 0.96735263,
      "num_input_tokens_seen": 2189020,
      "step": 101,
      "time_per_iteration": 2.7621188163757324
    },
    {
      "auxiliary_loss_clip": 0.01600968,
      "auxiliary_loss_mlp": 0.01132079,
      "balance_loss_clip": 1.20292962,
      "balance_loss_mlp": 1.06718159,
      "epoch": 0.012264774845187279,
      "flos": 15122020014720.0,
      "grad_norm": 2.7451334783640036,
      "language_loss": 0.80832958,
      "learning_rate": 3.350542823419951e-06,
      "loss": 0.83566004,
      "num_input_tokens_seen": 2205620,
      "step": 102,
      "time_per_iteration": 2.8932337760925293
    },
    {
      "auxiliary_loss_clip": 0.01594108,
      "auxiliary_loss_mlp": 0.01144785,
      "balance_loss_clip": 1.19647837,
      "balance_loss_mlp": 1.0808413,
      "epoch": 0.012385017735826368,
      "flos": 13949248959360.0,
      "grad_norm": 4.682482170957838,
      "language_loss": 0.87308657,
      "learning_rate": 3.3576106448465615e-06,
      "loss": 0.90047538,
      "num_input_tokens_seen": 2219000,
      "step": 103,
      "time_per_iteration": 2.689960479736328
    },
    {
      "auxiliary_loss_clip": 0.01583521,
      "auxiliary_loss_mlp": 0.01134063,
      "balance_loss_clip": 1.1947403,
      "balance_loss_mlp": 1.06921339,
      "epoch": 0.01250526062646546,
      "flos": 23623260428160.0,
      "grad_norm": 2.0109268279869092,
      "language_loss": 0.8820861,
      "learning_rate": 3.3646101770757797e-06,
      "loss": 0.90926194,
      "num_input_tokens_seen": 2237790,
      "step": 104,
      "time_per_iteration": 2.7636325359344482
    },
    {
      "auxiliary_loss_clip": 0.01576463,
      "auxiliary_loss_mlp": 0.01133903,
      "balance_loss_clip": 1.19086623,
      "balance_loss_mlp": 1.06585836,
      "epoch": 0.012625503517104552,
      "flos": 34640078958720.0,
      "grad_norm": 1.869165703671756,
      "language_loss": 0.85603428,
      "learning_rate": 3.371542727108104e-06,
      "loss": 0.88313794,
      "num_input_tokens_seen": 2259965,
      "step": 105,
      "time_per_iteration": 2.8976495265960693
    },
    {
      "auxiliary_loss_clip": 0.01578297,
      "auxiliary_loss_mlp": 0.01173763,
      "balance_loss_clip": 1.1931932,
      "balance_loss_mlp": 1.10791147,
      "epoch": 0.012745746407743641,
      "flos": 17821891837440.0,
      "grad_norm": 4.211353926744972,
      "language_loss": 0.89737689,
      "learning_rate": 3.3784095647770114e-06,
      "loss": 0.92489755,
      "num_input_tokens_seen": 2278610,
      "step": 106,
      "time_per_iteration": 2.7471508979797363
    },
    {
      "auxiliary_loss_clip": 0.01570641,
      "auxiliary_loss_mlp": 0.01135146,
      "balance_loss_clip": 1.18525732,
      "balance_loss_mlp": 1.07034433,
      "epoch": 0.012865989298382733,
      "flos": 20595057361920.0,
      "grad_norm": 1.9061386926554948,
      "language_loss": 0.88679534,
      "learning_rate": 3.3852119241449547e-06,
      "loss": 0.91385317,
      "num_input_tokens_seen": 2297730,
      "step": 107,
      "time_per_iteration": 2.7918107509613037
    },
    {
      "auxiliary_loss_clip": 0.01565141,
      "auxiliary_loss_mlp": 0.01128436,
      "balance_loss_clip": 1.18379807,
      "balance_loss_mlp": 1.06654274,
      "epoch": 0.012986232189021825,
      "flos": 23948969978880.0,
      "grad_norm": 3.6654584676992803,
      "language_loss": 0.96207058,
      "learning_rate": 3.3919510048344295e-06,
      "loss": 0.98900628,
      "num_input_tokens_seen": 2315740,
      "step": 108,
      "time_per_iteration": 2.8288490772247314
    },
    {
      "auxiliary_loss_clip": 0.0155556,
      "auxiliary_loss_mlp": 0.01123258,
      "balance_loss_clip": 1.17968702,
      "balance_loss_mlp": 1.06651413,
      "epoch": 0.013106475079660914,
      "flos": 23725425686400.0,
      "grad_norm": 2.0601187812985646,
      "language_loss": 0.86726987,
      "learning_rate": 3.3986279732976907e-06,
      "loss": 0.89405799,
      "num_input_tokens_seen": 2334215,
      "step": 109,
      "time_per_iteration": 2.9138190746307373
    },
    {
      "auxiliary_loss_clip": 0.01551265,
      "auxiliary_loss_mlp": 0.01107544,
      "balance_loss_clip": 1.17824244,
      "balance_loss_mlp": 1.05153942,
      "epoch": 0.013226717970300006,
      "flos": 21102438925440.0,
      "grad_norm": 2.2812966362338143,
      "language_loss": 0.95389366,
      "learning_rate": 3.4052439640284983e-06,
      "loss": 0.98048174,
      "num_input_tokens_seen": 2353130,
      "step": 110,
      "time_per_iteration": 2.8273158073425293
    },
    {
      "auxiliary_loss_clip": 0.01552231,
      "auxiliary_loss_mlp": 0.01125928,
      "balance_loss_clip": 1.18120956,
      "balance_loss_mlp": 1.06815922,
      "epoch": 0.013346960860939098,
      "flos": 24863902231680.0,
      "grad_norm": 1.8217300101508092,
      "language_loss": 0.81111324,
      "learning_rate": 3.4118000807190217e-06,
      "loss": 0.8378948,
      "num_input_tokens_seen": 2374010,
      "step": 111,
      "time_per_iteration": 2.9265265464782715
    },
    {
      "auxiliary_loss_clip": 0.01554999,
      "auxiliary_loss_mlp": 0.01122773,
      "balance_loss_clip": 1.17961705,
      "balance_loss_mlp": 1.06698346,
      "epoch": 0.013467203751578187,
      "flos": 28181940140160.0,
      "grad_norm": 2.244319661274039,
      "language_loss": 0.75792426,
      "learning_rate": 3.4182973973648723e-06,
      "loss": 0.78470194,
      "num_input_tokens_seen": 2395220,
      "step": 112,
      "time_per_iteration": 2.8461244106292725
    },
    {
      "auxiliary_loss_clip": 0.01542487,
      "auxiliary_loss_mlp": 0.01140312,
      "balance_loss_clip": 1.17528915,
      "balance_loss_mlp": 1.08490396,
      "epoch": 0.013587446642217279,
      "flos": 18916233546240.0,
      "grad_norm": 2.5584060410191998,
      "language_loss": 0.94861293,
      "learning_rate": 3.424736959321014e-06,
      "loss": 0.97544098,
      "num_input_tokens_seen": 2413025,
      "step": 113,
      "time_per_iteration": 2.728675127029419
    },
    {
      "auxiliary_loss_clip": 0.0154671,
      "auxiliary_loss_mlp": 0.01135983,
      "balance_loss_clip": 1.17592978,
      "balance_loss_mlp": 1.07900131,
      "epoch": 0.01370768953285637,
      "flos": 23988615615360.0,
      "grad_norm": 1.948387595041043,
      "language_loss": 0.88813108,
      "learning_rate": 3.431119784311155e-06,
      "loss": 0.914958,
      "num_input_tokens_seen": 2432700,
      "step": 114,
      "time_per_iteration": 2.815469741821289
    },
    {
      "auxiliary_loss_clip": 0.01532334,
      "auxiliary_loss_mlp": 0.01124578,
      "balance_loss_clip": 1.17121243,
      "balance_loss_mlp": 1.07169688,
      "epoch": 0.01382793242349546,
      "flos": 39202565512320.0,
      "grad_norm": 2.4841329687901155,
      "language_loss": 0.77702457,
      "learning_rate": 3.43744686339307e-06,
      "loss": 0.80359364,
      "num_input_tokens_seen": 2455020,
      "step": 115,
      "time_per_iteration": 2.8683135509490967
    },
    {
      "auxiliary_loss_clip": 0.0152728,
      "auxiliary_loss_mlp": 0.01088774,
      "balance_loss_clip": 1.16545391,
      "balance_loss_mlp": 1.03965998,
      "epoch": 0.013948175314134552,
      "flos": 41353506714240.0,
      "grad_norm": 2.142140868903816,
      "language_loss": 0.91037285,
      "learning_rate": 3.44371916188212e-06,
      "loss": 0.93653345,
      "num_input_tokens_seen": 2475775,
      "step": 116,
      "time_per_iteration": 2.9094302654266357
    },
    {
      "auxiliary_loss_clip": 0.01524132,
      "auxiliary_loss_mlp": 0.01104199,
      "balance_loss_clip": 1.16634381,
      "balance_loss_mlp": 1.05706406,
      "epoch": 0.014068418204773643,
      "flos": 22453542028800.0,
      "grad_norm": 5.743230579926946,
      "language_loss": 0.86295396,
      "learning_rate": 3.449937620235143e-06,
      "loss": 0.88923728,
      "num_input_tokens_seen": 2496370,
      "step": 117,
      "time_per_iteration": 2.731107473373413
    },
    {
      "auxiliary_loss_clip": 0.01525784,
      "auxiliary_loss_mlp": 0.0110992,
      "balance_loss_clip": 1.1674602,
      "balance_loss_mlp": 1.06049562,
      "epoch": 0.014188661095412733,
      "flos": 23805147922560.0,
      "grad_norm": 1.6302851245338519,
      "language_loss": 0.89428717,
      "learning_rate": 3.456103154896722e-06,
      "loss": 0.92064422,
      "num_input_tokens_seen": 2517645,
      "step": 118,
      "time_per_iteration": 2.845167398452759
    },
    {
      "auxiliary_loss_clip": 0.01513825,
      "auxiliary_loss_mlp": 0.01118777,
      "balance_loss_clip": 1.16037691,
      "balance_loss_mlp": 1.07164168,
      "epoch": 0.014308903986051825,
      "flos": 23660248458240.0,
      "grad_norm": 1.927899974760965,
      "language_loss": 0.92630631,
      "learning_rate": 3.462216659109757e-06,
      "loss": 0.95263243,
      "num_input_tokens_seen": 2537825,
      "step": 119,
      "time_per_iteration": 2.7461724281311035
    },
    {
      "auxiliary_loss_clip": 0.01533591,
      "auxiliary_loss_mlp": 0.01129803,
      "balance_loss_clip": 1.17021585,
      "balance_loss_mlp": 1.08192813,
      "epoch": 0.014429146876690916,
      "flos": 20667991927680.0,
      "grad_norm": 2.3595338761091225,
      "language_loss": 0.85433638,
      "learning_rate": 3.4682790036921077e-06,
      "loss": 0.88097036,
      "num_input_tokens_seen": 2556485,
      "step": 120,
      "time_per_iteration": 2.760356903076172
    },
    {
      "auxiliary_loss_clip": 0.01509103,
      "auxiliary_loss_mlp": 0.01105594,
      "balance_loss_clip": 1.1616385,
      "balance_loss_mlp": 1.06570637,
      "epoch": 0.014549389767330006,
      "flos": 20229199384320.0,
      "grad_norm": 1.9836852671485767,
      "language_loss": 0.83210635,
      "learning_rate": 3.4742910377810193e-06,
      "loss": 0.85825336,
      "num_input_tokens_seen": 2573945,
      "step": 121,
      "time_per_iteration": 2.7860090732574463
    },
    {
      "auxiliary_loss_clip": 0.01508445,
      "auxiliary_loss_mlp": 0.0111727,
      "balance_loss_clip": 1.16094863,
      "balance_loss_mlp": 1.0742358,
      "epoch": 0.014669632657969098,
      "flos": 18004174381440.0,
      "grad_norm": 2.356210490389133,
      "language_loss": 0.88766885,
      "learning_rate": 3.4802535895469042e-06,
      "loss": 0.91392601,
      "num_input_tokens_seen": 2592695,
      "step": 122,
      "time_per_iteration": 2.6975319385528564
    },
    {
      "auxiliary_loss_clip": 0.01511069,
      "auxiliary_loss_mlp": 0.01108716,
      "balance_loss_clip": 1.15972662,
      "balance_loss_mlp": 1.06508517,
      "epoch": 0.01478987554860819,
      "flos": 22741796672640.0,
      "grad_norm": 2.0612076045424805,
      "language_loss": 0.89491719,
      "learning_rate": 3.4861674668779934e-06,
      "loss": 0.92111504,
      "num_input_tokens_seen": 2610925,
      "step": 123,
      "time_per_iteration": 2.784285545349121
    },
    {
      "auxiliary_loss_clip": 0.01502669,
      "auxiliary_loss_mlp": 0.0110103,
      "balance_loss_clip": 1.1553731,
      "balance_loss_mlp": 1.05661261,
      "epoch": 0.01491011843924728,
      "flos": 17198590106880.0,
      "grad_norm": 2.037538627232873,
      "language_loss": 0.84087837,
      "learning_rate": 3.492033458037272e-06,
      "loss": 0.86691535,
      "num_input_tokens_seen": 2629495,
      "step": 124,
      "time_per_iteration": 5.581273317337036
    },
    {
      "auxiliary_loss_clip": 0.01498207,
      "auxiliary_loss_mlp": 0.01107425,
      "balance_loss_clip": 1.15232182,
      "balance_loss_mlp": 1.06708419,
      "epoch": 0.01503036132988637,
      "flos": 17673867889920.0,
      "grad_norm": 2.3211601292151913,
      "language_loss": 0.87003112,
      "learning_rate": 3.497852332293018e-06,
      "loss": 0.89608741,
      "num_input_tokens_seen": 2645070,
      "step": 125,
      "time_per_iteration": 3.59490704536438
    },
    {
      "auxiliary_loss_clip": 0.01496939,
      "auxiliary_loss_mlp": 0.01109982,
      "balance_loss_clip": 1.15398228,
      "balance_loss_mlp": 1.07083416,
      "epoch": 0.015150604220525462,
      "flos": 18878239935360.0,
      "grad_norm": 2.388215428074628,
      "language_loss": 0.96359652,
      "learning_rate": 3.5036248405242356e-06,
      "loss": 0.98966569,
      "num_input_tokens_seen": 2663825,
      "step": 126,
      "time_per_iteration": 2.669245481491089
    },
    {
      "auxiliary_loss_clip": 0.01497479,
      "auxiliary_loss_mlp": 0.01106413,
      "balance_loss_clip": 1.15287364,
      "balance_loss_mlp": 1.0637362,
      "epoch": 0.015270847111164552,
      "flos": 39420184060800.0,
      "grad_norm": 1.9129141565803736,
      "language_loss": 0.82928354,
      "learning_rate": 3.509351715802146e-06,
      "loss": 0.85532248,
      "num_input_tokens_seen": 2684710,
      "step": 127,
      "time_per_iteration": 2.8840103149414062
    },
    {
      "auxiliary_loss_clip": 0.01495467,
      "auxiliary_loss_mlp": 0.01118877,
      "balance_loss_clip": 1.15134811,
      "balance_loss_mlp": 1.07491255,
      "epoch": 0.015391090001803644,
      "flos": 43762466286720.0,
      "grad_norm": 2.338565350887651,
      "language_loss": 0.78592438,
      "learning_rate": 3.5150336739488763e-06,
      "loss": 0.81206775,
      "num_input_tokens_seen": 2706995,
      "step": 128,
      "time_per_iteration": 2.872817039489746
    },
    {
      "auxiliary_loss_clip": 0.01491207,
      "auxiliary_loss_mlp": 0.01085666,
      "balance_loss_clip": 1.15190148,
      "balance_loss_mlp": 1.04904497,
      "epoch": 0.015511332892442733,
      "flos": 18916341287040.0,
      "grad_norm": 1.8810526920302832,
      "language_loss": 0.84168535,
      "learning_rate": 3.5206714140744143e-06,
      "loss": 0.86745411,
      "num_input_tokens_seen": 2727050,
      "step": 129,
      "time_per_iteration": 2.702915906906128
    },
    {
      "auxiliary_loss_clip": 0.01494116,
      "auxiliary_loss_mlp": 0.01114898,
      "balance_loss_clip": 1.15510595,
      "balance_loss_mlp": 1.07520127,
      "epoch": 0.015631575783081827,
      "flos": 24535283679360.0,
      "grad_norm": 2.9104277112865553,
      "language_loss": 0.87563878,
      "learning_rate": 3.5262656190928208e-06,
      "loss": 0.90172893,
      "num_input_tokens_seen": 2745350,
      "step": 130,
      "time_per_iteration": 2.76291561126709
    },
    {
      "auxiliary_loss_clip": 0.01468285,
      "auxiliary_loss_mlp": 0.0102369,
      "balance_loss_clip": 1.18346858,
      "balance_loss_mlp": 1.00757313,
      "epoch": 0.015751818673720917,
      "flos": 62328536098560.0,
      "grad_norm": 1.0433007842048359,
      "language_loss": 0.71528476,
      "learning_rate": 3.5318169562186737e-06,
      "loss": 0.74020451,
      "num_input_tokens_seen": 2814195,
      "step": 131,
      "time_per_iteration": 3.2637038230895996
    },
    {
      "auxiliary_loss_clip": 0.01479537,
      "auxiliary_loss_mlp": 0.01117253,
      "balance_loss_clip": 1.14708257,
      "balance_loss_mlp": 1.08075118,
      "epoch": 0.015872061564360006,
      "flos": 23878549365120.0,
      "grad_norm": 2.6027774950912907,
      "language_loss": 0.82260787,
      "learning_rate": 3.5373260774446292e-06,
      "loss": 0.84857571,
      "num_input_tokens_seen": 2834645,
      "step": 132,
      "time_per_iteration": 2.701519012451172
    },
    {
      "auxiliary_loss_clip": 0.01477795,
      "auxiliary_loss_mlp": 0.01111042,
      "balance_loss_clip": 1.14596391,
      "balance_loss_mlp": 1.07439685,
      "epoch": 0.0159923044549991,
      "flos": 23367899664000.0,
      "grad_norm": 1.7460470291133263,
      "language_loss": 0.90433133,
      "learning_rate": 3.542793620000961e-06,
      "loss": 0.93021971,
      "num_input_tokens_seen": 2854120,
      "step": 133,
      "time_per_iteration": 2.757899045944214
    },
    {
      "auxiliary_loss_clip": 0.01475354,
      "auxiliary_loss_mlp": 0.01103747,
      "balance_loss_clip": 1.14492011,
      "balance_loss_mlp": 1.06626749,
      "epoch": 0.01611254734563819,
      "flos": 17858305249920.0,
      "grad_norm": 2.1689442897443705,
      "language_loss": 0.87030244,
      "learning_rate": 3.5482202067978894e-06,
      "loss": 0.89609349,
      "num_input_tokens_seen": 2871330,
      "step": 134,
      "time_per_iteration": 2.6864850521087646
    },
    {
      "auxiliary_loss_clip": 0.01472573,
      "auxiliary_loss_mlp": 0.0109635,
      "balance_loss_clip": 1.14488745,
      "balance_loss_mlp": 1.05992007,
      "epoch": 0.01623279023627728,
      "flos": 20954774113920.0,
      "grad_norm": 2.6293490573315195,
      "language_loss": 0.76236743,
      "learning_rate": 3.553606446851471e-06,
      "loss": 0.78805661,
      "num_input_tokens_seen": 2888070,
      "step": 135,
      "time_per_iteration": 2.6714701652526855
    },
    {
      "auxiliary_loss_clip": 0.01460495,
      "auxiliary_loss_mlp": 0.01095846,
      "balance_loss_clip": 1.13716817,
      "balance_loss_mlp": 1.0602262,
      "epoch": 0.016353033126916373,
      "flos": 15742412743680.0,
      "grad_norm": 2.6769646483389193,
      "language_loss": 0.83286893,
      "learning_rate": 3.5589529356937613e-06,
      "loss": 0.85843235,
      "num_input_tokens_seen": 2906465,
      "step": 136,
      "time_per_iteration": 2.7559194564819336
    },
    {
      "auxiliary_loss_clip": 0.01469956,
      "auxiliary_loss_mlp": 0.01097379,
      "balance_loss_clip": 1.14114356,
      "balance_loss_mlp": 1.06147313,
      "epoch": 0.016473276017555463,
      "flos": 18807280617600.0,
      "grad_norm": 2.257750821097481,
      "language_loss": 0.76942945,
      "learning_rate": 3.5642602557679627e-06,
      "loss": 0.79510272,
      "num_input_tokens_seen": 2924915,
      "step": 137,
      "time_per_iteration": 2.7101097106933594
    },
    {
      "auxiliary_loss_clip": 0.01464071,
      "auxiliary_loss_mlp": 0.01085636,
      "balance_loss_clip": 1.14641809,
      "balance_loss_mlp": 1.05547631,
      "epoch": 0.016593518908194552,
      "flos": 24352641999360.0,
      "grad_norm": 2.495061958284871,
      "language_loss": 0.84350836,
      "learning_rate": 3.569528976809202e-06,
      "loss": 0.86900544,
      "num_input_tokens_seen": 2942130,
      "step": 138,
      "time_per_iteration": 2.760922431945801
    },
    {
      "auxiliary_loss_clip": 0.01464837,
      "auxiliary_loss_mlp": 0.01105519,
      "balance_loss_clip": 1.1413238,
      "balance_loss_mlp": 1.06906497,
      "epoch": 0.016713761798833646,
      "flos": 22346133384960.0,
      "grad_norm": 1.753648278943012,
      "language_loss": 0.89756119,
      "learning_rate": 3.5747596562115522e-06,
      "loss": 0.92326474,
      "num_input_tokens_seen": 2962745,
      "step": 139,
      "time_per_iteration": 2.7051162719726562
    },
    {
      "auxiliary_loss_clip": 0.01469464,
      "auxiliary_loss_mlp": 0.01100375,
      "balance_loss_clip": 1.1434015,
      "balance_loss_mlp": 1.06589949,
      "epoch": 0.016834004689472735,
      "flos": 17821820010240.0,
      "grad_norm": 2.589273422096452,
      "language_loss": 0.90806085,
      "learning_rate": 3.5799528393819138e-06,
      "loss": 0.93375927,
      "num_input_tokens_seen": 2981825,
      "step": 140,
      "time_per_iteration": 2.7234466075897217
    },
    {
      "auxiliary_loss_clip": 0.0145219,
      "auxiliary_loss_mlp": 0.01098213,
      "balance_loss_clip": 1.13401794,
      "balance_loss_mlp": 1.06597853,
      "epoch": 0.016954247580111825,
      "flos": 20519501103360.0,
      "grad_norm": 3.7955451329503074,
      "language_loss": 0.87996149,
      "learning_rate": 3.585109060081286e-06,
      "loss": 0.90546554,
      "num_input_tokens_seen": 3001625,
      "step": 141,
      "time_per_iteration": 2.680722951889038
    },
    {
      "auxiliary_loss_clip": 0.01458798,
      "auxiliary_loss_mlp": 0.01097859,
      "balance_loss_clip": 1.13819242,
      "balance_loss_mlp": 1.06525564,
      "epoch": 0.017074490470750915,
      "flos": 22088869200000.0,
      "grad_norm": 2.1452041211680406,
      "language_loss": 0.78612632,
      "learning_rate": 3.590228840753992e-06,
      "loss": 0.81169283,
      "num_input_tokens_seen": 3022055,
      "step": 142,
      "time_per_iteration": 2.7220449447631836
    },
    {
      "auxiliary_loss_clip": 0.01449426,
      "auxiliary_loss_mlp": 0.01098736,
      "balance_loss_clip": 1.13515687,
      "balance_loss_mlp": 1.06829023,
      "epoch": 0.01719473336139001,
      "flos": 15997270717440.0,
      "grad_norm": 2.4732292601047114,
      "language_loss": 0.87539285,
      "learning_rate": 3.5953126928453423e-06,
      "loss": 0.9008745,
      "num_input_tokens_seen": 3039605,
      "step": 143,
      "time_per_iteration": 2.658341646194458
    },
    {
      "auxiliary_loss_clip": 0.01446441,
      "auxiliary_loss_mlp": 0.01082591,
      "balance_loss_clip": 1.13162756,
      "balance_loss_mlp": 1.05318177,
      "epoch": 0.017314976252029098,
      "flos": 22492038430080.0,
      "grad_norm": 2.4248332127950403,
      "language_loss": 0.80562556,
      "learning_rate": 3.600361117108239e-06,
      "loss": 0.83091581,
      "num_input_tokens_seen": 3059405,
      "step": 144,
      "time_per_iteration": 2.7081542015075684
    },
    {
      "auxiliary_loss_clip": 0.01452205,
      "auxiliary_loss_mlp": 0.0108819,
      "balance_loss_clip": 1.13371134,
      "balance_loss_mlp": 1.05690956,
      "epoch": 0.017435219142668188,
      "flos": 22018053536640.0,
      "grad_norm": 1.866466206661156,
      "language_loss": 0.97175223,
      "learning_rate": 3.6053746038991616e-06,
      "loss": 0.99715614,
      "num_input_tokens_seen": 3078490,
      "step": 145,
      "time_per_iteration": 2.8035378456115723
    },
    {
      "auxiliary_loss_clip": 0.01420949,
      "auxiliary_loss_mlp": 0.0101249,
      "balance_loss_clip": 1.16711628,
      "balance_loss_mlp": 1.00033092,
      "epoch": 0.01755546203330728,
      "flos": 72240526149120.0,
      "grad_norm": 1.055979533319725,
      "language_loss": 0.58486211,
      "learning_rate": 3.6103536334639843e-06,
      "loss": 0.60919654,
      "num_input_tokens_seen": 3131755,
      "step": 146,
      "time_per_iteration": 3.204871892929077
    },
    {
      "auxiliary_loss_clip": 0.01442222,
      "auxiliary_loss_mlp": 0.01082659,
      "balance_loss_clip": 1.13115454,
      "balance_loss_mlp": 1.05388212,
      "epoch": 0.01767570492394637,
      "flos": 25337061112320.0,
      "grad_norm": 2.021915097241205,
      "language_loss": 0.85587215,
      "learning_rate": 3.615298676214041e-06,
      "loss": 0.88112092,
      "num_input_tokens_seen": 3152035,
      "step": 147,
      "time_per_iteration": 2.705190658569336
    },
    {
      "auxiliary_loss_clip": 0.01439985,
      "auxiliary_loss_mlp": 0.01096617,
      "balance_loss_clip": 1.12905931,
      "balance_loss_mlp": 1.06826949,
      "epoch": 0.01779594781458546,
      "flos": 20449188230400.0,
      "grad_norm": 2.1374252749818368,
      "language_loss": 0.88601542,
      "learning_rate": 3.6202101929928317e-06,
      "loss": 0.91138148,
      "num_input_tokens_seen": 3170625,
      "step": 148,
      "time_per_iteration": 2.675208568572998
    },
    {
      "auxiliary_loss_clip": 0.01434374,
      "auxiliary_loss_mlp": 0.01091237,
      "balance_loss_clip": 1.12678969,
      "balance_loss_mlp": 1.06408119,
      "epoch": 0.017916190705224554,
      "flos": 16253601148800.0,
      "grad_norm": 1.9388730278393012,
      "language_loss": 0.88405055,
      "learning_rate": 3.6250886353337413e-06,
      "loss": 0.90930659,
      "num_input_tokens_seen": 3188155,
      "step": 149,
      "time_per_iteration": 2.6308436393737793
    },
    {
      "auxiliary_loss_clip": 0.01447637,
      "auxiliary_loss_mlp": 0.01094093,
      "balance_loss_clip": 1.13388312,
      "balance_loss_mlp": 1.06678247,
      "epoch": 0.018036433595863644,
      "flos": 23330588411520.0,
      "grad_norm": 1.9349917923098365,
      "language_loss": 0.86409307,
      "learning_rate": 3.6299344457091488e-06,
      "loss": 0.88951033,
      "num_input_tokens_seen": 3209015,
      "step": 150,
      "time_per_iteration": 2.6541969776153564
    },
    {
      "auxiliary_loss_clip": 0.01440057,
      "auxiliary_loss_mlp": 0.01082505,
      "balance_loss_clip": 1.13095355,
      "balance_loss_mlp": 1.05626702,
      "epoch": 0.018156676486502734,
      "flos": 18588010043520.0,
      "grad_norm": 60.74782179621216,
      "language_loss": 0.93918055,
      "learning_rate": 3.634748057771256e-06,
      "loss": 0.96440613,
      "num_input_tokens_seen": 3224955,
      "step": 151,
      "time_per_iteration": 3.6421430110931396
    },
    {
      "auxiliary_loss_clip": 0.01432905,
      "auxiliary_loss_mlp": 0.01088295,
      "balance_loss_clip": 1.1281563,
      "balance_loss_mlp": 1.06227124,
      "epoch": 0.018276919377141827,
      "flos": 25448707560960.0,
      "grad_norm": 1.8155231758319879,
      "language_loss": 0.85475034,
      "learning_rate": 3.639529896584965e-06,
      "loss": 0.87996233,
      "num_input_tokens_seen": 3246330,
      "step": 152,
      "time_per_iteration": 3.6537747383117676
    },
    {
      "auxiliary_loss_clip": 0.01435852,
      "auxiliary_loss_mlp": 0.01079379,
      "balance_loss_clip": 1.12838233,
      "balance_loss_mlp": 1.05174649,
      "epoch": 0.018397162267780917,
      "flos": 20047311889920.0,
      "grad_norm": 4.035985031723493,
      "language_loss": 0.88845593,
      "learning_rate": 3.6442803788531233e-06,
      "loss": 0.91360825,
      "num_input_tokens_seen": 3264290,
      "step": 153,
      "time_per_iteration": 3.632229804992676
    },
    {
      "auxiliary_loss_clip": 0.01438202,
      "auxiliary_loss_mlp": 0.01090213,
      "balance_loss_clip": 1.12786961,
      "balance_loss_mlp": 1.06168628,
      "epoch": 0.018517405158420007,
      "flos": 27565282425600.0,
      "grad_norm": 2.1271517447872177,
      "language_loss": 0.95901823,
      "learning_rate": 3.6489999131344357e-06,
      "loss": 0.9843024,
      "num_input_tokens_seen": 3287065,
      "step": 154,
      "time_per_iteration": 2.79250168800354
    },
    {
      "auxiliary_loss_clip": 0.01426155,
      "auxiliary_loss_mlp": 0.01087848,
      "balance_loss_clip": 1.12432098,
      "balance_loss_mlp": 1.06306458,
      "epoch": 0.0186376480490591,
      "flos": 19354056422400.0,
      "grad_norm": 2.2664702982278726,
      "language_loss": 0.90549302,
      "learning_rate": 3.653688900054313e-06,
      "loss": 0.93063307,
      "num_input_tokens_seen": 3305595,
      "step": 155,
      "time_per_iteration": 2.75221848487854
    },
    {
      "auxiliary_loss_clip": 0.01429889,
      "auxiliary_loss_mlp": 0.01069202,
      "balance_loss_clip": 1.12275457,
      "balance_loss_mlp": 1.04296374,
      "epoch": 0.01875789093969819,
      "flos": 26687840993280.0,
      "grad_norm": 2.1317739791986186,
      "language_loss": 0.76108301,
      "learning_rate": 3.6583477325089526e-06,
      "loss": 0.78607392,
      "num_input_tokens_seen": 3326135,
      "step": 156,
      "time_per_iteration": 2.748176097869873
    },
    {
      "auxiliary_loss_clip": 0.01424508,
      "auxiliary_loss_mlp": 0.01077342,
      "balance_loss_clip": 1.12165499,
      "balance_loss_mlp": 1.05160451,
      "epoch": 0.01887813383033728,
      "flos": 24353001135360.0,
      "grad_norm": 2.3646364926193004,
      "language_loss": 1.04355454,
      "learning_rate": 3.6629767958628916e-06,
      "loss": 1.06857312,
      "num_input_tokens_seen": 3343510,
      "step": 157,
      "time_per_iteration": 2.6955764293670654
    },
    {
      "auxiliary_loss_clip": 0.01420568,
      "auxiliary_loss_mlp": 0.01077353,
      "balance_loss_clip": 1.12279558,
      "balance_loss_mlp": 1.05122232,
      "epoch": 0.018998376720976373,
      "flos": 14647532330880.0,
      "grad_norm": 2.364645937029895,
      "language_loss": 0.85562301,
      "learning_rate": 3.667576468140291e-06,
      "loss": 0.88060224,
      "num_input_tokens_seen": 3361325,
      "step": 158,
      "time_per_iteration": 2.644195318222046
    },
    {
      "auxiliary_loss_clip": 0.0141619,
      "auxiliary_loss_mlp": 0.01065955,
      "balance_loss_clip": 1.11799502,
      "balance_loss_mlp": 1.04193485,
      "epoch": 0.019118619611615463,
      "flos": 29305261146240.0,
      "grad_norm": 3.588182172504127,
      "language_loss": 0.88900834,
      "learning_rate": 3.672147120210184e-06,
      "loss": 0.9138298,
      "num_input_tokens_seen": 3377925,
      "step": 159,
      "time_per_iteration": 2.771780014038086
    },
    {
      "auxiliary_loss_clip": 0.01423369,
      "auxiliary_loss_mlp": 0.0107536,
      "balance_loss_clip": 1.12559485,
      "balance_loss_mlp": 1.05193555,
      "epoch": 0.019238862502254553,
      "flos": 20886723797760.0,
      "grad_norm": 2.54416846392155,
      "language_loss": 0.86640805,
      "learning_rate": 3.6766891159659177e-06,
      "loss": 0.89139533,
      "num_input_tokens_seen": 3396335,
      "step": 160,
      "time_per_iteration": 2.644413471221924
    },
    {
      "auxiliary_loss_clip": 0.01423055,
      "auxiliary_loss_mlp": 0.01079508,
      "balance_loss_clip": 1.12619364,
      "balance_loss_mlp": 1.05571377,
      "epoch": 0.019359105392893646,
      "flos": 21360672777600.0,
      "grad_norm": 2.667775828425892,
      "language_loss": 0.87692463,
      "learning_rate": 3.6812028124990075e-06,
      "loss": 0.90195024,
      "num_input_tokens_seen": 3413605,
      "step": 161,
      "time_per_iteration": 2.7047224044799805
    },
    {
      "auxiliary_loss_clip": 0.01417286,
      "auxiliary_loss_mlp": 0.01081682,
      "balance_loss_clip": 1.12193704,
      "balance_loss_mlp": 1.05868673,
      "epoch": 0.019479348283532736,
      "flos": 16283729681280.0,
      "grad_norm": 3.7288049554399705,
      "language_loss": 0.81665742,
      "learning_rate": 3.6856885602676016e-06,
      "loss": 0.84164715,
      "num_input_tokens_seen": 3429640,
      "step": 162,
      "time_per_iteration": 2.637683868408203
    },
    {
      "auxiliary_loss_clip": 0.01417546,
      "auxiliary_loss_mlp": 0.01081885,
      "balance_loss_clip": 1.12317467,
      "balance_loss_mlp": 1.05922318,
      "epoch": 0.019599591174171826,
      "flos": 22091239497600.0,
      "grad_norm": 3.3490860286741237,
      "language_loss": 0.94057214,
      "learning_rate": 3.6901467032597733e-06,
      "loss": 0.9655664,
      "num_input_tokens_seen": 3448125,
      "step": 163,
      "time_per_iteration": 2.684487819671631
    },
    {
      "auxiliary_loss_clip": 0.01419963,
      "auxiliary_loss_mlp": 0.01068309,
      "balance_loss_clip": 1.12176824,
      "balance_loss_mlp": 1.04320359,
      "epoch": 0.01971983406481092,
      "flos": 19609668581760.0,
      "grad_norm": 3.392175801211264,
      "language_loss": 0.87212205,
      "learning_rate": 3.694577579151804e-06,
      "loss": 0.89700472,
      "num_input_tokens_seen": 3466535,
      "step": 164,
      "time_per_iteration": 2.685551166534424
    },
    {
      "auxiliary_loss_clip": 0.01419299,
      "auxiliary_loss_mlp": 0.01077409,
      "balance_loss_clip": 1.12331033,
      "balance_loss_mlp": 1.05325747,
      "epoch": 0.01984007695545001,
      "flos": 19099342103040.0,
      "grad_norm": 2.339034853334444,
      "language_loss": 0.73820829,
      "learning_rate": 3.6989815194616703e-06,
      "loss": 0.76317537,
      "num_input_tokens_seen": 3483730,
      "step": 165,
      "time_per_iteration": 2.7290780544281006
    },
    {
      "auxiliary_loss_clip": 0.01419113,
      "auxiliary_loss_mlp": 0.01081123,
      "balance_loss_clip": 1.12009597,
      "balance_loss_mlp": 1.05555248,
      "epoch": 0.0199603198460891,
      "flos": 20848406964480.0,
      "grad_norm": 2.1887185315771314,
      "language_loss": 0.79910827,
      "learning_rate": 3.703358849697888e-06,
      "loss": 0.82411063,
      "num_input_tokens_seen": 3503640,
      "step": 166,
      "time_per_iteration": 2.6929640769958496
    },
    {
      "auxiliary_loss_clip": 0.01414819,
      "auxiliary_loss_mlp": 0.01090163,
      "balance_loss_clip": 1.12252629,
      "balance_loss_mlp": 1.06779969,
      "epoch": 0.020080562736728192,
      "flos": 21870747861120.0,
      "grad_norm": 1.7095516970431228,
      "language_loss": 0.82670939,
      "learning_rate": 3.7077098895038803e-06,
      "loss": 0.8517592,
      "num_input_tokens_seen": 3523010,
      "step": 167,
      "time_per_iteration": 2.75655460357666
    },
    {
      "auxiliary_loss_clip": 0.0141455,
      "auxiliary_loss_mlp": 0.01076822,
      "balance_loss_clip": 1.12100387,
      "balance_loss_mlp": 1.05374265,
      "epoch": 0.020200805627367282,
      "flos": 21688788539520.0,
      "grad_norm": 2.181177540162999,
      "language_loss": 0.96936131,
      "learning_rate": 3.712034952798045e-06,
      "loss": 0.99427509,
      "num_input_tokens_seen": 3541125,
      "step": 168,
      "time_per_iteration": 2.7798078060150146
    },
    {
      "auxiliary_loss_clip": 0.01411482,
      "auxiliary_loss_mlp": 0.01082497,
      "balance_loss_clip": 1.11599302,
      "balance_loss_mlp": 1.05866718,
      "epoch": 0.02032104851800637,
      "flos": 33543043729920.0,
      "grad_norm": 2.323881710532503,
      "language_loss": 0.84760606,
      "learning_rate": 3.7163343479096656e-06,
      "loss": 0.87254584,
      "num_input_tokens_seen": 3562700,
      "step": 169,
      "time_per_iteration": 2.783356189727783
    },
    {
      "auxiliary_loss_clip": 0.01409335,
      "auxiliary_loss_mlp": 0.01078321,
      "balance_loss_clip": 1.11955953,
      "balance_loss_mlp": 1.05745935,
      "epoch": 0.020441291408645465,
      "flos": 31686965274240.0,
      "grad_norm": 2.148479148543879,
      "language_loss": 0.82996488,
      "learning_rate": 3.720608377710802e-06,
      "loss": 0.85484147,
      "num_input_tokens_seen": 3582790,
      "step": 170,
      "time_per_iteration": 2.7649166584014893
    },
    {
      "auxiliary_loss_clip": 0.01402499,
      "auxiliary_loss_mlp": 0.01085255,
      "balance_loss_clip": 1.11286259,
      "balance_loss_mlp": 1.06167603,
      "epoch": 0.020561534299284555,
      "flos": 20886687884160.0,
      "grad_norm": 2.1479474948592387,
      "language_loss": 0.86346686,
      "learning_rate": 3.7248573397443277e-06,
      "loss": 0.88834441,
      "num_input_tokens_seen": 3601715,
      "step": 171,
      "time_per_iteration": 2.7097723484039307
    },
    {
      "auxiliary_loss_clip": 0.01407558,
      "auxiliary_loss_mlp": 0.01090291,
      "balance_loss_clip": 1.11914945,
      "balance_loss_mlp": 1.06566274,
      "epoch": 0.020681777189923645,
      "flos": 20996610480000.0,
      "grad_norm": 2.3652997089914223,
      "language_loss": 0.97546226,
      "learning_rate": 3.729081526348224e-06,
      "loss": 1.00044072,
      "num_input_tokens_seen": 3620245,
      "step": 172,
      "time_per_iteration": 2.7945075035095215
    },
    {
      "auxiliary_loss_clip": 0.01408867,
      "auxiliary_loss_mlp": 0.0106833,
      "balance_loss_clip": 1.11786735,
      "balance_loss_mlp": 1.04693151,
      "epoch": 0.020802020080562738,
      "flos": 28257532312320.0,
      "grad_norm": 1.9980996958794162,
      "language_loss": 0.8489517,
      "learning_rate": 3.7332812247762777e-06,
      "loss": 0.87372369,
      "num_input_tokens_seen": 3641545,
      "step": 173,
      "time_per_iteration": 2.725947380065918
    },
    {
      "auxiliary_loss_clip": 0.01409526,
      "auxiliary_loss_mlp": 0.01067448,
      "balance_loss_clip": 1.12106454,
      "balance_loss_mlp": 1.04540658,
      "epoch": 0.020922262971201828,
      "flos": 19681274344320.0,
      "grad_norm": 2.5275926931378874,
      "language_loss": 0.95584893,
      "learning_rate": 3.737456717315293e-06,
      "loss": 0.98061872,
      "num_input_tokens_seen": 3660510,
      "step": 174,
      "time_per_iteration": 2.7185251712799072
    },
    {
      "auxiliary_loss_clip": 0.01398185,
      "auxiliary_loss_mlp": 0.01087161,
      "balance_loss_clip": 1.11622453,
      "balance_loss_mlp": 1.06505966,
      "epoch": 0.021042505861840918,
      "flos": 15666353694720.0,
      "grad_norm": 1.8186781240993874,
      "language_loss": 0.90642405,
      "learning_rate": 3.7416082813989552e-06,
      "loss": 0.93127751,
      "num_input_tokens_seen": 3677505,
      "step": 175,
      "time_per_iteration": 2.6648807525634766
    },
    {
      "auxiliary_loss_clip": 0.01406341,
      "auxiliary_loss_mlp": 0.01080207,
      "balance_loss_clip": 1.11812997,
      "balance_loss_mlp": 1.05758095,
      "epoch": 0.02116274875248001,
      "flos": 21142012734720.0,
      "grad_norm": 2.200840193604082,
      "language_loss": 0.8937974,
      "learning_rate": 3.745736189718439e-06,
      "loss": 0.91866285,
      "num_input_tokens_seen": 3696760,
      "step": 176,
      "time_per_iteration": 2.6946747303009033
    },
    {
      "auxiliary_loss_clip": 0.01397038,
      "auxiliary_loss_mlp": 0.01066664,
      "balance_loss_clip": 1.11355984,
      "balance_loss_mlp": 1.04487288,
      "epoch": 0.0212829916431191,
      "flos": 24715770543360.0,
      "grad_norm": 5.890637722855994,
      "language_loss": 0.72764993,
      "learning_rate": 3.749840710329894e-06,
      "loss": 0.75228691,
      "num_input_tokens_seen": 3717465,
      "step": 177,
      "time_per_iteration": 2.7420859336853027
    },
    {
      "auxiliary_loss_clip": 0.01408487,
      "auxiliary_loss_mlp": 0.01086094,
      "balance_loss_clip": 1.11704588,
      "balance_loss_mlp": 1.06185925,
      "epoch": 0.02140323453375819,
      "flos": 16645493508480.0,
      "grad_norm": 3.643850153258798,
      "language_loss": 0.97924364,
      "learning_rate": 3.7539221067588938e-06,
      "loss": 1.00418949,
      "num_input_tokens_seen": 3731440,
      "step": 178,
      "time_per_iteration": 4.4607133865356445
    },
    {
      "auxiliary_loss_clip": 0.01403435,
      "auxiliary_loss_mlp": 0.01083046,
      "balance_loss_clip": 1.11531568,
      "balance_loss_mlp": 1.05970466,
      "epoch": 0.021523477424397284,
      "flos": 20299332689280.0,
      "grad_norm": 3.781526401117534,
      "language_loss": 0.93401647,
      "learning_rate": 3.757980638101964e-06,
      "loss": 0.95888126,
      "num_input_tokens_seen": 3744935,
      "step": 179,
      "time_per_iteration": 3.627661943435669
    },
    {
      "auxiliary_loss_clip": 0.01405726,
      "auxiliary_loss_mlp": 0.01077177,
      "balance_loss_clip": 1.11727667,
      "balance_loss_mlp": 1.05257225,
      "epoch": 0.021643720315036374,
      "flos": 26104005331200.0,
      "grad_norm": 2.490933159583478,
      "language_loss": 0.89204723,
      "learning_rate": 3.7620165591252806e-06,
      "loss": 0.91687626,
      "num_input_tokens_seen": 3763035,
      "step": 180,
      "time_per_iteration": 3.6285765171051025
    },
    {
      "auxiliary_loss_clip": 0.01395999,
      "auxiliary_loss_mlp": 0.01072721,
      "balance_loss_clip": 1.11575317,
      "balance_loss_mlp": 1.05123901,
      "epoch": 0.021763963205675464,
      "flos": 24787663614720.0,
      "grad_norm": 1.9913210565059012,
      "language_loss": 0.94293463,
      "learning_rate": 3.766030120360636e-06,
      "loss": 0.9676218,
      "num_input_tokens_seen": 3782665,
      "step": 181,
      "time_per_iteration": 2.700010061264038
    },
    {
      "auxiliary_loss_clip": 0.01401043,
      "auxiliary_loss_mlp": 0.01075745,
      "balance_loss_clip": 1.11527622,
      "balance_loss_mlp": 1.05423939,
      "epoch": 0.021884206096314557,
      "flos": 25813559957760.0,
      "grad_norm": 2.3901968722110936,
      "language_loss": 0.90185088,
      "learning_rate": 3.7700215681987578e-06,
      "loss": 0.92661875,
      "num_input_tokens_seen": 3802435,
      "step": 182,
      "time_per_iteration": 2.7289299964904785
    },
    {
      "auxiliary_loss_clip": 0.01395113,
      "auxiliary_loss_mlp": 0.01087349,
      "balance_loss_clip": 1.11341381,
      "balance_loss_mlp": 1.0644486,
      "epoch": 0.022004448986953647,
      "flos": 20082719721600.0,
      "grad_norm": 1.9804866748497072,
      "language_loss": 0.8233434,
      "learning_rate": 3.7739911449800767e-06,
      "loss": 0.84816802,
      "num_input_tokens_seen": 3822490,
      "step": 183,
      "time_per_iteration": 2.6960322856903076
    },
    {
      "auxiliary_loss_clip": 0.01396131,
      "auxiliary_loss_mlp": 0.010858,
      "balance_loss_clip": 1.11269391,
      "balance_loss_mlp": 1.06521201,
      "epoch": 0.022124691877592736,
      "flos": 20480609652480.0,
      "grad_norm": 1.8957355256751662,
      "language_loss": 0.80684805,
      "learning_rate": 3.7779390890830114e-06,
      "loss": 0.8316673,
      "num_input_tokens_seen": 3841140,
      "step": 184,
      "time_per_iteration": 2.7065329551696777
    },
    {
      "auxiliary_loss_clip": 0.01396528,
      "auxiliary_loss_mlp": 0.0108184,
      "balance_loss_clip": 1.11215496,
      "balance_loss_mlp": 1.05914295,
      "epoch": 0.02224493476823183,
      "flos": 23586847015680.0,
      "grad_norm": 2.0748493005153703,
      "language_loss": 0.85895932,
      "learning_rate": 3.7818656350098723e-06,
      "loss": 0.88374305,
      "num_input_tokens_seen": 3862090,
      "step": 185,
      "time_per_iteration": 2.7281975746154785
    },
    {
      "auxiliary_loss_clip": 0.01390901,
      "auxiliary_loss_mlp": 0.01078447,
      "balance_loss_clip": 1.10829759,
      "balance_loss_mlp": 1.05503464,
      "epoch": 0.02236517765887092,
      "flos": 16909940413440.0,
      "grad_norm": 2.4439100866056496,
      "language_loss": 0.77353776,
      "learning_rate": 3.7857710134704447e-06,
      "loss": 0.79823124,
      "num_input_tokens_seen": 3881025,
      "step": 186,
      "time_per_iteration": 2.6709048748016357
    },
    {
      "auxiliary_loss_clip": 0.013904,
      "auxiliary_loss_mlp": 0.01059152,
      "balance_loss_clip": 1.11275935,
      "balance_loss_mlp": 1.03851724,
      "epoch": 0.02248542054951001,
      "flos": 43508182930560.0,
      "grad_norm": 2.208342333547801,
      "language_loss": 0.79367304,
      "learning_rate": 3.7896554514633234e-06,
      "loss": 0.81816852,
      "num_input_tokens_seen": 3905310,
      "step": 187,
      "time_per_iteration": 2.8505561351776123
    },
    {
      "auxiliary_loss_clip": 0.01389328,
      "auxiliary_loss_mlp": 0.01071942,
      "balance_loss_clip": 1.1111424,
      "balance_loss_mlp": 1.05090201,
      "epoch": 0.022605663440149103,
      "flos": 23367648268800.0,
      "grad_norm": 2.046542662845841,
      "language_loss": 0.84474427,
      "learning_rate": 3.7935191723550955e-06,
      "loss": 0.86935699,
      "num_input_tokens_seen": 3924265,
      "step": 188,
      "time_per_iteration": 2.704414129257202
    },
    {
      "auxiliary_loss_clip": 0.01388921,
      "auxiliary_loss_mlp": 0.01072141,
      "balance_loss_clip": 1.11036325,
      "balance_loss_mlp": 1.05187535,
      "epoch": 0.022725906330788193,
      "flos": 29019915504000.0,
      "grad_norm": 1.9661297850452932,
      "language_loss": 0.88267815,
      "learning_rate": 3.797362395957408e-06,
      "loss": 0.90728867,
      "num_input_tokens_seen": 3944830,
      "step": 189,
      "time_per_iteration": 2.797250270843506
    },
    {
      "auxiliary_loss_clip": 0.01396887,
      "auxiliary_loss_mlp": 0.01070583,
      "balance_loss_clip": 1.11550355,
      "balance_loss_mlp": 1.04929233,
      "epoch": 0.022846149221427282,
      "flos": 24496176746880.0,
      "grad_norm": 2.295860803886912,
      "language_loss": 0.78015423,
      "learning_rate": 3.8011853386020055e-06,
      "loss": 0.804829,
      "num_input_tokens_seen": 3965735,
      "step": 190,
      "time_per_iteration": 2.715747117996216
    },
    {
      "auxiliary_loss_clip": 0.01392012,
      "auxiliary_loss_mlp": 0.01083549,
      "balance_loss_clip": 1.1126349,
      "balance_loss_mlp": 1.0614121,
      "epoch": 0.022966392112066376,
      "flos": 15523537219200.0,
      "grad_norm": 2.5382561173993743,
      "language_loss": 0.89806026,
      "learning_rate": 3.804988213213804e-06,
      "loss": 0.92281586,
      "num_input_tokens_seen": 3983975,
      "step": 191,
      "time_per_iteration": 2.7323524951934814
    },
    {
      "auxiliary_loss_clip": 0.01366592,
      "auxiliary_loss_mlp": 0.01024245,
      "balance_loss_clip": 1.1445905,
      "balance_loss_mlp": 1.01280105,
      "epoch": 0.023086635002705466,
      "flos": 55650408433920.0,
      "grad_norm": 1.019457335800412,
      "language_loss": 0.63195932,
      "learning_rate": 3.808771229382049e-06,
      "loss": 0.6558677,
      "num_input_tokens_seen": 4043440,
      "step": 192,
      "time_per_iteration": 3.138338088989258
    },
    {
      "auxiliary_loss_clip": 0.01384543,
      "auxiliary_loss_mlp": 0.01081573,
      "balance_loss_clip": 1.11086309,
      "balance_loss_mlp": 1.06174862,
      "epoch": 0.023206877893344555,
      "flos": 19313441118720.0,
      "grad_norm": 2.0472775462233774,
      "language_loss": 0.84434521,
      "learning_rate": 3.8125345934296324e-06,
      "loss": 0.8690064,
      "num_input_tokens_seen": 4061750,
      "step": 193,
      "time_per_iteration": 2.6477746963500977
    },
    {
      "auxiliary_loss_clip": 0.01385895,
      "auxiliary_loss_mlp": 0.01076271,
      "balance_loss_clip": 1.10987711,
      "balance_loss_mlp": 1.05412173,
      "epoch": 0.02332712078398365,
      "flos": 23072965090560.0,
      "grad_norm": 1.9798519364784741,
      "language_loss": 0.88082647,
      "learning_rate": 3.81627850848061e-06,
      "loss": 0.9054482,
      "num_input_tokens_seen": 4082345,
      "step": 194,
      "time_per_iteration": 2.7456421852111816
    },
    {
      "auxiliary_loss_clip": 0.01380445,
      "auxiliary_loss_mlp": 0.010664,
      "balance_loss_clip": 1.10619831,
      "balance_loss_mlp": 1.04623008,
      "epoch": 0.02344736367462274,
      "flos": 24425971614720.0,
      "grad_norm": 2.141720749716008,
      "language_loss": 0.86126029,
      "learning_rate": 3.820003174525994e-06,
      "loss": 0.88572866,
      "num_input_tokens_seen": 4101770,
      "step": 195,
      "time_per_iteration": 2.689518451690674
    },
    {
      "auxiliary_loss_clip": 0.01384382,
      "auxiliary_loss_mlp": 0.01078606,
      "balance_loss_clip": 1.11095476,
      "balance_loss_mlp": 1.05815017,
      "epoch": 0.02356760656526183,
      "flos": 21579799697280.0,
      "grad_norm": 2.349333882713823,
      "language_loss": 0.82741582,
      "learning_rate": 3.823708788487851e-06,
      "loss": 0.85204566,
      "num_input_tokens_seen": 4118770,
      "step": 196,
      "time_per_iteration": 2.693510055541992
    },
    {
      "auxiliary_loss_clip": 0.01380327,
      "auxiliary_loss_mlp": 0.01082142,
      "balance_loss_clip": 1.10832953,
      "balance_loss_mlp": 1.06287766,
      "epoch": 0.02368784945590092,
      "flos": 25193598192000.0,
      "grad_norm": 2.093739907396898,
      "language_loss": 0.84406555,
      "learning_rate": 3.827395544281781e-06,
      "loss": 0.86869025,
      "num_input_tokens_seen": 4141110,
      "step": 197,
      "time_per_iteration": 2.791916847229004
    },
    {
      "auxiliary_loss_clip": 0.01386369,
      "auxiliary_loss_mlp": 0.01082311,
      "balance_loss_clip": 1.11139083,
      "balance_loss_mlp": 1.06226015,
      "epoch": 0.02380809234654001,
      "flos": 27562481164800.0,
      "grad_norm": 2.01150832375434,
      "language_loss": 0.79083872,
      "learning_rate": 3.831063632877802e-06,
      "loss": 0.81552553,
      "num_input_tokens_seen": 4161430,
      "step": 198,
      "time_per_iteration": 2.723952293395996
    },
    {
      "auxiliary_loss_clip": 0.0138267,
      "auxiliary_loss_mlp": 0.01075363,
      "balance_loss_clip": 1.11459041,
      "balance_loss_mlp": 1.05687332,
      "epoch": 0.0239283352371791,
      "flos": 18259786540800.0,
      "grad_norm": 3.418963258664074,
      "language_loss": 0.75713021,
      "learning_rate": 3.834713242359712e-06,
      "loss": 0.78171051,
      "num_input_tokens_seen": 4179260,
      "step": 199,
      "time_per_iteration": 2.6102654933929443
    },
    {
      "auxiliary_loss_clip": 0.01385329,
      "auxiliary_loss_mlp": 0.01074603,
      "balance_loss_clip": 1.10850573,
      "balance_loss_mlp": 1.05269265,
      "epoch": 0.02404857812781819,
      "flos": 21395110942080.0,
      "grad_norm": 2.7251068765085233,
      "language_loss": 0.8725813,
      "learning_rate": 3.838344557982959e-06,
      "loss": 0.89718068,
      "num_input_tokens_seen": 4200640,
      "step": 200,
      "time_per_iteration": 2.8857808113098145
    },
    {
      "auxiliary_loss_clip": 0.01379306,
      "auxiliary_loss_mlp": 0.0107856,
      "balance_loss_clip": 1.10760927,
      "balance_loss_mlp": 1.05723369,
      "epoch": 0.024168821018457284,
      "flos": 16654256426880.0,
      "grad_norm": 7.022232063743792,
      "language_loss": 0.85146761,
      "learning_rate": 3.841957762231063e-06,
      "loss": 0.8760463,
      "num_input_tokens_seen": 4218170,
      "step": 201,
      "time_per_iteration": 2.6776058673858643
    },
    {
      "auxiliary_loss_clip": 0.01376146,
      "auxiliary_loss_mlp": 0.01064694,
      "balance_loss_clip": 1.10541153,
      "balance_loss_mlp": 1.04464269,
      "epoch": 0.024289063909096374,
      "flos": 22820872464000.0,
      "grad_norm": 1.8792648746036849,
      "language_loss": 0.87722462,
      "learning_rate": 3.8455530348706454e-06,
      "loss": 0.90163296,
      "num_input_tokens_seen": 4237770,
      "step": 202,
      "time_per_iteration": 2.6722023487091064
    },
    {
      "auxiliary_loss_clip": 0.01375799,
      "auxiliary_loss_mlp": 0.01076925,
      "balance_loss_clip": 1.10678399,
      "balance_loss_mlp": 1.05823255,
      "epoch": 0.024409306799735464,
      "flos": 17748598135680.0,
      "grad_norm": 1.9419002155781018,
      "language_loss": 0.77178466,
      "learning_rate": 3.849130553005099e-06,
      "loss": 0.79631186,
      "num_input_tokens_seen": 4255985,
      "step": 203,
      "time_per_iteration": 2.6749582290649414
    },
    {
      "auxiliary_loss_clip": 0.01377384,
      "auxiliary_loss_mlp": 0.01067638,
      "balance_loss_clip": 1.10488939,
      "balance_loss_mlp": 1.04869604,
      "epoch": 0.024529549690374557,
      "flos": 21616213109760.0,
      "grad_norm": 1.8737592631250246,
      "language_loss": 0.83649242,
      "learning_rate": 3.852690491126933e-06,
      "loss": 0.86094266,
      "num_input_tokens_seen": 4276035,
      "step": 204,
      "time_per_iteration": 2.681607723236084
    },
    {
      "auxiliary_loss_clip": 0.01371668,
      "auxiliary_loss_mlp": 0.01062326,
      "balance_loss_clip": 1.10142112,
      "balance_loss_mlp": 1.04207253,
      "epoch": 0.024649792581013647,
      "flos": 25551662918400.0,
      "grad_norm": 2.47162084469832,
      "language_loss": 0.91275615,
      "learning_rate": 3.856233021168845e-06,
      "loss": 0.93709612,
      "num_input_tokens_seen": 4295730,
      "step": 205,
      "time_per_iteration": 4.479662895202637
    },
    {
      "auxiliary_loss_clip": 0.01365889,
      "auxiliary_loss_mlp": 0.01055111,
      "balance_loss_clip": 1.10214317,
      "balance_loss_mlp": 1.03745592,
      "epoch": 0.024770035471652737,
      "flos": 34495574544000.0,
      "grad_norm": 2.66739390471763,
      "language_loss": 0.91206568,
      "learning_rate": 3.859758312553544e-06,
      "loss": 0.93627572,
      "num_input_tokens_seen": 4317950,
      "step": 206,
      "time_per_iteration": 3.6736178398132324
    },
    {
      "auxiliary_loss_clip": 0.01375517,
      "auxiliary_loss_mlp": 0.01069948,
      "balance_loss_clip": 1.10803223,
      "balance_loss_mlp": 1.05174434,
      "epoch": 0.02489027836229183,
      "flos": 21505428587520.0,
      "grad_norm": 2.143360921851919,
      "language_loss": 0.9198271,
      "learning_rate": 3.8632665322423735e-06,
      "loss": 0.94428176,
      "num_input_tokens_seen": 4337605,
      "step": 207,
      "time_per_iteration": 3.570754289627075
    },
    {
      "auxiliary_loss_clip": 0.01373021,
      "auxiliary_loss_mlp": 0.01068159,
      "balance_loss_clip": 1.10509872,
      "balance_loss_mlp": 1.04888248,
      "epoch": 0.02501052125293092,
      "flos": 23219013790080.0,
      "grad_norm": 1.946605226864214,
      "language_loss": 0.86058128,
      "learning_rate": 3.866757844782762e-06,
      "loss": 0.88499308,
      "num_input_tokens_seen": 4358110,
      "step": 208,
      "time_per_iteration": 2.719597578048706
    },
    {
      "auxiliary_loss_clip": 0.01372329,
      "auxiliary_loss_mlp": 0.01064111,
      "balance_loss_clip": 1.10602355,
      "balance_loss_mlp": 1.0451808,
      "epoch": 0.02513076414357001,
      "flos": 26388920010240.0,
      "grad_norm": 2.39019503640336,
      "language_loss": 0.91268545,
      "learning_rate": 3.870232412354527e-06,
      "loss": 0.93704981,
      "num_input_tokens_seen": 4374955,
      "step": 209,
      "time_per_iteration": 2.6946091651916504
    },
    {
      "auxiliary_loss_clip": 0.0136794,
      "auxiliary_loss_mlp": 0.01062786,
      "balance_loss_clip": 1.10257041,
      "balance_loss_mlp": 1.04415369,
      "epoch": 0.025251007034209103,
      "flos": 13590430047360.0,
      "grad_norm": 2.040828224703171,
      "language_loss": 0.92758226,
      "learning_rate": 3.873690394815086e-06,
      "loss": 0.95188951,
      "num_input_tokens_seen": 4391535,
      "step": 210,
      "time_per_iteration": 2.666837453842163
    },
    {
      "auxiliary_loss_clip": 0.01365956,
      "auxiliary_loss_mlp": 0.01058304,
      "balance_loss_clip": 1.09952545,
      "balance_loss_mlp": 1.03999305,
      "epoch": 0.025371249924848193,
      "flos": 15049229103360.0,
      "grad_norm": 2.7506884763230195,
      "language_loss": 0.91248894,
      "learning_rate": 3.877131949743587e-06,
      "loss": 0.93673152,
      "num_input_tokens_seen": 4408400,
      "step": 211,
      "time_per_iteration": 2.6792168617248535
    },
    {
      "auxiliary_loss_clip": 0.01366826,
      "auxiliary_loss_mlp": 0.01076965,
      "balance_loss_clip": 1.10248852,
      "balance_loss_mlp": 1.05823767,
      "epoch": 0.025491492815487283,
      "flos": 25553853648000.0,
      "grad_norm": 2.2247067782808903,
      "language_loss": 0.78012383,
      "learning_rate": 3.880557232483993e-06,
      "loss": 0.80456173,
      "num_input_tokens_seen": 4427840,
      "step": 212,
      "time_per_iteration": 2.7173378467559814
    },
    {
      "auxiliary_loss_clip": 0.01365236,
      "auxiliary_loss_mlp": 0.01061008,
      "balance_loss_clip": 1.09925103,
      "balance_loss_mlp": 1.0420891,
      "epoch": 0.025611735706126376,
      "flos": 20630752502400.0,
      "grad_norm": 1.8829296350748523,
      "language_loss": 0.86684024,
      "learning_rate": 3.883966396187164e-06,
      "loss": 0.89110267,
      "num_input_tokens_seen": 4447110,
      "step": 213,
      "time_per_iteration": 2.687816619873047
    },
    {
      "auxiliary_loss_clip": 0.0136719,
      "auxiliary_loss_mlp": 0.01062258,
      "balance_loss_clip": 1.10321426,
      "balance_loss_mlp": 1.04429281,
      "epoch": 0.025731978596765466,
      "flos": 19062282245760.0,
      "grad_norm": 2.1154463439456426,
      "language_loss": 0.89913821,
      "learning_rate": 3.887359591851937e-06,
      "loss": 0.92343271,
      "num_input_tokens_seen": 4464715,
      "step": 214,
      "time_per_iteration": 2.6238648891448975
    },
    {
      "auxiliary_loss_clip": 0.01362074,
      "auxiliary_loss_mlp": 0.01055667,
      "balance_loss_clip": 1.10065579,
      "balance_loss_mlp": 1.03692698,
      "epoch": 0.025852221487404556,
      "flos": 22163814927360.0,
      "grad_norm": 1.5663418666755529,
      "language_loss": 0.9221977,
      "learning_rate": 3.890736968365265e-06,
      "loss": 0.94637513,
      "num_input_tokens_seen": 4485030,
      "step": 215,
      "time_per_iteration": 2.6874797344207764
    },
    {
      "auxiliary_loss_clip": 0.01362846,
      "auxiliary_loss_mlp": 0.01059668,
      "balance_loss_clip": 1.09907341,
      "balance_loss_mlp": 1.03993893,
      "epoch": 0.02597246437804365,
      "flos": 26541971861760.0,
      "grad_norm": 1.7772951099778618,
      "language_loss": 0.8474251,
      "learning_rate": 3.894098672541412e-06,
      "loss": 0.87165022,
      "num_input_tokens_seen": 4505935,
      "step": 216,
      "time_per_iteration": 2.667559862136841
    },
    {
      "auxiliary_loss_clip": 0.01362865,
      "auxiliary_loss_mlp": 0.01067293,
      "balance_loss_clip": 1.09939265,
      "balance_loss_mlp": 1.04748058,
      "epoch": 0.02609270726868274,
      "flos": 32671671696000.0,
      "grad_norm": 2.9008734623247387,
      "language_loss": 0.7528137,
      "learning_rate": 3.89744484916025e-06,
      "loss": 0.77711535,
      "num_input_tokens_seen": 4527045,
      "step": 217,
      "time_per_iteration": 2.748222827911377
    },
    {
      "auxiliary_loss_clip": 0.013646,
      "auxiliary_loss_mlp": 0.01066196,
      "balance_loss_clip": 1.10040259,
      "balance_loss_mlp": 1.04651463,
      "epoch": 0.02621295015932183,
      "flos": 26243553669120.0,
      "grad_norm": 2.1982813953729865,
      "language_loss": 0.87439555,
      "learning_rate": 3.900775641004673e-06,
      "loss": 0.89870346,
      "num_input_tokens_seen": 4546360,
      "step": 218,
      "time_per_iteration": 2.6575980186462402
    },
    {
      "auxiliary_loss_clip": 0.01369246,
      "auxiliary_loss_mlp": 0.01072522,
      "balance_loss_clip": 1.10237503,
      "balance_loss_mlp": 1.05022979,
      "epoch": 0.026333193049960922,
      "flos": 42921402353280.0,
      "grad_norm": 2.5826404410805135,
      "language_loss": 0.74340618,
      "learning_rate": 3.904091188897156e-06,
      "loss": 0.76782382,
      "num_input_tokens_seen": 4565495,
      "step": 219,
      "time_per_iteration": 2.829437732696533
    },
    {
      "auxiliary_loss_clip": 0.01361316,
      "auxiliary_loss_mlp": 0.01070768,
      "balance_loss_clip": 1.09911013,
      "balance_loss_mlp": 1.05100322,
      "epoch": 0.026453435940600012,
      "flos": 17963846386560.0,
      "grad_norm": 2.2013609139393298,
      "language_loss": 0.81786919,
      "learning_rate": 3.90739163173548e-06,
      "loss": 0.84219009,
      "num_input_tokens_seen": 4583330,
      "step": 220,
      "time_per_iteration": 2.6075642108917236
    },
    {
      "auxiliary_loss_clip": 0.01359395,
      "auxiliary_loss_mlp": 0.01069048,
      "balance_loss_clip": 1.09874642,
      "balance_loss_mlp": 1.05032051,
      "epoch": 0.026573678831239102,
      "flos": 18984319776000.0,
      "grad_norm": 2.638861950572057,
      "language_loss": 0.88458073,
      "learning_rate": 3.910677106527646e-06,
      "loss": 0.90886515,
      "num_input_tokens_seen": 4600520,
      "step": 221,
      "time_per_iteration": 2.66658353805542
    },
    {
      "auxiliary_loss_clip": 0.01356406,
      "auxiliary_loss_mlp": 0.01069392,
      "balance_loss_clip": 1.09786618,
      "balance_loss_mlp": 1.05161786,
      "epoch": 0.026693921721878195,
      "flos": 29241448634880.0,
      "grad_norm": 2.2766940253418735,
      "language_loss": 0.84200555,
      "learning_rate": 3.913947748426004e-06,
      "loss": 0.86626351,
      "num_input_tokens_seen": 4617340,
      "step": 222,
      "time_per_iteration": 2.6975743770599365
    },
    {
      "auxiliary_loss_clip": 0.01361456,
      "auxiliary_loss_mlp": 0.01067968,
      "balance_loss_clip": 1.10050893,
      "balance_loss_mlp": 1.05014634,
      "epoch": 0.026814164612517285,
      "flos": 14128083797760.0,
      "grad_norm": 4.542766593496982,
      "language_loss": 0.76505029,
      "learning_rate": 3.9172036907606136e-06,
      "loss": 0.78934455,
      "num_input_tokens_seen": 4630820,
      "step": 223,
      "time_per_iteration": 2.682870626449585
    },
    {
      "auxiliary_loss_clip": 0.01360816,
      "auxiliary_loss_mlp": 0.01064611,
      "balance_loss_clip": 1.09793222,
      "balance_loss_mlp": 1.04610991,
      "epoch": 0.026934407503156375,
      "flos": 23511973115520.0,
      "grad_norm": 1.7291655506609105,
      "language_loss": 0.94936579,
      "learning_rate": 3.920445065071855e-06,
      "loss": 0.97362012,
      "num_input_tokens_seen": 4651985,
      "step": 224,
      "time_per_iteration": 2.7348005771636963
    },
    {
      "auxiliary_loss_clip": 0.01356761,
      "auxiliary_loss_mlp": 0.01068894,
      "balance_loss_clip": 1.09770906,
      "balance_loss_mlp": 1.0503335,
      "epoch": 0.027054650393795468,
      "flos": 28950356816640.0,
      "grad_norm": 2.5008515101344293,
      "language_loss": 0.80043864,
      "learning_rate": 3.923672001142322e-06,
      "loss": 0.82469517,
      "num_input_tokens_seen": 4672295,
      "step": 225,
      "time_per_iteration": 2.6929314136505127
    },
    {
      "auxiliary_loss_clip": 0.01353823,
      "auxiliary_loss_mlp": 0.01076394,
      "balance_loss_clip": 1.0963968,
      "balance_loss_mlp": 1.05753517,
      "epoch": 0.027174893284434558,
      "flos": 31431568596480.0,
      "grad_norm": 1.8133559667111454,
      "language_loss": 0.84260684,
      "learning_rate": 3.926884627027996e-06,
      "loss": 0.86690903,
      "num_input_tokens_seen": 4696065,
      "step": 226,
      "time_per_iteration": 2.778541088104248
    },
    {
      "auxiliary_loss_clip": 0.01355022,
      "auxiliary_loss_mlp": 0.01069103,
      "balance_loss_clip": 1.09557915,
      "balance_loss_mlp": 1.05119801,
      "epoch": 0.027295136175073648,
      "flos": 22054466949120.0,
      "grad_norm": 2.085147031556694,
      "language_loss": 0.77417493,
      "learning_rate": 3.930083069088744e-06,
      "loss": 0.79841614,
      "num_input_tokens_seen": 4716065,
      "step": 227,
      "time_per_iteration": 2.6528007984161377
    },
    {
      "auxiliary_loss_clip": 0.01320708,
      "auxiliary_loss_mlp": 0.01010348,
      "balance_loss_clip": 1.11756825,
      "balance_loss_mlp": 0.99899924,
      "epoch": 0.02741537906571274,
      "flos": 60800752972800.0,
      "grad_norm": 2.1687144181246225,
      "language_loss": 0.59331959,
      "learning_rate": 3.933267452018137e-06,
      "loss": 0.61663014,
      "num_input_tokens_seen": 4775860,
      "step": 228,
      "time_per_iteration": 3.183230400085449
    },
    {
      "auxiliary_loss_clip": 0.0135363,
      "auxiliary_loss_mlp": 0.01061196,
      "balance_loss_clip": 1.09726095,
      "balance_loss_mlp": 1.04289734,
      "epoch": 0.02753562195635183,
      "flos": 24606278910720.0,
      "grad_norm": 2.2596565134025766,
      "language_loss": 0.84328228,
      "learning_rate": 3.936437898872622e-06,
      "loss": 0.86743057,
      "num_input_tokens_seen": 4795835,
      "step": 229,
      "time_per_iteration": 2.6772241592407227
    },
    {
      "auxiliary_loss_clip": 0.01353822,
      "auxiliary_loss_mlp": 0.01055797,
      "balance_loss_clip": 1.09660125,
      "balance_loss_mlp": 1.03840446,
      "epoch": 0.02765586484699092,
      "flos": 34094236907520.0,
      "grad_norm": 2.981695398685085,
      "language_loss": 0.7989105,
      "learning_rate": 3.9395945311000525e-06,
      "loss": 0.82300675,
      "num_input_tokens_seen": 4817460,
      "step": 230,
      "time_per_iteration": 2.772352695465088
    },
    {
      "auxiliary_loss_clip": 0.0135492,
      "auxiliary_loss_mlp": 0.01069802,
      "balance_loss_clip": 1.09727359,
      "balance_loss_mlp": 1.05175352,
      "epoch": 0.027776107737630014,
      "flos": 14829922615680.0,
      "grad_norm": 2.1114948249289736,
      "language_loss": 0.90723765,
      "learning_rate": 3.942737468567608e-06,
      "loss": 0.93148482,
      "num_input_tokens_seen": 4835475,
      "step": 231,
      "time_per_iteration": 2.5835177898406982
    },
    {
      "auxiliary_loss_clip": 0.0135314,
      "auxiliary_loss_mlp": 0.0106867,
      "balance_loss_clip": 1.09692466,
      "balance_loss_mlp": 1.05099118,
      "epoch": 0.027896350628269104,
      "flos": 47920347066240.0,
      "grad_norm": 2.0857639582128105,
      "language_loss": 0.86027831,
      "learning_rate": 3.9458668295891026e-06,
      "loss": 0.88449639,
      "num_input_tokens_seen": 4857760,
      "step": 232,
      "time_per_iteration": 2.880688428878784
    },
    {
      "auxiliary_loss_clip": 0.0134801,
      "auxiliary_loss_mlp": 0.01061239,
      "balance_loss_clip": 1.09144175,
      "balance_loss_mlp": 1.04220176,
      "epoch": 0.028016593518908194,
      "flos": 21684550734720.0,
      "grad_norm": 2.39938083872044,
      "language_loss": 0.86896724,
      "learning_rate": 3.948982730951712e-06,
      "loss": 0.89305973,
      "num_input_tokens_seen": 4875855,
      "step": 233,
      "time_per_iteration": 3.616084337234497
    },
    {
      "auxiliary_loss_clip": 0.01351798,
      "auxiliary_loss_mlp": 0.0106299,
      "balance_loss_clip": 1.09434104,
      "balance_loss_mlp": 1.04422688,
      "epoch": 0.028136836409547287,
      "flos": 18439483305600.0,
      "grad_norm": 2.403557458972179,
      "language_loss": 0.81858885,
      "learning_rate": 3.9520852879421254e-06,
      "loss": 0.84273672,
      "num_input_tokens_seen": 4893200,
      "step": 234,
      "time_per_iteration": 3.555677652359009
    },
    {
      "auxiliary_loss_clip": 0.01348607,
      "auxiliary_loss_mlp": 0.0106572,
      "balance_loss_clip": 1.09525013,
      "balance_loss_mlp": 1.04867315,
      "epoch": 0.028257079300186377,
      "flos": 31576934937600.0,
      "grad_norm": 2.5765402240824424,
      "language_loss": 0.81788802,
      "learning_rate": 3.955174614372137e-06,
      "loss": 0.84203136,
      "num_input_tokens_seen": 4912965,
      "step": 235,
      "time_per_iteration": 3.6337788105010986
    },
    {
      "auxiliary_loss_clip": 0.01350165,
      "auxiliary_loss_mlp": 0.01067378,
      "balance_loss_clip": 1.09546709,
      "balance_loss_mlp": 1.0491147,
      "epoch": 0.028377322190825467,
      "flos": 23513337832320.0,
      "grad_norm": 2.071324894125773,
      "language_loss": 0.84490645,
      "learning_rate": 3.9582508226037045e-06,
      "loss": 0.86908191,
      "num_input_tokens_seen": 4933105,
      "step": 236,
      "time_per_iteration": 2.7000463008880615
    },
    {
      "auxiliary_loss_clip": 0.01356313,
      "auxiliary_loss_mlp": 0.01068063,
      "balance_loss_clip": 1.09652829,
      "balance_loss_mlp": 1.04903674,
      "epoch": 0.02849756508146456,
      "flos": 20479604071680.0,
      "grad_norm": 2.5930975429792977,
      "language_loss": 0.9417783,
      "learning_rate": 3.9613140235734636e-06,
      "loss": 0.96602201,
      "num_input_tokens_seen": 4950085,
      "step": 237,
      "time_per_iteration": 2.6634085178375244
    },
    {
      "auxiliary_loss_clip": 0.01347567,
      "auxiliary_loss_mlp": 0.0106062,
      "balance_loss_clip": 1.09272861,
      "balance_loss_mlp": 1.04184484,
      "epoch": 0.02861780797210365,
      "flos": 14283362292480.0,
      "grad_norm": 1.9672334233832236,
      "language_loss": 0.81337488,
      "learning_rate": 3.96436432681674e-06,
      "loss": 0.83745676,
      "num_input_tokens_seen": 4968075,
      "step": 238,
      "time_per_iteration": 2.6750121116638184
    },
    {
      "auxiliary_loss_clip": 0.01347413,
      "auxiliary_loss_mlp": 0.01065276,
      "balance_loss_clip": 1.09356093,
      "balance_loss_mlp": 1.04700136,
      "epoch": 0.02873805086274274,
      "flos": 25808532053760.0,
      "grad_norm": 2.0800939110143988,
      "language_loss": 0.89090914,
      "learning_rate": 3.967401840491044e-06,
      "loss": 0.91503596,
      "num_input_tokens_seen": 4987355,
      "step": 239,
      "time_per_iteration": 2.6861329078674316
    },
    {
      "auxiliary_loss_clip": 0.01346179,
      "auxiliary_loss_mlp": 0.01063687,
      "balance_loss_clip": 1.09588623,
      "balance_loss_mlp": 1.04755843,
      "epoch": 0.028858293753381833,
      "flos": 17304238984320.0,
      "grad_norm": 2.1099576972006524,
      "language_loss": 0.87773097,
      "learning_rate": 3.97042667139909e-06,
      "loss": 0.90182966,
      "num_input_tokens_seen": 5004680,
      "step": 240,
      "time_per_iteration": 2.651397943496704
    },
    {
      "auxiliary_loss_clip": 0.01347398,
      "auxiliary_loss_mlp": 0.01058355,
      "balance_loss_clip": 1.0950501,
      "balance_loss_mlp": 1.04084277,
      "epoch": 0.028978536644020923,
      "flos": 23038347358080.0,
      "grad_norm": 2.1294892146847517,
      "language_loss": 0.87479508,
      "learning_rate": 3.973438925011327e-06,
      "loss": 0.89885259,
      "num_input_tokens_seen": 5022965,
      "step": 241,
      "time_per_iteration": 2.6352479457855225
    },
    {
      "auxiliary_loss_clip": 0.0134648,
      "auxiliary_loss_mlp": 0.01052387,
      "balance_loss_clip": 1.09219146,
      "balance_loss_mlp": 1.03432679,
      "epoch": 0.029098779534660012,
      "flos": 28329712692480.0,
      "grad_norm": 2.383613682826015,
      "language_loss": 0.91311383,
      "learning_rate": 3.976438705488002e-06,
      "loss": 0.93710256,
      "num_input_tokens_seen": 5042625,
      "step": 242,
      "time_per_iteration": 2.7367398738861084
    },
    {
      "auxiliary_loss_clip": 0.0134489,
      "auxiliary_loss_mlp": 0.01060547,
      "balance_loss_clip": 1.09476411,
      "balance_loss_mlp": 1.0441792,
      "epoch": 0.029219022425299106,
      "flos": 13881665520000.0,
      "grad_norm": 3.126318453379722,
      "language_loss": 0.93549973,
      "learning_rate": 3.9794261157007744e-06,
      "loss": 0.95955408,
      "num_input_tokens_seen": 5060380,
      "step": 243,
      "time_per_iteration": 2.6801209449768066
    },
    {
      "auxiliary_loss_clip": 0.01349412,
      "auxiliary_loss_mlp": 0.01056293,
      "balance_loss_clip": 1.09561324,
      "balance_loss_mlp": 1.03749371,
      "epoch": 0.029339265315938196,
      "flos": 19422501788160.0,
      "grad_norm": 2.20913497152989,
      "language_loss": 0.84903252,
      "learning_rate": 3.982401257253887e-06,
      "loss": 0.87308955,
      "num_input_tokens_seen": 5078720,
      "step": 244,
      "time_per_iteration": 2.724752426147461
    },
    {
      "auxiliary_loss_clip": 0.01346235,
      "auxiliary_loss_mlp": 0.01057312,
      "balance_loss_clip": 1.0935452,
      "balance_loss_mlp": 1.04055107,
      "epoch": 0.029459508206577285,
      "flos": 15669550005120.0,
      "grad_norm": 2.7076614978858795,
      "language_loss": 0.89816201,
      "learning_rate": 3.985364230504893e-06,
      "loss": 0.92219758,
      "num_input_tokens_seen": 5096605,
      "step": 245,
      "time_per_iteration": 2.645578145980835
    },
    {
      "auxiliary_loss_clip": 0.01352941,
      "auxiliary_loss_mlp": 0.01061706,
      "balance_loss_clip": 1.0989815,
      "balance_loss_mlp": 1.04527915,
      "epoch": 0.02957975109721638,
      "flos": 28220975245440.0,
      "grad_norm": 2.255953792905796,
      "language_loss": 0.84478939,
      "learning_rate": 3.988315134584976e-06,
      "loss": 0.86893582,
      "num_input_tokens_seen": 5116285,
      "step": 246,
      "time_per_iteration": 2.714426040649414
    },
    {
      "auxiliary_loss_clip": 0.01349014,
      "auxiliary_loss_mlp": 0.01068267,
      "balance_loss_clip": 1.09581792,
      "balance_loss_mlp": 1.05064821,
      "epoch": 0.02969999398785547,
      "flos": 24315869450880.0,
      "grad_norm": 1.8513024819139392,
      "language_loss": 0.80267406,
      "learning_rate": 3.991254067418851e-06,
      "loss": 0.8268469,
      "num_input_tokens_seen": 5136825,
      "step": 247,
      "time_per_iteration": 2.700528144836426
    },
    {
      "auxiliary_loss_clip": 0.01340779,
      "auxiliary_loss_mlp": 0.01063345,
      "balance_loss_clip": 1.09398198,
      "balance_loss_mlp": 1.04690647,
      "epoch": 0.02982023687849456,
      "flos": 35078584193280.0,
      "grad_norm": 2.459161173673978,
      "language_loss": 0.82921481,
      "learning_rate": 3.994181125744254e-06,
      "loss": 0.85325605,
      "num_input_tokens_seen": 5158630,
      "step": 248,
      "time_per_iteration": 2.7382946014404297
    },
    {
      "auxiliary_loss_clip": 0.01342814,
      "auxiliary_loss_mlp": 0.0105717,
      "balance_loss_clip": 1.09304357,
      "balance_loss_mlp": 1.04091001,
      "epoch": 0.02994047976913365,
      "flos": 26177155378560.0,
      "grad_norm": 1.8717818801919344,
      "language_loss": 0.73972654,
      "learning_rate": 3.99709640513106e-06,
      "loss": 0.76372635,
      "num_input_tokens_seen": 5179510,
      "step": 249,
      "time_per_iteration": 2.7715437412261963
    },
    {
      "auxiliary_loss_clip": 0.01343035,
      "auxiliary_loss_mlp": 0.01073497,
      "balance_loss_clip": 1.09000385,
      "balance_loss_mlp": 1.05515051,
      "epoch": 0.03006072265977274,
      "flos": 25625028447360.0,
      "grad_norm": 3.8111200713277404,
      "language_loss": 0.85284364,
      "learning_rate": 4e-06,
      "loss": 0.87700897,
      "num_input_tokens_seen": 5199345,
      "step": 250,
      "time_per_iteration": 2.6763250827789307
    },
    {
      "auxiliary_loss_clip": 0.01345063,
      "auxiliary_loss_mlp": 0.01055208,
      "balance_loss_clip": 1.09447062,
      "balance_loss_mlp": 1.03881669,
      "epoch": 0.03018096555041183,
      "flos": 22127078292480.0,
      "grad_norm": 2.489464104546009,
      "language_loss": 0.88150907,
      "learning_rate": 3.999999848300794e-06,
      "loss": 0.90551174,
      "num_input_tokens_seen": 5218330,
      "step": 251,
      "time_per_iteration": 2.668487787246704
    },
    {
      "auxiliary_loss_clip": 0.01337663,
      "auxiliary_loss_mlp": 0.01055439,
      "balance_loss_clip": 1.08884287,
      "balance_loss_mlp": 1.03869033,
      "epoch": 0.030301208441050925,
      "flos": 30188197359360.0,
      "grad_norm": 2.501745407821338,
      "language_loss": 0.89069784,
      "learning_rate": 3.999999393203203e-06,
      "loss": 0.91462886,
      "num_input_tokens_seen": 5240740,
      "step": 252,
      "time_per_iteration": 2.713688850402832
    },
    {
      "auxiliary_loss_clip": 0.01337698,
      "auxiliary_loss_mlp": 0.01056093,
      "balance_loss_clip": 1.08726346,
      "balance_loss_mlp": 1.0396421,
      "epoch": 0.030421451331690014,
      "flos": 23621392920960.0,
      "grad_norm": 1.8878167373114878,
      "language_loss": 0.85260916,
      "learning_rate": 3.999998634707293e-06,
      "loss": 0.8765471,
      "num_input_tokens_seen": 5260290,
      "step": 253,
      "time_per_iteration": 2.639984369277954
    },
    {
      "auxiliary_loss_clip": 0.01346001,
      "auxiliary_loss_mlp": 0.01061992,
      "balance_loss_clip": 1.09486198,
      "balance_loss_mlp": 1.04492116,
      "epoch": 0.030541694222329104,
      "flos": 27928446883200.0,
      "grad_norm": 2.4625703894967526,
      "language_loss": 0.96464044,
      "learning_rate": 3.999997572813182e-06,
      "loss": 0.98872036,
      "num_input_tokens_seen": 5278100,
      "step": 254,
      "time_per_iteration": 2.681966781616211
    },
    {
      "auxiliary_loss_clip": 0.01340474,
      "auxiliary_loss_mlp": 0.01063252,
      "balance_loss_clip": 1.08974671,
      "balance_loss_mlp": 1.04661071,
      "epoch": 0.030661937112968194,
      "flos": 18588441006720.0,
      "grad_norm": 2.0919668484401743,
      "language_loss": 0.87561178,
      "learning_rate": 3.999996207521028e-06,
      "loss": 0.89964902,
      "num_input_tokens_seen": 5296810,
      "step": 255,
      "time_per_iteration": 2.6863253116607666
    },
    {
      "auxiliary_loss_clip": 0.01343039,
      "auxiliary_loss_mlp": 0.01055336,
      "balance_loss_clip": 1.0887785,
      "balance_loss_mlp": 1.03722847,
      "epoch": 0.030782180003607287,
      "flos": 12969139478400.0,
      "grad_norm": 2.3039757460560684,
      "language_loss": 0.82090002,
      "learning_rate": 3.999994538831039e-06,
      "loss": 0.84488374,
      "num_input_tokens_seen": 5313395,
      "step": 256,
      "time_per_iteration": 2.747922420501709
    },
    {
      "auxiliary_loss_clip": 0.01340643,
      "auxiliary_loss_mlp": 0.01055926,
      "balance_loss_clip": 1.08976531,
      "balance_loss_mlp": 1.03774619,
      "epoch": 0.030902422894246377,
      "flos": 23335364920320.0,
      "grad_norm": 2.7060263427939404,
      "language_loss": 0.85825264,
      "learning_rate": 3.99999256674347e-06,
      "loss": 0.8822183,
      "num_input_tokens_seen": 5333545,
      "step": 257,
      "time_per_iteration": 2.744527816772461
    },
    {
      "auxiliary_loss_clip": 0.01298547,
      "auxiliary_loss_mlp": 0.01013924,
      "balance_loss_clip": 1.10362279,
      "balance_loss_mlp": 1.00219393,
      "epoch": 0.031022665784885467,
      "flos": 55094151438720.0,
      "grad_norm": 1.0258928603402335,
      "language_loss": 0.53582597,
      "learning_rate": 3.999990291258618e-06,
      "loss": 0.55895066,
      "num_input_tokens_seen": 5392235,
      "step": 258,
      "time_per_iteration": 3.1286189556121826
    },
    {
      "auxiliary_loss_clip": 0.01340155,
      "auxiliary_loss_mlp": 0.01057391,
      "balance_loss_clip": 1.09021556,
      "balance_loss_mlp": 1.04045105,
      "epoch": 0.03114290867552456,
      "flos": 19317786664320.0,
      "grad_norm": 2.9054715744891326,
      "language_loss": 0.86667788,
      "learning_rate": 3.999987712376829e-06,
      "loss": 0.89065331,
      "num_input_tokens_seen": 5410555,
      "step": 259,
      "time_per_iteration": 2.640153169631958
    },
    {
      "auxiliary_loss_clip": 0.01339282,
      "auxiliary_loss_mlp": 0.01057125,
      "balance_loss_clip": 1.09141672,
      "balance_loss_mlp": 1.04020941,
      "epoch": 0.031263151566163654,
      "flos": 20959442881920.0,
      "grad_norm": 2.7397007365531207,
      "language_loss": 0.82201421,
      "learning_rate": 3.999984830098494e-06,
      "loss": 0.84597832,
      "num_input_tokens_seen": 5430135,
      "step": 260,
      "time_per_iteration": 3.5857839584350586
    },
    {
      "auxiliary_loss_clip": 0.01336088,
      "auxiliary_loss_mlp": 0.01063742,
      "balance_loss_clip": 1.08783889,
      "balance_loss_mlp": 1.04633701,
      "epoch": 0.03138339445680274,
      "flos": 14793006412800.0,
      "grad_norm": 3.46280460672016,
      "language_loss": 0.97745872,
      "learning_rate": 3.999981644424051e-06,
      "loss": 1.00145698,
      "num_input_tokens_seen": 5444935,
      "step": 261,
      "time_per_iteration": 4.463522434234619
    },
    {
      "auxiliary_loss_clip": 0.01339378,
      "auxiliary_loss_mlp": 0.01069247,
      "balance_loss_clip": 1.09212101,
      "balance_loss_mlp": 1.05074584,
      "epoch": 0.03150363734744183,
      "flos": 11655599022720.0,
      "grad_norm": 2.7977881144240695,
      "language_loss": 0.86087382,
      "learning_rate": 3.999978155353982e-06,
      "loss": 0.88496011,
      "num_input_tokens_seen": 5462080,
      "step": 262,
      "time_per_iteration": 2.60974383354187
    },
    {
      "auxiliary_loss_clip": 0.01335537,
      "auxiliary_loss_mlp": 0.01062858,
      "balance_loss_clip": 1.08732772,
      "balance_loss_mlp": 1.04479766,
      "epoch": 0.03162388023808092,
      "flos": 33727732485120.0,
      "grad_norm": 2.319468458369819,
      "language_loss": 0.80562586,
      "learning_rate": 3.9999743628888186e-06,
      "loss": 0.82960981,
      "num_input_tokens_seen": 5483870,
      "step": 263,
      "time_per_iteration": 3.6830005645751953
    },
    {
      "auxiliary_loss_clip": 0.01329393,
      "auxiliary_loss_mlp": 0.01057665,
      "balance_loss_clip": 1.08401775,
      "balance_loss_mlp": 1.04015279,
      "epoch": 0.03174412312872001,
      "flos": 20810952057600.0,
      "grad_norm": 6.059573783327329,
      "language_loss": 0.89456326,
      "learning_rate": 3.999970267029133e-06,
      "loss": 0.91843385,
      "num_input_tokens_seen": 5502830,
      "step": 264,
      "time_per_iteration": 2.6728200912475586
    },
    {
      "auxiliary_loss_clip": 0.01332787,
      "auxiliary_loss_mlp": 0.01054245,
      "balance_loss_clip": 1.08736646,
      "balance_loss_mlp": 1.03711498,
      "epoch": 0.0318643660193591,
      "flos": 23727939638400.0,
      "grad_norm": 2.1265296151641637,
      "language_loss": 0.80218452,
      "learning_rate": 3.999965867775548e-06,
      "loss": 0.82605487,
      "num_input_tokens_seen": 5523225,
      "step": 265,
      "time_per_iteration": 2.6331467628479004
    },
    {
      "auxiliary_loss_clip": 0.01334392,
      "auxiliary_loss_mlp": 0.01061785,
      "balance_loss_clip": 1.08691001,
      "balance_loss_mlp": 1.04472589,
      "epoch": 0.0319846089099982,
      "flos": 13917863450880.0,
      "grad_norm": 2.6741509462366593,
      "language_loss": 0.86850512,
      "learning_rate": 3.9999611651287315e-06,
      "loss": 0.8924669,
      "num_input_tokens_seen": 5541380,
      "step": 266,
      "time_per_iteration": 2.668396472930908
    },
    {
      "auxiliary_loss_clip": 0.01337152,
      "auxiliary_loss_mlp": 0.01057237,
      "balance_loss_clip": 1.08925986,
      "balance_loss_mlp": 1.04028535,
      "epoch": 0.03210485180063729,
      "flos": 14753253035520.0,
      "grad_norm": 2.364227147119494,
      "language_loss": 0.78836089,
      "learning_rate": 3.999956159089396e-06,
      "loss": 0.81230479,
      "num_input_tokens_seen": 5558830,
      "step": 267,
      "time_per_iteration": 2.649045705795288
    },
    {
      "auxiliary_loss_clip": 0.01335069,
      "auxiliary_loss_mlp": 0.01061707,
      "balance_loss_clip": 1.08837819,
      "balance_loss_mlp": 1.04425466,
      "epoch": 0.03222509469127638,
      "flos": 28913153304960.0,
      "grad_norm": 2.106099301231671,
      "language_loss": 0.7968725,
      "learning_rate": 3.999950849658302e-06,
      "loss": 0.82084024,
      "num_input_tokens_seen": 5577750,
      "step": 268,
      "time_per_iteration": 2.7429590225219727
    },
    {
      "auxiliary_loss_clip": 0.01341353,
      "auxiliary_loss_mlp": 0.01063408,
      "balance_loss_clip": 1.09036088,
      "balance_loss_mlp": 1.04640889,
      "epoch": 0.03234533758191547,
      "flos": 16946389739520.0,
      "grad_norm": 2.099338927460233,
      "language_loss": 0.84224808,
      "learning_rate": 3.999945236836254e-06,
      "loss": 0.86629564,
      "num_input_tokens_seen": 5596715,
      "step": 269,
      "time_per_iteration": 2.606489419937134
    },
    {
      "auxiliary_loss_clip": 0.01340946,
      "auxiliary_loss_mlp": 0.01064131,
      "balance_loss_clip": 1.09159112,
      "balance_loss_mlp": 1.04543865,
      "epoch": 0.03246558047255456,
      "flos": 18989096284800.0,
      "grad_norm": 2.573486288193937,
      "language_loss": 0.94445693,
      "learning_rate": 3.999939320624103e-06,
      "loss": 0.96850777,
      "num_input_tokens_seen": 5611865,
      "step": 270,
      "time_per_iteration": 2.668107509613037
    },
    {
      "auxiliary_loss_clip": 0.01337228,
      "auxiliary_loss_mlp": 0.0105871,
      "balance_loss_clip": 1.08984327,
      "balance_loss_mlp": 1.04138863,
      "epoch": 0.03258582336319365,
      "flos": 23728334688000.0,
      "grad_norm": 2.0280765614907885,
      "language_loss": 0.89657652,
      "learning_rate": 3.999933101022749e-06,
      "loss": 0.92053592,
      "num_input_tokens_seen": 5632270,
      "step": 271,
      "time_per_iteration": 2.64052152633667
    },
    {
      "auxiliary_loss_clip": 0.01335515,
      "auxiliary_loss_mlp": 0.01061832,
      "balance_loss_clip": 1.08883166,
      "balance_loss_mlp": 1.0445224,
      "epoch": 0.032706066253832745,
      "flos": 27670823562240.0,
      "grad_norm": 1.8164468645438316,
      "language_loss": 0.86831838,
      "learning_rate": 3.999926578033132e-06,
      "loss": 0.89229184,
      "num_input_tokens_seen": 5652085,
      "step": 272,
      "time_per_iteration": 2.727051258087158
    },
    {
      "auxiliary_loss_clip": 0.01333988,
      "auxiliary_loss_mlp": 0.01062995,
      "balance_loss_clip": 1.08534276,
      "balance_loss_mlp": 1.04532814,
      "epoch": 0.032826309144471835,
      "flos": 45624685968000.0,
      "grad_norm": 2.2526943988225034,
      "language_loss": 0.62760359,
      "learning_rate": 3.999919751656244e-06,
      "loss": 0.65157342,
      "num_input_tokens_seen": 5678985,
      "step": 273,
      "time_per_iteration": 2.8644216060638428
    },
    {
      "auxiliary_loss_clip": 0.01329611,
      "auxiliary_loss_mlp": 0.01055845,
      "balance_loss_clip": 1.08413887,
      "balance_loss_mlp": 1.03743887,
      "epoch": 0.032946552035110925,
      "flos": 25812374808960.0,
      "grad_norm": 2.274063846740029,
      "language_loss": 0.75589293,
      "learning_rate": 3.9999126218931195e-06,
      "loss": 0.77974749,
      "num_input_tokens_seen": 5697020,
      "step": 274,
      "time_per_iteration": 2.68550705909729
    },
    {
      "auxiliary_loss_clip": 0.01336485,
      "auxiliary_loss_mlp": 0.01050746,
      "balance_loss_clip": 1.08969378,
      "balance_loss_mlp": 1.03330588,
      "epoch": 0.033066794925750015,
      "flos": 15121984101120.0,
      "grad_norm": 2.2083926036150445,
      "language_loss": 0.89378953,
      "learning_rate": 3.99990518874484e-06,
      "loss": 0.91766179,
      "num_input_tokens_seen": 5713460,
      "step": 275,
      "time_per_iteration": 2.6182687282562256
    },
    {
      "auxiliary_loss_clip": 0.01334676,
      "auxiliary_loss_mlp": 0.01066869,
      "balance_loss_clip": 1.08866179,
      "balance_loss_mlp": 1.04972661,
      "epoch": 0.033187037816389105,
      "flos": 22776593973120.0,
      "grad_norm": 2.497853624096942,
      "language_loss": 0.92470264,
      "learning_rate": 3.999897452212534e-06,
      "loss": 0.94871807,
      "num_input_tokens_seen": 5730790,
      "step": 276,
      "time_per_iteration": 2.649456262588501
    },
    {
      "auxiliary_loss_clip": 0.01331034,
      "auxiliary_loss_mlp": 0.01058629,
      "balance_loss_clip": 1.08657682,
      "balance_loss_mlp": 1.04089057,
      "epoch": 0.033307280707028195,
      "flos": 23331414424320.0,
      "grad_norm": 1.984748587482508,
      "language_loss": 1.00358295,
      "learning_rate": 3.999889412297374e-06,
      "loss": 1.02747953,
      "num_input_tokens_seen": 5750215,
      "step": 277,
      "time_per_iteration": 2.695356607437134
    },
    {
      "auxiliary_loss_clip": 0.01330703,
      "auxiliary_loss_mlp": 0.01044504,
      "balance_loss_clip": 1.0850184,
      "balance_loss_mlp": 1.02816021,
      "epoch": 0.03342752359766729,
      "flos": 28840290566400.0,
      "grad_norm": 3.632928047190656,
      "language_loss": 0.78950381,
      "learning_rate": 3.999881069000581e-06,
      "loss": 0.81325591,
      "num_input_tokens_seen": 5769945,
      "step": 278,
      "time_per_iteration": 2.677590847015381
    },
    {
      "auxiliary_loss_clip": 0.01332296,
      "auxiliary_loss_mlp": 0.01054599,
      "balance_loss_clip": 1.08512211,
      "balance_loss_mlp": 1.03601456,
      "epoch": 0.03354776648830638,
      "flos": 19384544090880.0,
      "grad_norm": 2.7129550262127307,
      "language_loss": 0.8704685,
      "learning_rate": 3.99987242232342e-06,
      "loss": 0.89433748,
      "num_input_tokens_seen": 5784950,
      "step": 279,
      "time_per_iteration": 2.6613900661468506
    },
    {
      "auxiliary_loss_clip": 0.01335487,
      "auxiliary_loss_mlp": 0.01063675,
      "balance_loss_clip": 1.08964241,
      "balance_loss_mlp": 1.04554307,
      "epoch": 0.03366800937894547,
      "flos": 17858628472320.0,
      "grad_norm": 1.7493086138852192,
      "language_loss": 0.79657757,
      "learning_rate": 3.9998634722672026e-06,
      "loss": 0.82056922,
      "num_input_tokens_seen": 5805005,
      "step": 280,
      "time_per_iteration": 2.6390607357025146
    },
    {
      "auxiliary_loss_clip": 0.01334185,
      "auxiliary_loss_mlp": 0.01058962,
      "balance_loss_clip": 1.08875322,
      "balance_loss_mlp": 1.04202271,
      "epoch": 0.03378825226958456,
      "flos": 35951033635200.0,
      "grad_norm": 2.3554459232650524,
      "language_loss": 0.78393871,
      "learning_rate": 3.999854218833286e-06,
      "loss": 0.80787021,
      "num_input_tokens_seen": 5825825,
      "step": 281,
      "time_per_iteration": 2.8130767345428467
    },
    {
      "auxiliary_loss_clip": 0.01333005,
      "auxiliary_loss_mlp": 0.010606,
      "balance_loss_clip": 1.08806884,
      "balance_loss_mlp": 1.04301643,
      "epoch": 0.03390849516022365,
      "flos": 25702488126720.0,
      "grad_norm": 2.0212411418725678,
      "language_loss": 0.82066971,
      "learning_rate": 3.999844662023075e-06,
      "loss": 0.8446058,
      "num_input_tokens_seen": 5845700,
      "step": 282,
      "time_per_iteration": 2.681990623474121
    },
    {
      "auxiliary_loss_clip": 0.01325462,
      "auxiliary_loss_mlp": 0.01056583,
      "balance_loss_clip": 1.08451581,
      "balance_loss_mlp": 1.03938115,
      "epoch": 0.03402873805086274,
      "flos": 21284505987840.0,
      "grad_norm": 2.0125879102470874,
      "language_loss": 0.92350531,
      "learning_rate": 3.999834801838018e-06,
      "loss": 0.94732571,
      "num_input_tokens_seen": 5864680,
      "step": 283,
      "time_per_iteration": 2.6648290157318115
    },
    {
      "auxiliary_loss_clip": 0.01326316,
      "auxiliary_loss_mlp": 0.01054204,
      "balance_loss_clip": 1.08465767,
      "balance_loss_mlp": 1.03682351,
      "epoch": 0.03414898094150183,
      "flos": 22710913954560.0,
      "grad_norm": 2.0904795127995413,
      "language_loss": 0.7416954,
      "learning_rate": 3.9998246382796115e-06,
      "loss": 0.76550066,
      "num_input_tokens_seen": 5884260,
      "step": 284,
      "time_per_iteration": 2.65207576751709
    },
    {
      "auxiliary_loss_clip": 0.01332141,
      "auxiliary_loss_mlp": 0.01053733,
      "balance_loss_clip": 1.08470321,
      "balance_loss_mlp": 1.03542197,
      "epoch": 0.03426922383214093,
      "flos": 18879927874560.0,
      "grad_norm": 2.1196859561926717,
      "language_loss": 0.90886545,
      "learning_rate": 3.999814171349399e-06,
      "loss": 0.93272418,
      "num_input_tokens_seen": 5902120,
      "step": 285,
      "time_per_iteration": 2.651594400405884
    },
    {
      "auxiliary_loss_clip": 0.01328249,
      "auxiliary_loss_mlp": 0.01057183,
      "balance_loss_clip": 1.08592618,
      "balance_loss_mlp": 1.04056549,
      "epoch": 0.03438946672278002,
      "flos": 34752012716160.0,
      "grad_norm": 1.9601920601689813,
      "language_loss": 0.7350443,
      "learning_rate": 3.9998034010489655e-06,
      "loss": 0.75889862,
      "num_input_tokens_seen": 5925810,
      "step": 286,
      "time_per_iteration": 2.712480068206787
    },
    {
      "auxiliary_loss_clip": 0.01327743,
      "auxiliary_loss_mlp": 0.01060401,
      "balance_loss_clip": 1.08694434,
      "balance_loss_mlp": 1.04358006,
      "epoch": 0.03450970961341911,
      "flos": 22164102236160.0,
      "grad_norm": 2.213816510446719,
      "language_loss": 0.76066369,
      "learning_rate": 3.999792327379946e-06,
      "loss": 0.78454512,
      "num_input_tokens_seen": 5945185,
      "step": 287,
      "time_per_iteration": 3.6035640239715576
    },
    {
      "auxiliary_loss_clip": 0.01335611,
      "auxiliary_loss_mlp": 0.01063698,
      "balance_loss_clip": 1.09270263,
      "balance_loss_mlp": 1.04671049,
      "epoch": 0.034629952504058197,
      "flos": 21725740656000.0,
      "grad_norm": 2.281414438516137,
      "language_loss": 0.96439213,
      "learning_rate": 3.999780950344021e-06,
      "loss": 0.9883852,
      "num_input_tokens_seen": 5963375,
      "step": 288,
      "time_per_iteration": 3.5618526935577393
    },
    {
      "auxiliary_loss_clip": 0.01335311,
      "auxiliary_loss_mlp": 0.01064361,
      "balance_loss_clip": 1.08854651,
      "balance_loss_mlp": 1.04626465,
      "epoch": 0.034750195394697286,
      "flos": 20047994248320.0,
      "grad_norm": 2.002936315615876,
      "language_loss": 0.8273859,
      "learning_rate": 3.999769269942916e-06,
      "loss": 0.85138261,
      "num_input_tokens_seen": 5983415,
      "step": 289,
      "time_per_iteration": 2.664250373840332
    },
    {
      "auxiliary_loss_clip": 0.01329564,
      "auxiliary_loss_mlp": 0.01053702,
      "balance_loss_clip": 1.08649755,
      "balance_loss_mlp": 1.03610635,
      "epoch": 0.034870438285336376,
      "flos": 27965865876480.0,
      "grad_norm": 1.9237530727098424,
      "language_loss": 0.81287634,
      "learning_rate": 3.999757286178402e-06,
      "loss": 0.83670896,
      "num_input_tokens_seen": 6005850,
      "step": 290,
      "time_per_iteration": 3.6787497997283936
    },
    {
      "auxiliary_loss_clip": 0.01333072,
      "auxiliary_loss_mlp": 0.01054549,
      "balance_loss_clip": 1.08919001,
      "balance_loss_mlp": 1.03703666,
      "epoch": 0.03499068117597547,
      "flos": 22017514832640.0,
      "grad_norm": 1.9328539518901988,
      "language_loss": 0.90564573,
      "learning_rate": 3.999744999052299e-06,
      "loss": 0.92952192,
      "num_input_tokens_seen": 6027240,
      "step": 291,
      "time_per_iteration": 2.6998767852783203
    },
    {
      "auxiliary_loss_clip": 0.01287519,
      "auxiliary_loss_mlp": 0.01012583,
      "balance_loss_clip": 1.10304284,
      "balance_loss_mlp": 1.00166368,
      "epoch": 0.03511092406661456,
      "flos": 57242147725440.0,
      "grad_norm": 0.9668164635051221,
      "language_loss": 0.6122725,
      "learning_rate": 3.9997324085664675e-06,
      "loss": 0.63527358,
      "num_input_tokens_seen": 6087470,
      "step": 292,
      "time_per_iteration": 3.1550822257995605
    },
    {
      "auxiliary_loss_clip": 0.01327218,
      "auxiliary_loss_mlp": 0.01060244,
      "balance_loss_clip": 1.08447433,
      "balance_loss_mlp": 1.04227936,
      "epoch": 0.03523116695725365,
      "flos": 22928065626240.0,
      "grad_norm": 2.3924289087509867,
      "language_loss": 0.92160743,
      "learning_rate": 3.999719514722821e-06,
      "loss": 0.94548208,
      "num_input_tokens_seen": 6107600,
      "step": 293,
      "time_per_iteration": 2.68707537651062
    },
    {
      "auxiliary_loss_clip": 0.01325176,
      "auxiliary_loss_mlp": 0.01053765,
      "balance_loss_clip": 1.08485496,
      "balance_loss_mlp": 1.03712332,
      "epoch": 0.03535140984789274,
      "flos": 36903241226880.0,
      "grad_norm": 2.576314395757117,
      "language_loss": 0.74739826,
      "learning_rate": 3.999706317523314e-06,
      "loss": 0.77118766,
      "num_input_tokens_seen": 6126160,
      "step": 294,
      "time_per_iteration": 2.7441561222076416
    },
    {
      "auxiliary_loss_clip": 0.01326268,
      "auxiliary_loss_mlp": 0.01051182,
      "balance_loss_clip": 1.08577216,
      "balance_loss_mlp": 1.03465962,
      "epoch": 0.03547165273853183,
      "flos": 20449152316800.0,
      "grad_norm": 2.2062969423675463,
      "language_loss": 0.85968941,
      "learning_rate": 3.999692816969948e-06,
      "loss": 0.88346392,
      "num_input_tokens_seen": 6145695,
      "step": 295,
      "time_per_iteration": 2.6573500633239746
    },
    {
      "auxiliary_loss_clip": 0.01272452,
      "auxiliary_loss_mlp": 0.01013764,
      "balance_loss_clip": 1.09009862,
      "balance_loss_mlp": 1.00284421,
      "epoch": 0.03559189562917092,
      "flos": 69850564871040.0,
      "grad_norm": 1.0516663703987414,
      "language_loss": 0.69484371,
      "learning_rate": 3.999679013064772e-06,
      "loss": 0.71770585,
      "num_input_tokens_seen": 6212440,
      "step": 296,
      "time_per_iteration": 3.241339921951294
    },
    {
      "auxiliary_loss_clip": 0.01328491,
      "auxiliary_loss_mlp": 0.0105486,
      "balance_loss_clip": 1.08647859,
      "balance_loss_mlp": 1.03792024,
      "epoch": 0.03571213851981002,
      "flos": 21651944163840.0,
      "grad_norm": 2.4967300225527107,
      "language_loss": 0.85484248,
      "learning_rate": 3.99966490580988e-06,
      "loss": 0.878676,
      "num_input_tokens_seen": 6229800,
      "step": 297,
      "time_per_iteration": 2.6126229763031006
    },
    {
      "auxiliary_loss_clip": 0.01330406,
      "auxiliary_loss_mlp": 0.01058649,
      "balance_loss_clip": 1.08608389,
      "balance_loss_mlp": 1.04160237,
      "epoch": 0.03583238141044911,
      "flos": 43945610757120.0,
      "grad_norm": 3.6477141024878614,
      "language_loss": 0.65733492,
      "learning_rate": 3.999650495207411e-06,
      "loss": 0.68122548,
      "num_input_tokens_seen": 6255825,
      "step": 298,
      "time_per_iteration": 2.8482000827789307
    },
    {
      "auxiliary_loss_clip": 0.01322859,
      "auxiliary_loss_mlp": 0.01061203,
      "balance_loss_clip": 1.08413863,
      "balance_loss_mlp": 1.04383397,
      "epoch": 0.0359526243010882,
      "flos": 18910810592640.0,
      "grad_norm": 2.3676631957160823,
      "language_loss": 0.90276647,
      "learning_rate": 3.999635781259553e-06,
      "loss": 0.92660713,
      "num_input_tokens_seen": 6271090,
      "step": 299,
      "time_per_iteration": 2.5826237201690674
    },
    {
      "auxiliary_loss_clip": 0.01255825,
      "auxiliary_loss_mlp": 0.01012024,
      "balance_loss_clip": 1.07762194,
      "balance_loss_mlp": 1.00110424,
      "epoch": 0.03607286719172729,
      "flos": 61668892782720.0,
      "grad_norm": 0.9263967720796876,
      "language_loss": 0.52344573,
      "learning_rate": 3.999620763968535e-06,
      "loss": 0.54612416,
      "num_input_tokens_seen": 6329965,
      "step": 300,
      "time_per_iteration": 3.0407047271728516
    },
    {
      "auxiliary_loss_clip": 0.01322942,
      "auxiliary_loss_mlp": 0.0105322,
      "balance_loss_clip": 1.08491647,
      "balance_loss_mlp": 1.03580356,
      "epoch": 0.03619311008236638,
      "flos": 27819062991360.0,
      "grad_norm": 1.647491474523333,
      "language_loss": 0.86296582,
      "learning_rate": 3.999605443336638e-06,
      "loss": 0.88672739,
      "num_input_tokens_seen": 6352095,
      "step": 301,
      "time_per_iteration": 2.718418836593628
    },
    {
      "auxiliary_loss_clip": 0.01329299,
      "auxiliary_loss_mlp": 0.01059603,
      "balance_loss_clip": 1.08762431,
      "balance_loss_mlp": 1.04234171,
      "epoch": 0.03631335297300547,
      "flos": 13621133197440.0,
      "grad_norm": 2.8987628751801457,
      "language_loss": 0.89245874,
      "learning_rate": 3.999589819366185e-06,
      "loss": 0.91634774,
      "num_input_tokens_seen": 6365885,
      "step": 302,
      "time_per_iteration": 2.622633934020996
    },
    {
      "auxiliary_loss_clip": 0.01328606,
      "auxiliary_loss_mlp": 0.01056024,
      "balance_loss_clip": 1.08602774,
      "balance_loss_mlp": 1.038059,
      "epoch": 0.036433595863644565,
      "flos": 27631788456960.0,
      "grad_norm": 2.4235722064731617,
      "language_loss": 0.85026562,
      "learning_rate": 3.999573892059547e-06,
      "loss": 0.87411189,
      "num_input_tokens_seen": 6385015,
      "step": 303,
      "time_per_iteration": 2.668869733810425
    },
    {
      "auxiliary_loss_clip": 0.01331164,
      "auxiliary_loss_mlp": 0.01060863,
      "balance_loss_clip": 1.08678961,
      "balance_loss_mlp": 1.04218352,
      "epoch": 0.036553838754283655,
      "flos": 24572020314240.0,
      "grad_norm": 1.9493067762633212,
      "language_loss": 0.81083059,
      "learning_rate": 3.999557661419138e-06,
      "loss": 0.83475089,
      "num_input_tokens_seen": 6405165,
      "step": 304,
      "time_per_iteration": 2.6923069953918457
    },
    {
      "auxiliary_loss_clip": 0.01330491,
      "auxiliary_loss_mlp": 0.01053737,
      "balance_loss_clip": 1.08860016,
      "balance_loss_mlp": 1.03750014,
      "epoch": 0.036674081644922744,
      "flos": 23404313076480.0,
      "grad_norm": 1.892033766540908,
      "language_loss": 0.8147701,
      "learning_rate": 3.9995411274474225e-06,
      "loss": 0.83861238,
      "num_input_tokens_seen": 6424445,
      "step": 305,
      "time_per_iteration": 2.6390857696533203
    },
    {
      "auxiliary_loss_clip": 0.01327452,
      "auxiliary_loss_mlp": 0.01064888,
      "balance_loss_clip": 1.08515453,
      "balance_loss_mlp": 1.04641032,
      "epoch": 0.036794324535561834,
      "flos": 27489690253440.0,
      "grad_norm": 2.553315817329672,
      "language_loss": 0.8142941,
      "learning_rate": 3.999524290146908e-06,
      "loss": 0.8382175,
      "num_input_tokens_seen": 6444650,
      "step": 306,
      "time_per_iteration": 2.6644692420959473
    },
    {
      "auxiliary_loss_clip": 0.01326539,
      "auxiliary_loss_mlp": 0.010634,
      "balance_loss_clip": 1.08802783,
      "balance_loss_mlp": 1.04603148,
      "epoch": 0.036914567426200924,
      "flos": 19463476227840.0,
      "grad_norm": 2.7890545608089155,
      "language_loss": 0.92451787,
      "learning_rate": 3.9995071495201485e-06,
      "loss": 0.94841725,
      "num_input_tokens_seen": 6461755,
      "step": 307,
      "time_per_iteration": 2.5955259799957275
    },
    {
      "auxiliary_loss_clip": 0.01324739,
      "auxiliary_loss_mlp": 0.01057146,
      "balance_loss_clip": 1.08603644,
      "balance_loss_mlp": 1.03924108,
      "epoch": 0.037034810316840014,
      "flos": 22309324922880.0,
      "grad_norm": 2.634051142064039,
      "language_loss": 0.97629082,
      "learning_rate": 3.999489705569744e-06,
      "loss": 1.00010967,
      "num_input_tokens_seen": 6479455,
      "step": 308,
      "time_per_iteration": 2.6629934310913086
    },
    {
      "auxiliary_loss_clip": 0.01321513,
      "auxiliary_loss_mlp": 0.01057241,
      "balance_loss_clip": 1.08195066,
      "balance_loss_mlp": 1.04033685,
      "epoch": 0.03715505320747911,
      "flos": 18588333265920.0,
      "grad_norm": 2.700452304714524,
      "language_loss": 0.8641988,
      "learning_rate": 3.999471958298341e-06,
      "loss": 0.88798642,
      "num_input_tokens_seen": 6498365,
      "step": 309,
      "time_per_iteration": 2.6396284103393555
    },
    {
      "auxiliary_loss_clip": 0.01329275,
      "auxiliary_loss_mlp": 0.01066449,
      "balance_loss_clip": 1.08779216,
      "balance_loss_mlp": 1.04807854,
      "epoch": 0.0372752960981182,
      "flos": 35955343267200.0,
      "grad_norm": 1.9111949790978375,
      "language_loss": 0.7583102,
      "learning_rate": 3.999453907708631e-06,
      "loss": 0.78226745,
      "num_input_tokens_seen": 6520770,
      "step": 310,
      "time_per_iteration": 2.729296922683716
    },
    {
      "auxiliary_loss_clip": 0.01326364,
      "auxiliary_loss_mlp": 0.01050163,
      "balance_loss_clip": 1.08623314,
      "balance_loss_mlp": 1.03362846,
      "epoch": 0.03739553898875729,
      "flos": 20814040627200.0,
      "grad_norm": 2.517325640595576,
      "language_loss": 0.81405878,
      "learning_rate": 3.999435553803353e-06,
      "loss": 0.83782411,
      "num_input_tokens_seen": 6540170,
      "step": 311,
      "time_per_iteration": 2.681500196456909
    },
    {
      "auxiliary_loss_clip": 0.0132187,
      "auxiliary_loss_mlp": 0.01061268,
      "balance_loss_clip": 1.08403993,
      "balance_loss_mlp": 1.04437602,
      "epoch": 0.03751578187939638,
      "flos": 20264140339200.0,
      "grad_norm": 2.4062009690623185,
      "language_loss": 0.83340812,
      "learning_rate": 3.999416896585292e-06,
      "loss": 0.85723954,
      "num_input_tokens_seen": 6557200,
      "step": 312,
      "time_per_iteration": 2.630185842514038
    },
    {
      "auxiliary_loss_clip": 0.01324795,
      "auxiliary_loss_mlp": 0.0105592,
      "balance_loss_clip": 1.08441758,
      "balance_loss_mlp": 1.0386821,
      "epoch": 0.03763602477003547,
      "flos": 20668063754880.0,
      "grad_norm": 4.102682524580284,
      "language_loss": 0.85893083,
      "learning_rate": 3.9993979360572775e-06,
      "loss": 0.88273805,
      "num_input_tokens_seen": 6577340,
      "step": 313,
      "time_per_iteration": 2.7654476165771484
    },
    {
      "auxiliary_loss_clip": 0.01332345,
      "auxiliary_loss_mlp": 0.01060855,
      "balance_loss_clip": 1.08930731,
      "balance_loss_mlp": 1.043486,
      "epoch": 0.03775626766067456,
      "flos": 16691352197760.0,
      "grad_norm": 2.5103934033276025,
      "language_loss": 0.82642031,
      "learning_rate": 3.999378672222185e-06,
      "loss": 0.85035223,
      "num_input_tokens_seen": 6595125,
      "step": 314,
      "time_per_iteration": 3.5210869312286377
    },
    {
      "auxiliary_loss_clip": 0.01326016,
      "auxiliary_loss_mlp": 0.01056496,
      "balance_loss_clip": 1.08723402,
      "balance_loss_mlp": 1.03780389,
      "epoch": 0.03787651055131366,
      "flos": 21141797253120.0,
      "grad_norm": 2.0430984015790212,
      "language_loss": 0.82706654,
      "learning_rate": 3.9993591050829385e-06,
      "loss": 0.85089171,
      "num_input_tokens_seen": 6612990,
      "step": 315,
      "time_per_iteration": 3.571441888809204
    },
    {
      "auxiliary_loss_clip": 0.01326013,
      "auxiliary_loss_mlp": 0.0106401,
      "balance_loss_clip": 1.08732975,
      "balance_loss_mlp": 1.04611647,
      "epoch": 0.037996753441952746,
      "flos": 22018089450240.0,
      "grad_norm": 2.52982692807849,
      "language_loss": 0.7931121,
      "learning_rate": 3.999339234642506e-06,
      "loss": 0.81701231,
      "num_input_tokens_seen": 6632740,
      "step": 316,
      "time_per_iteration": 2.7277400493621826
    },
    {
      "auxiliary_loss_clip": 0.01327584,
      "auxiliary_loss_mlp": 0.01050529,
      "balance_loss_clip": 1.08880401,
      "balance_loss_mlp": 1.03198051,
      "epoch": 0.038116996332591836,
      "flos": 27709391790720.0,
      "grad_norm": 2.3544040212987114,
      "language_loss": 0.8376497,
      "learning_rate": 3.9993190609038994e-06,
      "loss": 0.86143076,
      "num_input_tokens_seen": 6651505,
      "step": 317,
      "time_per_iteration": 2.633148193359375
    },
    {
      "auxiliary_loss_clip": 0.01318918,
      "auxiliary_loss_mlp": 0.01051527,
      "balance_loss_clip": 1.08339977,
      "balance_loss_mlp": 1.0340035,
      "epoch": 0.038237239223230926,
      "flos": 21178067011200.0,
      "grad_norm": 2.3146947315547943,
      "language_loss": 0.83068633,
      "learning_rate": 3.999298583870182e-06,
      "loss": 0.85439074,
      "num_input_tokens_seen": 6671090,
      "step": 318,
      "time_per_iteration": 3.5569136142730713
    },
    {
      "auxiliary_loss_clip": 0.01321483,
      "auxiliary_loss_mlp": 0.01056807,
      "balance_loss_clip": 1.08381891,
      "balance_loss_mlp": 1.03952169,
      "epoch": 0.038357482113870016,
      "flos": 25556618995200.0,
      "grad_norm": 1.8839049189440786,
      "language_loss": 0.77565634,
      "learning_rate": 3.999277803544458e-06,
      "loss": 0.79943919,
      "num_input_tokens_seen": 6691245,
      "step": 319,
      "time_per_iteration": 2.622443199157715
    },
    {
      "auxiliary_loss_clip": 0.01234888,
      "auxiliary_loss_mlp": 0.01029189,
      "balance_loss_clip": 1.06328619,
      "balance_loss_mlp": 1.01660061,
      "epoch": 0.038477725004509106,
      "flos": 59227578034560.0,
      "grad_norm": 0.9792727208711397,
      "language_loss": 0.62453133,
      "learning_rate": 3.999256719929882e-06,
      "loss": 0.64717209,
      "num_input_tokens_seen": 6752520,
      "step": 320,
      "time_per_iteration": 3.1581408977508545
    },
    {
      "auxiliary_loss_clip": 0.01231991,
      "auxiliary_loss_mlp": 0.01018793,
      "balance_loss_clip": 1.06097293,
      "balance_loss_mlp": 1.00649095,
      "epoch": 0.0385979678951482,
      "flos": 67317676398720.0,
      "grad_norm": 1.2308967342690122,
      "language_loss": 0.6718297,
      "learning_rate": 3.999235333029651e-06,
      "loss": 0.69433761,
      "num_input_tokens_seen": 6806460,
      "step": 321,
      "time_per_iteration": 3.029639482498169
    },
    {
      "auxiliary_loss_clip": 0.01319911,
      "auxiliary_loss_mlp": 0.01059741,
      "balance_loss_clip": 1.08551931,
      "balance_loss_mlp": 1.04303932,
      "epoch": 0.03871821078578729,
      "flos": 22746752749440.0,
      "grad_norm": 1.9161065870511802,
      "language_loss": 0.81925619,
      "learning_rate": 3.999213642847009e-06,
      "loss": 0.84305269,
      "num_input_tokens_seen": 6827045,
      "step": 322,
      "time_per_iteration": 2.6293816566467285
    },
    {
      "auxiliary_loss_clip": 0.01320991,
      "auxiliary_loss_mlp": 0.01055366,
      "balance_loss_clip": 1.08403552,
      "balance_loss_mlp": 1.03862941,
      "epoch": 0.03883845367642638,
      "flos": 26280613526400.0,
      "grad_norm": 2.68377907983788,
      "language_loss": 0.91136551,
      "learning_rate": 3.999191649385247e-06,
      "loss": 0.93512917,
      "num_input_tokens_seen": 6848220,
      "step": 323,
      "time_per_iteration": 2.862316608428955
    },
    {
      "auxiliary_loss_clip": 0.01226148,
      "auxiliary_loss_mlp": 0.01012849,
      "balance_loss_clip": 1.05784595,
      "balance_loss_mlp": 1.00026023,
      "epoch": 0.03895869656706547,
      "flos": 56962835568000.0,
      "grad_norm": 0.9080510767273545,
      "language_loss": 0.59809917,
      "learning_rate": 3.999169352647702e-06,
      "loss": 0.62048918,
      "num_input_tokens_seen": 6909400,
      "step": 324,
      "time_per_iteration": 3.0914359092712402
    },
    {
      "auxiliary_loss_clip": 0.0132296,
      "auxiliary_loss_mlp": 0.010806,
      "balance_loss_clip": 1.0852344,
      "balance_loss_mlp": 1.06189609,
      "epoch": 0.03907893945770456,
      "flos": 24863363527680.0,
      "grad_norm": 1.8557292659122324,
      "language_loss": 0.83021826,
      "learning_rate": 3.999146752637755e-06,
      "loss": 0.85425383,
      "num_input_tokens_seen": 6930445,
      "step": 325,
      "time_per_iteration": 2.6211185455322266
    },
    {
      "auxiliary_loss_clip": 0.01320682,
      "auxiliary_loss_mlp": 0.01057234,
      "balance_loss_clip": 1.08396697,
      "balance_loss_mlp": 1.0394125,
      "epoch": 0.03919918234834365,
      "flos": 18368595815040.0,
      "grad_norm": 2.886550910353189,
      "language_loss": 0.89466548,
      "learning_rate": 3.999123849358836e-06,
      "loss": 0.91844463,
      "num_input_tokens_seen": 6948110,
      "step": 326,
      "time_per_iteration": 2.6137118339538574
    },
    {
      "auxiliary_loss_clip": 0.01321354,
      "auxiliary_loss_mlp": 0.01060032,
      "balance_loss_clip": 1.08513725,
      "balance_loss_mlp": 1.04174519,
      "epoch": 0.03931942523898275,
      "flos": 25225414663680.0,
      "grad_norm": 2.5061993450474844,
      "language_loss": 0.7477628,
      "learning_rate": 3.999100642814418e-06,
      "loss": 0.77157658,
      "num_input_tokens_seen": 6968550,
      "step": 327,
      "time_per_iteration": 2.6519956588745117
    },
    {
      "auxiliary_loss_clip": 0.0131985,
      "auxiliary_loss_mlp": 0.01057054,
      "balance_loss_clip": 1.08475888,
      "balance_loss_mlp": 1.03892231,
      "epoch": 0.03943966812962184,
      "flos": 23257905240960.0,
      "grad_norm": 2.4449706463781133,
      "language_loss": 0.88374174,
      "learning_rate": 3.999077133008022e-06,
      "loss": 0.90751076,
      "num_input_tokens_seen": 6987135,
      "step": 328,
      "time_per_iteration": 2.6128017902374268
    },
    {
      "auxiliary_loss_clip": 0.01322901,
      "auxiliary_loss_mlp": 0.0106466,
      "balance_loss_clip": 1.08597422,
      "balance_loss_mlp": 1.0447284,
      "epoch": 0.03955991102026093,
      "flos": 29168837291520.0,
      "grad_norm": 1.8292955655905583,
      "language_loss": 0.90507501,
      "learning_rate": 3.9990533199432145e-06,
      "loss": 0.92895061,
      "num_input_tokens_seen": 7008630,
      "step": 329,
      "time_per_iteration": 2.6695449352264404
    },
    {
      "auxiliary_loss_clip": 0.01320064,
      "auxiliary_loss_mlp": 0.01058095,
      "balance_loss_clip": 1.0838002,
      "balance_loss_mlp": 1.04021335,
      "epoch": 0.03968015391090002,
      "flos": 17602441695360.0,
      "grad_norm": 2.1612030153510946,
      "language_loss": 0.75880176,
      "learning_rate": 3.999029203623608e-06,
      "loss": 0.78258336,
      "num_input_tokens_seen": 7026350,
      "step": 330,
      "time_per_iteration": 2.6431169509887695
    },
    {
      "auxiliary_loss_clip": 0.01315386,
      "auxiliary_loss_mlp": 0.01051294,
      "balance_loss_clip": 1.08276975,
      "balance_loss_mlp": 1.03344882,
      "epoch": 0.03980039680153911,
      "flos": 21799285752960.0,
      "grad_norm": 5.43726836289835,
      "language_loss": 0.86761189,
      "learning_rate": 3.99900478405286e-06,
      "loss": 0.89127868,
      "num_input_tokens_seen": 7045660,
      "step": 331,
      "time_per_iteration": 2.593463897705078
    },
    {
      "auxiliary_loss_clip": 0.01319256,
      "auxiliary_loss_mlp": 0.01057102,
      "balance_loss_clip": 1.08764517,
      "balance_loss_mlp": 1.04140198,
      "epoch": 0.0399206396921782,
      "flos": 15195134148480.0,
      "grad_norm": 2.4063228264154235,
      "language_loss": 0.82210231,
      "learning_rate": 3.998980061234676e-06,
      "loss": 0.84586596,
      "num_input_tokens_seen": 7063575,
      "step": 332,
      "time_per_iteration": 2.6380016803741455
    },
    {
      "auxiliary_loss_clip": 0.0132261,
      "auxiliary_loss_mlp": 0.01053088,
      "balance_loss_clip": 1.08412707,
      "balance_loss_mlp": 1.03453898,
      "epoch": 0.040040882582817294,
      "flos": 14422910630400.0,
      "grad_norm": 2.716650241954061,
      "language_loss": 0.7572397,
      "learning_rate": 3.9989550351728055e-06,
      "loss": 0.78099668,
      "num_input_tokens_seen": 7080505,
      "step": 333,
      "time_per_iteration": 2.541846752166748
    },
    {
      "auxiliary_loss_clip": 0.01319136,
      "auxiliary_loss_mlp": 0.01054205,
      "balance_loss_clip": 1.08603859,
      "balance_loss_mlp": 1.03734922,
      "epoch": 0.040161125473456384,
      "flos": 19280906375040.0,
      "grad_norm": 2.489164731608753,
      "language_loss": 0.84679055,
      "learning_rate": 3.998929705871046e-06,
      "loss": 0.87052393,
      "num_input_tokens_seen": 7097860,
      "step": 334,
      "time_per_iteration": 2.6191959381103516
    },
    {
      "auxiliary_loss_clip": 0.01317412,
      "auxiliary_loss_mlp": 0.01058466,
      "balance_loss_clip": 1.08606029,
      "balance_loss_mlp": 1.0415144,
      "epoch": 0.040281368364095474,
      "flos": 17821101738240.0,
      "grad_norm": 3.0228056244505592,
      "language_loss": 0.88921702,
      "learning_rate": 3.99890407333324e-06,
      "loss": 0.91297579,
      "num_input_tokens_seen": 7116390,
      "step": 335,
      "time_per_iteration": 2.5929923057556152
    },
    {
      "auxiliary_loss_clip": 0.01313798,
      "auxiliary_loss_mlp": 0.01057079,
      "balance_loss_clip": 1.08007336,
      "balance_loss_mlp": 1.03950763,
      "epoch": 0.040401611254734564,
      "flos": 19573757959680.0,
      "grad_norm": 1.9818282885752578,
      "language_loss": 0.87127095,
      "learning_rate": 3.998878137563275e-06,
      "loss": 0.89497972,
      "num_input_tokens_seen": 7135940,
      "step": 336,
      "time_per_iteration": 2.6696622371673584
    },
    {
      "auxiliary_loss_clip": 0.01315572,
      "auxiliary_loss_mlp": 0.01052725,
      "balance_loss_clip": 1.08246708,
      "balance_loss_mlp": 1.03486705,
      "epoch": 0.040521854145373654,
      "flos": 22054466949120.0,
      "grad_norm": 2.245541878004471,
      "language_loss": 0.8522684,
      "learning_rate": 3.998851898565085e-06,
      "loss": 0.87595135,
      "num_input_tokens_seen": 7155745,
      "step": 337,
      "time_per_iteration": 2.633291006088257
    },
    {
      "auxiliary_loss_clip": 0.01313064,
      "auxiliary_loss_mlp": 0.01049778,
      "balance_loss_clip": 1.0813508,
      "balance_loss_mlp": 1.03311241,
      "epoch": 0.04064209703601274,
      "flos": 22674644196480.0,
      "grad_norm": 2.075317379331567,
      "language_loss": 0.83078766,
      "learning_rate": 3.998825356342653e-06,
      "loss": 0.85441613,
      "num_input_tokens_seen": 7175920,
      "step": 338,
      "time_per_iteration": 2.6819536685943604
    },
    {
      "auxiliary_loss_clip": 0.01316852,
      "auxiliary_loss_mlp": 0.01064319,
      "balance_loss_clip": 1.08239007,
      "balance_loss_mlp": 1.0471406,
      "epoch": 0.04076233992665183,
      "flos": 38582172783360.0,
      "grad_norm": 2.3065432643313404,
      "language_loss": 0.72924727,
      "learning_rate": 3.998798510900003e-06,
      "loss": 0.75305891,
      "num_input_tokens_seen": 7198720,
      "step": 339,
      "time_per_iteration": 2.7588069438934326
    },
    {
      "auxiliary_loss_clip": 0.01315942,
      "auxiliary_loss_mlp": 0.01053191,
      "balance_loss_clip": 1.08204627,
      "balance_loss_mlp": 1.03611994,
      "epoch": 0.04088258281729093,
      "flos": 25885309374720.0,
      "grad_norm": 2.6638444611385586,
      "language_loss": 0.83839536,
      "learning_rate": 3.998771362241207e-06,
      "loss": 0.86208671,
      "num_input_tokens_seen": 7219125,
      "step": 340,
      "time_per_iteration": 2.7085659503936768
    },
    {
      "auxiliary_loss_clip": 0.01310246,
      "auxiliary_loss_mlp": 0.01056521,
      "balance_loss_clip": 1.08023632,
      "balance_loss_mlp": 1.03959346,
      "epoch": 0.04100282570793002,
      "flos": 19789832223360.0,
      "grad_norm": 1.9346532159345855,
      "language_loss": 0.87850654,
      "learning_rate": 3.998743910370385e-06,
      "loss": 0.90217423,
      "num_input_tokens_seen": 7237985,
      "step": 341,
      "time_per_iteration": 3.5850679874420166
    },
    {
      "auxiliary_loss_clip": 0.0132035,
      "auxiliary_loss_mlp": 0.01050323,
      "balance_loss_clip": 1.08989799,
      "balance_loss_mlp": 1.03184557,
      "epoch": 0.04112306859856911,
      "flos": 22565152563840.0,
      "grad_norm": 2.074147521657071,
      "language_loss": 0.73350441,
      "learning_rate": 3.998716155291702e-06,
      "loss": 0.75721109,
      "num_input_tokens_seen": 7255825,
      "step": 342,
      "time_per_iteration": 4.419248342514038
    },
    {
      "auxiliary_loss_clip": 0.01316426,
      "auxiliary_loss_mlp": 0.0105948,
      "balance_loss_clip": 1.08626771,
      "balance_loss_mlp": 1.0418849,
      "epoch": 0.0412433114892082,
      "flos": 25040654081280.0,
      "grad_norm": 1.7228295792701045,
      "language_loss": 0.90576184,
      "learning_rate": 3.998688097009366e-06,
      "loss": 0.92952091,
      "num_input_tokens_seen": 7276590,
      "step": 343,
      "time_per_iteration": 2.654273748397827
    },
    {
      "auxiliary_loss_clip": 0.01314423,
      "auxiliary_loss_mlp": 0.01054512,
      "balance_loss_clip": 1.08279991,
      "balance_loss_mlp": 1.03857386,
      "epoch": 0.04136355437984729,
      "flos": 25191371548800.0,
      "grad_norm": 2.67031612032337,
      "language_loss": 0.79789329,
      "learning_rate": 3.998659735527636e-06,
      "loss": 0.82158256,
      "num_input_tokens_seen": 7295680,
      "step": 344,
      "time_per_iteration": 2.6593985557556152
    },
    {
      "auxiliary_loss_clip": 0.01313472,
      "auxiliary_loss_mlp": 0.01056384,
      "balance_loss_clip": 1.08199382,
      "balance_loss_mlp": 1.03919435,
      "epoch": 0.04148379727048638,
      "flos": 22966777509120.0,
      "grad_norm": 2.9265696134865524,
      "language_loss": 0.7752133,
      "learning_rate": 3.998631070850813e-06,
      "loss": 0.79891193,
      "num_input_tokens_seen": 7316300,
      "step": 345,
      "time_per_iteration": 3.5941050052642822
    },
    {
      "auxiliary_loss_clip": 0.01312172,
      "auxiliary_loss_mlp": 0.01065153,
      "balance_loss_clip": 1.08450747,
      "balance_loss_mlp": 1.04954886,
      "epoch": 0.041604040161125476,
      "flos": 14063481187200.0,
      "grad_norm": 9.669920748042394,
      "language_loss": 0.83459818,
      "learning_rate": 3.9986021029832455e-06,
      "loss": 0.85837138,
      "num_input_tokens_seen": 7333615,
      "step": 346,
      "time_per_iteration": 2.6331076622009277
    },
    {
      "auxiliary_loss_clip": 0.01311588,
      "auxiliary_loss_mlp": 0.01054979,
      "balance_loss_clip": 1.0804745,
      "balance_loss_mlp": 1.03633428,
      "epoch": 0.041724283051764566,
      "flos": 12091877614080.0,
      "grad_norm": 3.1434985285290455,
      "language_loss": 0.91515702,
      "learning_rate": 3.9985728319293285e-06,
      "loss": 0.93882269,
      "num_input_tokens_seen": 7347590,
      "step": 347,
      "time_per_iteration": 2.543609142303467
    },
    {
      "auxiliary_loss_clip": 0.01315901,
      "auxiliary_loss_mlp": 0.01054174,
      "balance_loss_clip": 1.08138919,
      "balance_loss_mlp": 1.03667402,
      "epoch": 0.041844525942403656,
      "flos": 12385303816320.0,
      "grad_norm": 3.8108890094834353,
      "language_loss": 0.84985483,
      "learning_rate": 3.998543257693501e-06,
      "loss": 0.8735556,
      "num_input_tokens_seen": 7364345,
      "step": 348,
      "time_per_iteration": 2.6474947929382324
    },
    {
      "auxiliary_loss_clip": 0.01311354,
      "auxiliary_loss_mlp": 0.01064573,
      "balance_loss_clip": 1.0826478,
      "balance_loss_mlp": 1.04809856,
      "epoch": 0.041964768833042745,
      "flos": 23769345041280.0,
      "grad_norm": 3.529409922750284,
      "language_loss": 0.87857246,
      "learning_rate": 3.998513380280251e-06,
      "loss": 0.90233171,
      "num_input_tokens_seen": 7384625,
      "step": 349,
      "time_per_iteration": 2.6514475345611572
    },
    {
      "auxiliary_loss_clip": 0.01316873,
      "auxiliary_loss_mlp": 0.01073037,
      "balance_loss_clip": 1.08453143,
      "balance_loss_mlp": 1.05457139,
      "epoch": 0.042085011723681835,
      "flos": 11875336473600.0,
      "grad_norm": 3.672724278475927,
      "language_loss": 0.95130575,
      "learning_rate": 3.99848319969411e-06,
      "loss": 0.97520494,
      "num_input_tokens_seen": 7402225,
      "step": 350,
      "time_per_iteration": 2.6287927627563477
    },
    {
      "auxiliary_loss_clip": 0.01318783,
      "auxiliary_loss_mlp": 0.01060811,
      "balance_loss_clip": 1.08621323,
      "balance_loss_mlp": 1.04264331,
      "epoch": 0.042205254614320925,
      "flos": 16873957964160.0,
      "grad_norm": 2.680385273323739,
      "language_loss": 0.79362053,
      "learning_rate": 3.9984527159396564e-06,
      "loss": 0.81741649,
      "num_input_tokens_seen": 7420865,
      "step": 351,
      "time_per_iteration": 2.553727149963379
    },
    {
      "auxiliary_loss_clip": 0.0131367,
      "auxiliary_loss_mlp": 0.01055195,
      "balance_loss_clip": 1.08083344,
      "balance_loss_mlp": 1.03880346,
      "epoch": 0.04232549750496002,
      "flos": 25118508810240.0,
      "grad_norm": 2.600457258332655,
      "language_loss": 0.844832,
      "learning_rate": 3.9984219290215154e-06,
      "loss": 0.86852068,
      "num_input_tokens_seen": 7441040,
      "step": 352,
      "time_per_iteration": 2.6844756603240967
    },
    {
      "auxiliary_loss_clip": 0.01310442,
      "auxiliary_loss_mlp": 0.01050758,
      "balance_loss_clip": 1.08343387,
      "balance_loss_mlp": 1.03547573,
      "epoch": 0.04244574039559911,
      "flos": 26724541714560.0,
      "grad_norm": 1.7525776585142179,
      "language_loss": 0.89115554,
      "learning_rate": 3.998390838944356e-06,
      "loss": 0.9147675,
      "num_input_tokens_seen": 7462545,
      "step": 353,
      "time_per_iteration": 2.6615450382232666
    },
    {
      "auxiliary_loss_clip": 0.01312752,
      "auxiliary_loss_mlp": 0.01058389,
      "balance_loss_clip": 1.08385932,
      "balance_loss_mlp": 1.0421766,
      "epoch": 0.0425659832862382,
      "flos": 20923244951040.0,
      "grad_norm": 2.517121286811475,
      "language_loss": 0.90475357,
      "learning_rate": 3.998359445712895e-06,
      "loss": 0.92846501,
      "num_input_tokens_seen": 7481650,
      "step": 354,
      "time_per_iteration": 2.6801083087921143
    },
    {
      "auxiliary_loss_clip": 0.01311161,
      "auxiliary_loss_mlp": 0.01049483,
      "balance_loss_clip": 1.08098197,
      "balance_loss_mlp": 1.03372359,
      "epoch": 0.04268622617687729,
      "flos": 23331127115520.0,
      "grad_norm": 2.1684793421260946,
      "language_loss": 0.81118083,
      "learning_rate": 3.9983277493318955e-06,
      "loss": 0.83478725,
      "num_input_tokens_seen": 7500945,
      "step": 355,
      "time_per_iteration": 2.7060132026672363
    },
    {
      "auxiliary_loss_clip": 0.01313411,
      "auxiliary_loss_mlp": 0.01050624,
      "balance_loss_clip": 1.08021498,
      "balance_loss_mlp": 1.03444695,
      "epoch": 0.04280646906751638,
      "flos": 25994010908160.0,
      "grad_norm": 1.8249584420057783,
      "language_loss": 0.8123697,
      "learning_rate": 3.998295749806165e-06,
      "loss": 0.8360101,
      "num_input_tokens_seen": 7522170,
      "step": 356,
      "time_per_iteration": 2.6396872997283936
    },
    {
      "auxiliary_loss_clip": 0.01315192,
      "auxiliary_loss_mlp": 0.01062426,
      "balance_loss_clip": 1.08665609,
      "balance_loss_mlp": 1.04559386,
      "epoch": 0.04292671195815547,
      "flos": 26906824258560.0,
      "grad_norm": 1.9244036809622669,
      "language_loss": 0.831725,
      "learning_rate": 3.998263447140558e-06,
      "loss": 0.85550117,
      "num_input_tokens_seen": 7542370,
      "step": 357,
      "time_per_iteration": 2.7295985221862793
    },
    {
      "auxiliary_loss_clip": 0.01310574,
      "auxiliary_loss_mlp": 0.01048294,
      "balance_loss_clip": 1.08050609,
      "balance_loss_mlp": 1.03247499,
      "epoch": 0.04304695484879457,
      "flos": 39457315745280.0,
      "grad_norm": 1.9816575024332594,
      "language_loss": 0.81656575,
      "learning_rate": 3.998230841339976e-06,
      "loss": 0.84015447,
      "num_input_tokens_seen": 7564380,
      "step": 358,
      "time_per_iteration": 2.7430434226989746
    },
    {
      "auxiliary_loss_clip": 0.01310026,
      "auxiliary_loss_mlp": 0.01051019,
      "balance_loss_clip": 1.08493125,
      "balance_loss_mlp": 1.03512812,
      "epoch": 0.04316719773943366,
      "flos": 19646297475840.0,
      "grad_norm": 3.1155027298003577,
      "language_loss": 0.85117173,
      "learning_rate": 3.998197932409363e-06,
      "loss": 0.8747822,
      "num_input_tokens_seen": 7582390,
      "step": 359,
      "time_per_iteration": 2.6199841499328613
    },
    {
      "auxiliary_loss_clip": 0.01303974,
      "auxiliary_loss_mlp": 0.01054018,
      "balance_loss_clip": 1.07978261,
      "balance_loss_mlp": 1.03842568,
      "epoch": 0.04328744063007275,
      "flos": 22452320966400.0,
      "grad_norm": 2.261524887651382,
      "language_loss": 0.86431026,
      "learning_rate": 3.9981647203537125e-06,
      "loss": 0.88789016,
      "num_input_tokens_seen": 7599890,
      "step": 360,
      "time_per_iteration": 2.6075191497802734
    },
    {
      "auxiliary_loss_clip": 0.01307735,
      "auxiliary_loss_mlp": 0.01057943,
      "balance_loss_clip": 1.07911825,
      "balance_loss_mlp": 1.04264855,
      "epoch": 0.04340768352071184,
      "flos": 21283033530240.0,
      "grad_norm": 2.4821503522444894,
      "language_loss": 0.96026361,
      "learning_rate": 3.998131205178063e-06,
      "loss": 0.98392034,
      "num_input_tokens_seen": 7618360,
      "step": 361,
      "time_per_iteration": 2.6372666358947754
    },
    {
      "auxiliary_loss_clip": 0.01306035,
      "auxiliary_loss_mlp": 0.01057238,
      "balance_loss_clip": 1.07937527,
      "balance_loss_mlp": 1.041538,
      "epoch": 0.04352792641135093,
      "flos": 11583705951360.0,
      "grad_norm": 3.2413704551625644,
      "language_loss": 0.76968038,
      "learning_rate": 3.998097386887498e-06,
      "loss": 0.79331315,
      "num_input_tokens_seen": 7635435,
      "step": 362,
      "time_per_iteration": 2.5989463329315186
    },
    {
      "auxiliary_loss_clip": 0.01304669,
      "auxiliary_loss_mlp": 0.01061722,
      "balance_loss_clip": 1.0801574,
      "balance_loss_mlp": 1.0458436,
      "epoch": 0.04364816930199002,
      "flos": 23623547736960.0,
      "grad_norm": 1.9146675327073408,
      "language_loss": 0.84915894,
      "learning_rate": 3.998063265487148e-06,
      "loss": 0.87282288,
      "num_input_tokens_seen": 7656485,
      "step": 363,
      "time_per_iteration": 2.68257999420166
    },
    {
      "auxiliary_loss_clip": 0.01306892,
      "auxiliary_loss_mlp": 0.01053802,
      "balance_loss_clip": 1.08179235,
      "balance_loss_mlp": 1.03841257,
      "epoch": 0.043768412192629114,
      "flos": 14429734214400.0,
      "grad_norm": 2.4822064505549655,
      "language_loss": 0.81165266,
      "learning_rate": 3.99802884098219e-06,
      "loss": 0.83525956,
      "num_input_tokens_seen": 7674595,
      "step": 364,
      "time_per_iteration": 2.63155198097229
    },
    {
      "auxiliary_loss_clip": 0.01307018,
      "auxiliary_loss_mlp": 0.01046457,
      "balance_loss_clip": 1.07924104,
      "balance_loss_mlp": 1.03044713,
      "epoch": 0.043888655083268203,
      "flos": 26468893641600.0,
      "grad_norm": 2.5721327083215995,
      "language_loss": 0.82403159,
      "learning_rate": 3.997994113377845e-06,
      "loss": 0.84756637,
      "num_input_tokens_seen": 7693495,
      "step": 365,
      "time_per_iteration": 2.6758716106414795
    },
    {
      "auxiliary_loss_clip": 0.01306483,
      "auxiliary_loss_mlp": 0.01047664,
      "balance_loss_clip": 1.0798502,
      "balance_loss_mlp": 1.03164172,
      "epoch": 0.04400889797390729,
      "flos": 27235263242880.0,
      "grad_norm": 2.1632730658879202,
      "language_loss": 0.83007818,
      "learning_rate": 3.9979590826793815e-06,
      "loss": 0.85361964,
      "num_input_tokens_seen": 7714685,
      "step": 366,
      "time_per_iteration": 2.7164053916931152
    },
    {
      "auxiliary_loss_clip": 0.01308812,
      "auxiliary_loss_mlp": 0.01050242,
      "balance_loss_clip": 1.08215368,
      "balance_loss_mlp": 1.03429151,
      "epoch": 0.04412914086454638,
      "flos": 20119528183680.0,
      "grad_norm": 2.1105660367720733,
      "language_loss": 0.80987555,
      "learning_rate": 3.997923748892113e-06,
      "loss": 0.83346605,
      "num_input_tokens_seen": 7734005,
      "step": 367,
      "time_per_iteration": 3.5883920192718506
    },
    {
      "auxiliary_loss_clip": 0.01303969,
      "auxiliary_loss_mlp": 0.01051332,
      "balance_loss_clip": 1.08124137,
      "balance_loss_mlp": 1.03651381,
      "epoch": 0.04424938375518547,
      "flos": 22604618632320.0,
      "grad_norm": 2.230129199666655,
      "language_loss": 0.88768589,
      "learning_rate": 3.9978881120214015e-06,
      "loss": 0.91123891,
      "num_input_tokens_seen": 7755525,
      "step": 368,
      "time_per_iteration": 2.6816582679748535
    },
    {
      "auxiliary_loss_clip": 0.0130482,
      "auxiliary_loss_mlp": 0.0104731,
      "balance_loss_clip": 1.07804942,
      "balance_loss_mlp": 1.03124034,
      "epoch": 0.04436962664582456,
      "flos": 24132365844480.0,
      "grad_norm": 2.1701214121100953,
      "language_loss": 0.79309767,
      "learning_rate": 3.997852172072652e-06,
      "loss": 0.81661898,
      "num_input_tokens_seen": 7776740,
      "step": 369,
      "time_per_iteration": 2.6450181007385254
    },
    {
      "auxiliary_loss_clip": 0.01304524,
      "auxiliary_loss_mlp": 0.01060232,
      "balance_loss_clip": 1.0780853,
      "balance_loss_mlp": 1.04432988,
      "epoch": 0.04448986953646366,
      "flos": 18222906251520.0,
      "grad_norm": 6.8067332197436805,
      "language_loss": 0.89093733,
      "learning_rate": 3.9978159290513155e-06,
      "loss": 0.91458488,
      "num_input_tokens_seen": 7794820,
      "step": 370,
      "time_per_iteration": 4.46401047706604
    },
    {
      "auxiliary_loss_clip": 0.01306662,
      "auxiliary_loss_mlp": 0.01063061,
      "balance_loss_clip": 1.08009195,
      "balance_loss_mlp": 1.04675293,
      "epoch": 0.04461011242710275,
      "flos": 30117920400000.0,
      "grad_norm": 2.2806416651338726,
      "language_loss": 0.80062997,
      "learning_rate": 3.997779382962892e-06,
      "loss": 0.82432723,
      "num_input_tokens_seen": 7817705,
      "step": 371,
      "time_per_iteration": 2.716287612915039
    },
    {
      "auxiliary_loss_clip": 0.01300306,
      "auxiliary_loss_mlp": 0.01049359,
      "balance_loss_clip": 1.07703364,
      "balance_loss_mlp": 1.03331375,
      "epoch": 0.04473035531774184,
      "flos": 29752529299200.0,
      "grad_norm": 2.8713693165489214,
      "language_loss": 0.7404719,
      "learning_rate": 3.997742533812924e-06,
      "loss": 0.76396859,
      "num_input_tokens_seen": 7840970,
      "step": 372,
      "time_per_iteration": 3.597639322280884
    },
    {
      "auxiliary_loss_clip": 0.0130737,
      "auxiliary_loss_mlp": 0.01062919,
      "balance_loss_clip": 1.08264673,
      "balance_loss_mlp": 1.04739833,
      "epoch": 0.04485059820838093,
      "flos": 13151565676800.0,
      "grad_norm": 3.721350573565986,
      "language_loss": 0.92438346,
      "learning_rate": 3.997705381607001e-06,
      "loss": 0.94808638,
      "num_input_tokens_seen": 7857785,
      "step": 373,
      "time_per_iteration": 2.6370835304260254
    },
    {
      "auxiliary_loss_clip": 0.01240654,
      "auxiliary_loss_mlp": 0.01014332,
      "balance_loss_clip": 1.07194924,
      "balance_loss_mlp": 1.00264966,
      "epoch": 0.04497084109902002,
      "flos": 68094209548800.0,
      "grad_norm": 1.0130841362736653,
      "language_loss": 0.60316265,
      "learning_rate": 3.997667926350761e-06,
      "loss": 0.62571251,
      "num_input_tokens_seen": 7916115,
      "step": 374,
      "time_per_iteration": 3.064362049102783
    },
    {
      "auxiliary_loss_clip": 0.01235245,
      "auxiliary_loss_mlp": 0.01014926,
      "balance_loss_clip": 1.06872678,
      "balance_loss_mlp": 1.00391126,
      "epoch": 0.04509108398965911,
      "flos": 64342263346560.0,
      "grad_norm": 0.932756343235059,
      "language_loss": 0.57780635,
      "learning_rate": 3.997630168049886e-06,
      "loss": 0.60030806,
      "num_input_tokens_seen": 7974480,
      "step": 375,
      "time_per_iteration": 3.177156448364258
    },
    {
      "auxiliary_loss_clip": 0.01307006,
      "auxiliary_loss_mlp": 0.01057526,
      "balance_loss_clip": 1.07983327,
      "balance_loss_mlp": 1.04138529,
      "epoch": 0.045211326880298205,
      "flos": 22271115830400.0,
      "grad_norm": 2.1767695322061718,
      "language_loss": 0.77544922,
      "learning_rate": 3.997592106710101e-06,
      "loss": 0.79909456,
      "num_input_tokens_seen": 7993940,
      "step": 376,
      "time_per_iteration": 2.628694772720337
    },
    {
      "auxiliary_loss_clip": 0.01299126,
      "auxiliary_loss_mlp": 0.01047582,
      "balance_loss_clip": 1.07650769,
      "balance_loss_mlp": 1.03221571,
      "epoch": 0.045331569770937295,
      "flos": 32159441796480.0,
      "grad_norm": 2.651121759076787,
      "language_loss": 0.65941101,
      "learning_rate": 3.997553742337182e-06,
      "loss": 0.68287814,
      "num_input_tokens_seen": 8013365,
      "step": 377,
      "time_per_iteration": 2.6504950523376465
    },
    {
      "auxiliary_loss_clip": 0.01303262,
      "auxiliary_loss_mlp": 0.01051838,
      "balance_loss_clip": 1.07878208,
      "balance_loss_mlp": 1.03613842,
      "epoch": 0.045451812661576385,
      "flos": 22163455791360.0,
      "grad_norm": 1.8452475117039886,
      "language_loss": 0.91413713,
      "learning_rate": 3.997515074936949e-06,
      "loss": 0.93768811,
      "num_input_tokens_seen": 8034240,
      "step": 378,
      "time_per_iteration": 2.643339157104492
    },
    {
      "auxiliary_loss_clip": 0.01303313,
      "auxiliary_loss_mlp": 0.01054347,
      "balance_loss_clip": 1.07820749,
      "balance_loss_mlp": 1.03913641,
      "epoch": 0.045572055552215475,
      "flos": 16581968305920.0,
      "grad_norm": 3.9181129607126786,
      "language_loss": 0.86764485,
      "learning_rate": 3.997476104515268e-06,
      "loss": 0.89122146,
      "num_input_tokens_seen": 8052430,
      "step": 379,
      "time_per_iteration": 2.6107184886932373
    },
    {
      "auxiliary_loss_clip": 0.01301038,
      "auxiliary_loss_mlp": 0.0105295,
      "balance_loss_clip": 1.07977223,
      "balance_loss_mlp": 1.03812063,
      "epoch": 0.045692298442854565,
      "flos": 17603375448960.0,
      "grad_norm": 2.467920236688807,
      "language_loss": 0.7758432,
      "learning_rate": 3.9974368310780485e-06,
      "loss": 0.79938304,
      "num_input_tokens_seen": 8069605,
      "step": 380,
      "time_per_iteration": 2.623745918273926
    },
    {
      "auxiliary_loss_clip": 0.01309622,
      "auxiliary_loss_mlp": 0.01053533,
      "balance_loss_clip": 1.08126259,
      "balance_loss_mlp": 1.03642607,
      "epoch": 0.045812541333493655,
      "flos": 26761098781440.0,
      "grad_norm": 2.436186170523636,
      "language_loss": 0.74603862,
      "learning_rate": 3.997397254631251e-06,
      "loss": 0.76967019,
      "num_input_tokens_seen": 8090225,
      "step": 381,
      "time_per_iteration": 2.642836093902588
    },
    {
      "auxiliary_loss_clip": 0.01226789,
      "auxiliary_loss_mlp": 0.01029283,
      "balance_loss_clip": 1.06717908,
      "balance_loss_mlp": 1.01822078,
      "epoch": 0.04593278422413275,
      "flos": 60250349894400.0,
      "grad_norm": 0.8338471093938346,
      "language_loss": 0.60097086,
      "learning_rate": 3.997357375180878e-06,
      "loss": 0.62353158,
      "num_input_tokens_seen": 8154505,
      "step": 382,
      "time_per_iteration": 3.2914223670959473
    },
    {
      "auxiliary_loss_clip": 0.01303448,
      "auxiliary_loss_mlp": 0.01048632,
      "balance_loss_clip": 1.0774368,
      "balance_loss_mlp": 1.03232431,
      "epoch": 0.04605302711477184,
      "flos": 21799249839360.0,
      "grad_norm": 1.8658699069621885,
      "language_loss": 0.74872172,
      "learning_rate": 3.997317192732979e-06,
      "loss": 0.77224255,
      "num_input_tokens_seen": 8173285,
      "step": 383,
      "time_per_iteration": 2.6295504570007324
    },
    {
      "auxiliary_loss_clip": 0.0130447,
      "auxiliary_loss_mlp": 0.01057159,
      "balance_loss_clip": 1.07918286,
      "balance_loss_mlp": 1.04117346,
      "epoch": 0.04617327000541093,
      "flos": 19459705299840.0,
      "grad_norm": 2.1476817624981606,
      "language_loss": 0.82472771,
      "learning_rate": 3.99727670729365e-06,
      "loss": 0.84834403,
      "num_input_tokens_seen": 8191845,
      "step": 384,
      "time_per_iteration": 2.6378796100616455
    },
    {
      "auxiliary_loss_clip": 0.0130143,
      "auxiliary_loss_mlp": 0.01051381,
      "balance_loss_clip": 1.08129478,
      "balance_loss_mlp": 1.03639603,
      "epoch": 0.04629351289605002,
      "flos": 25411468135680.0,
      "grad_norm": 1.6975515691023697,
      "language_loss": 0.77660626,
      "learning_rate": 3.997235918869033e-06,
      "loss": 0.80013442,
      "num_input_tokens_seen": 8212880,
      "step": 385,
      "time_per_iteration": 2.665940046310425
    },
    {
      "auxiliary_loss_clip": 0.01302548,
      "auxiliary_loss_mlp": 0.01048212,
      "balance_loss_clip": 1.08127284,
      "balance_loss_mlp": 1.03381169,
      "epoch": 0.04641375578668911,
      "flos": 20558284813440.0,
      "grad_norm": 2.15279852419483,
      "language_loss": 0.8256557,
      "learning_rate": 3.997194827465315e-06,
      "loss": 0.84916323,
      "num_input_tokens_seen": 8231475,
      "step": 386,
      "time_per_iteration": 2.617692708969116
    },
    {
      "auxiliary_loss_clip": 0.01302221,
      "auxiliary_loss_mlp": 0.01045698,
      "balance_loss_clip": 1.07787466,
      "balance_loss_mlp": 1.03073716,
      "epoch": 0.0465339986773282,
      "flos": 13188661447680.0,
      "grad_norm": 6.235576590941023,
      "language_loss": 0.91232705,
      "learning_rate": 3.997153433088728e-06,
      "loss": 0.93580627,
      "num_input_tokens_seen": 8248600,
      "step": 387,
      "time_per_iteration": 2.5979676246643066
    },
    {
      "auxiliary_loss_clip": 0.01301638,
      "auxiliary_loss_mlp": 0.01047917,
      "balance_loss_clip": 1.07884145,
      "balance_loss_mlp": 1.03195477,
      "epoch": 0.0466542415679673,
      "flos": 25556547168000.0,
      "grad_norm": 2.885524261584598,
      "language_loss": 0.81159687,
      "learning_rate": 3.997111735745554e-06,
      "loss": 0.83509243,
      "num_input_tokens_seen": 8271570,
      "step": 388,
      "time_per_iteration": 2.6200788021087646
    },
    {
      "auxiliary_loss_clip": 0.01298932,
      "auxiliary_loss_mlp": 0.01055129,
      "balance_loss_clip": 1.07794368,
      "balance_loss_mlp": 1.03828526,
      "epoch": 0.04677448445860639,
      "flos": 22236749493120.0,
      "grad_norm": 2.150535611730706,
      "language_loss": 0.82617331,
      "learning_rate": 3.997069735442118e-06,
      "loss": 0.84971392,
      "num_input_tokens_seen": 8291265,
      "step": 389,
      "time_per_iteration": 2.617143154144287
    },
    {
      "auxiliary_loss_clip": 0.01299153,
      "auxiliary_loss_mlp": 0.01052262,
      "balance_loss_clip": 1.07805562,
      "balance_loss_mlp": 1.03713405,
      "epoch": 0.04689472734924548,
      "flos": 28147825198080.0,
      "grad_norm": 1.6297291704072627,
      "language_loss": 0.8030082,
      "learning_rate": 3.997027432184792e-06,
      "loss": 0.82652235,
      "num_input_tokens_seen": 8315925,
      "step": 390,
      "time_per_iteration": 2.718494176864624
    },
    {
      "auxiliary_loss_clip": 0.01299901,
      "auxiliary_loss_mlp": 0.01050994,
      "balance_loss_clip": 1.07847774,
      "balance_loss_mlp": 1.03655815,
      "epoch": 0.04701497023988457,
      "flos": 23148952312320.0,
      "grad_norm": 1.9645380012611247,
      "language_loss": 0.89395046,
      "learning_rate": 3.99698482597999e-06,
      "loss": 0.91745949,
      "num_input_tokens_seen": 8333605,
      "step": 391,
      "time_per_iteration": 2.665731430053711
    },
    {
      "auxiliary_loss_clip": 0.01217278,
      "auxiliary_loss_mlp": 0.01031215,
      "balance_loss_clip": 1.06213295,
      "balance_loss_mlp": 1.02143979,
      "epoch": 0.04713521313052366,
      "flos": 64827668764800.0,
      "grad_norm": 0.8718167318713126,
      "language_loss": 0.63896096,
      "learning_rate": 3.99694191683418e-06,
      "loss": 0.66144586,
      "num_input_tokens_seen": 8394405,
      "step": 392,
      "time_per_iteration": 3.194749116897583
    },
    {
      "auxiliary_loss_clip": 0.01303344,
      "auxiliary_loss_mlp": 0.01047637,
      "balance_loss_clip": 1.08180571,
      "balance_loss_mlp": 1.03212833,
      "epoch": 0.047255456021162746,
      "flos": 18771585477120.0,
      "grad_norm": 2.489187848980738,
      "language_loss": 0.81625026,
      "learning_rate": 3.996898704753867e-06,
      "loss": 0.83976007,
      "num_input_tokens_seen": 8412355,
      "step": 393,
      "time_per_iteration": 2.616530179977417
    },
    {
      "auxiliary_loss_clip": 0.01298025,
      "auxiliary_loss_mlp": 0.01045276,
      "balance_loss_clip": 1.07661057,
      "balance_loss_mlp": 1.03070247,
      "epoch": 0.04737569891180184,
      "flos": 22053820504320.0,
      "grad_norm": 2.3459453534729797,
      "language_loss": 0.87799358,
      "learning_rate": 3.996855189745609e-06,
      "loss": 0.90142661,
      "num_input_tokens_seen": 8431620,
      "step": 394,
      "time_per_iteration": 3.605282783508301
    },
    {
      "auxiliary_loss_clip": 0.01297658,
      "auxiliary_loss_mlp": 0.01052702,
      "balance_loss_clip": 1.0766542,
      "balance_loss_mlp": 1.03721642,
      "epoch": 0.04749594180244093,
      "flos": 29057370410880.0,
      "grad_norm": 2.188929514797865,
      "language_loss": 0.92666864,
      "learning_rate": 3.996811371816007e-06,
      "loss": 0.95017219,
      "num_input_tokens_seen": 8454045,
      "step": 395,
      "time_per_iteration": 2.7015435695648193
    },
    {
      "auxiliary_loss_clip": 0.01299281,
      "auxiliary_loss_mlp": 0.01056084,
      "balance_loss_clip": 1.07994199,
      "balance_loss_mlp": 1.04192221,
      "epoch": 0.04761618469308002,
      "flos": 35112268172160.0,
      "grad_norm": 2.1449571025072034,
      "language_loss": 0.77632356,
      "learning_rate": 3.996767250971707e-06,
      "loss": 0.79987723,
      "num_input_tokens_seen": 8476785,
      "step": 396,
      "time_per_iteration": 3.7139105796813965
    },
    {
      "auxiliary_loss_clip": 0.01302771,
      "auxiliary_loss_mlp": 0.01047977,
      "balance_loss_clip": 1.08159506,
      "balance_loss_mlp": 1.0326947,
      "epoch": 0.04773642758371911,
      "flos": 25630702796160.0,
      "grad_norm": 2.129692985405832,
      "language_loss": 0.86819255,
      "learning_rate": 3.996722827219403e-06,
      "loss": 0.89170003,
      "num_input_tokens_seen": 8498400,
      "step": 397,
      "time_per_iteration": 3.635160446166992
    },
    {
      "auxiliary_loss_clip": 0.01304487,
      "auxiliary_loss_mlp": 0.0105706,
      "balance_loss_clip": 1.08283186,
      "balance_loss_mlp": 1.04159832,
      "epoch": 0.0478566704743582,
      "flos": 20631506688000.0,
      "grad_norm": 2.4099673282326153,
      "language_loss": 0.82337797,
      "learning_rate": 3.996678100565833e-06,
      "loss": 0.84699345,
      "num_input_tokens_seen": 8517455,
      "step": 398,
      "time_per_iteration": 2.596437931060791
    },
    {
      "auxiliary_loss_clip": 0.01294864,
      "auxiliary_loss_mlp": 0.0105344,
      "balance_loss_clip": 1.07622218,
      "balance_loss_mlp": 1.03745353,
      "epoch": 0.04797691336499729,
      "flos": 18835721210880.0,
      "grad_norm": 4.1332630828220225,
      "language_loss": 0.88947976,
      "learning_rate": 3.996633071017783e-06,
      "loss": 0.91296285,
      "num_input_tokens_seen": 8534085,
      "step": 399,
      "time_per_iteration": 3.5083529949188232
    },
    {
      "auxiliary_loss_clip": 0.01295827,
      "auxiliary_loss_mlp": 0.01050832,
      "balance_loss_clip": 1.07759094,
      "balance_loss_mlp": 1.035496,
      "epoch": 0.04809715625563638,
      "flos": 21099673578240.0,
      "grad_norm": 2.3156492153757986,
      "language_loss": 0.81554347,
      "learning_rate": 3.996587738582084e-06,
      "loss": 0.83901,
      "num_input_tokens_seen": 8550885,
      "step": 400,
      "time_per_iteration": 2.6151187419891357
    },
    {
      "auxiliary_loss_clip": 0.0129502,
      "auxiliary_loss_mlp": 0.01041089,
      "balance_loss_clip": 1.07525694,
      "balance_loss_mlp": 1.02673602,
      "epoch": 0.04821739914627548,
      "flos": 23805650712960.0,
      "grad_norm": 14.731276881924632,
      "language_loss": 0.86220276,
      "learning_rate": 3.9965421032656115e-06,
      "loss": 0.88556385,
      "num_input_tokens_seen": 8570815,
      "step": 401,
      "time_per_iteration": 2.7041215896606445
    },
    {
      "auxiliary_loss_clip": 0.01296896,
      "auxiliary_loss_mlp": 0.01046446,
      "balance_loss_clip": 1.07614183,
      "balance_loss_mlp": 1.03041267,
      "epoch": 0.04833764203691457,
      "flos": 22200587475840.0,
      "grad_norm": 8.90084233581978,
      "language_loss": 0.94029123,
      "learning_rate": 3.99649616507529e-06,
      "loss": 0.96372473,
      "num_input_tokens_seen": 8589910,
      "step": 402,
      "time_per_iteration": 2.587742805480957
    },
    {
      "auxiliary_loss_clip": 0.01204678,
      "auxiliary_loss_mlp": 0.01013351,
      "balance_loss_clip": 1.05373991,
      "balance_loss_mlp": 1.00472057,
      "epoch": 0.04845788492755366,
      "flos": 65904376896000.0,
      "grad_norm": 0.9036124254548012,
      "language_loss": 0.631926,
      "learning_rate": 3.996449924018088e-06,
      "loss": 0.6541062,
      "num_input_tokens_seen": 8650370,
      "step": 403,
      "time_per_iteration": 3.1111562252044678
    },
    {
      "auxiliary_loss_clip": 0.01293671,
      "auxiliary_loss_mlp": 0.01052111,
      "balance_loss_clip": 1.07666731,
      "balance_loss_mlp": 1.03855109,
      "epoch": 0.04857812781819275,
      "flos": 19281301424640.0,
      "grad_norm": 2.1645317554984507,
      "language_loss": 0.79347521,
      "learning_rate": 3.99640338010102e-06,
      "loss": 0.81693304,
      "num_input_tokens_seen": 8669475,
      "step": 404,
      "time_per_iteration": 2.6097238063812256
    },
    {
      "auxiliary_loss_clip": 0.01292754,
      "auxiliary_loss_mlp": 0.01043702,
      "balance_loss_clip": 1.07449675,
      "balance_loss_mlp": 1.02833629,
      "epoch": 0.04869837070883184,
      "flos": 24062376193920.0,
      "grad_norm": 2.141331495575818,
      "language_loss": 0.78649139,
      "learning_rate": 3.996356533331146e-06,
      "loss": 0.80985594,
      "num_input_tokens_seen": 8691345,
      "step": 405,
      "time_per_iteration": 2.589722156524658
    },
    {
      "auxiliary_loss_clip": 0.01305055,
      "auxiliary_loss_mlp": 0.01042853,
      "balance_loss_clip": 1.0783397,
      "balance_loss_mlp": 1.02783287,
      "epoch": 0.04881861359947093,
      "flos": 25187169657600.0,
      "grad_norm": 4.163046764289541,
      "language_loss": 0.61563182,
      "learning_rate": 3.996309383715573e-06,
      "loss": 0.63911086,
      "num_input_tokens_seen": 8710125,
      "step": 406,
      "time_per_iteration": 2.6741931438446045
    },
    {
      "auxiliary_loss_clip": 0.01301958,
      "auxiliary_loss_mlp": 0.0104081,
      "balance_loss_clip": 1.07978034,
      "balance_loss_mlp": 1.02664208,
      "epoch": 0.048938856490110025,
      "flos": 16362913213440.0,
      "grad_norm": 2.0206622820672173,
      "language_loss": 0.73690397,
      "learning_rate": 3.996261931261454e-06,
      "loss": 0.76033163,
      "num_input_tokens_seen": 8728705,
      "step": 407,
      "time_per_iteration": 2.5919785499572754
    },
    {
      "auxiliary_loss_clip": 0.01297076,
      "auxiliary_loss_mlp": 0.01044636,
      "balance_loss_clip": 1.07849193,
      "balance_loss_mlp": 1.02996111,
      "epoch": 0.049059099380749115,
      "flos": 29895094379520.0,
      "grad_norm": 1.7599759901633647,
      "language_loss": 0.86384964,
      "learning_rate": 3.996214175975987e-06,
      "loss": 0.88726676,
      "num_input_tokens_seen": 8749225,
      "step": 408,
      "time_per_iteration": 2.6972897052764893
    },
    {
      "auxiliary_loss_clip": 0.01301189,
      "auxiliary_loss_mlp": 0.0105167,
      "balance_loss_clip": 1.08010852,
      "balance_loss_mlp": 1.03642356,
      "epoch": 0.049179342271388204,
      "flos": 35918858027520.0,
      "grad_norm": 2.7067458787902963,
      "language_loss": 0.79011166,
      "learning_rate": 3.996166117866417e-06,
      "loss": 0.8136403,
      "num_input_tokens_seen": 8771160,
      "step": 409,
      "time_per_iteration": 2.7547659873962402
    },
    {
      "auxiliary_loss_clip": 0.01293891,
      "auxiliary_loss_mlp": 0.01044703,
      "balance_loss_clip": 1.07579851,
      "balance_loss_mlp": 1.03029633,
      "epoch": 0.049299585162027294,
      "flos": 14611226659200.0,
      "grad_norm": 2.2818809017612334,
      "language_loss": 0.86740005,
      "learning_rate": 3.996117756940035e-06,
      "loss": 0.89078605,
      "num_input_tokens_seen": 8787845,
      "step": 410,
      "time_per_iteration": 2.640611171722412
    },
    {
      "auxiliary_loss_clip": 0.01297865,
      "auxiliary_loss_mlp": 0.01048514,
      "balance_loss_clip": 1.07881618,
      "balance_loss_mlp": 1.03417897,
      "epoch": 0.049419828052666384,
      "flos": 19567939956480.0,
      "grad_norm": 2.370835452333674,
      "language_loss": 0.97567695,
      "learning_rate": 3.996069093204175e-06,
      "loss": 0.99914074,
      "num_input_tokens_seen": 8803805,
      "step": 411,
      "time_per_iteration": 2.6025469303131104
    },
    {
      "auxiliary_loss_clip": 0.01303283,
      "auxiliary_loss_mlp": 0.01053015,
      "balance_loss_clip": 1.08123231,
      "balance_loss_mlp": 1.03757739,
      "epoch": 0.049540070943305474,
      "flos": 13659916907520.0,
      "grad_norm": 5.394416986978219,
      "language_loss": 0.87944829,
      "learning_rate": 3.996020126666221e-06,
      "loss": 0.90301132,
      "num_input_tokens_seen": 8820785,
      "step": 412,
      "time_per_iteration": 2.6566362380981445
    },
    {
      "auxiliary_loss_clip": 0.01297607,
      "auxiliary_loss_mlp": 0.01044814,
      "balance_loss_clip": 1.07866502,
      "balance_loss_mlp": 1.03058004,
      "epoch": 0.04966031383394457,
      "flos": 21832035978240.0,
      "grad_norm": 2.1039589901887843,
      "language_loss": 0.82274944,
      "learning_rate": 3.995970857333601e-06,
      "loss": 0.84617364,
      "num_input_tokens_seen": 8841195,
      "step": 413,
      "time_per_iteration": 2.653988838195801
    },
    {
      "auxiliary_loss_clip": 0.01297741,
      "auxiliary_loss_mlp": 0.01049105,
      "balance_loss_clip": 1.07647145,
      "balance_loss_mlp": 1.03419161,
      "epoch": 0.04978055672458366,
      "flos": 28618793349120.0,
      "grad_norm": 2.0464942565751647,
      "language_loss": 0.7959649,
      "learning_rate": 3.995921285213789e-06,
      "loss": 0.81943333,
      "num_input_tokens_seen": 8861455,
      "step": 414,
      "time_per_iteration": 2.711378812789917
    },
    {
      "auxiliary_loss_clip": 0.01293507,
      "auxiliary_loss_mlp": 0.01045318,
      "balance_loss_clip": 1.07577395,
      "balance_loss_mlp": 1.03154302,
      "epoch": 0.04990079961522275,
      "flos": 19828220883840.0,
      "grad_norm": 3.0656789623997387,
      "language_loss": 0.80776167,
      "learning_rate": 3.995871410314305e-06,
      "loss": 0.83114994,
      "num_input_tokens_seen": 8880015,
      "step": 415,
      "time_per_iteration": 2.601487636566162
    },
    {
      "auxiliary_loss_clip": 0.01189121,
      "auxiliary_loss_mlp": 0.01009629,
      "balance_loss_clip": 1.04143023,
      "balance_loss_mlp": 1.00195158,
      "epoch": 0.05002104250586184,
      "flos": 62735045293440.0,
      "grad_norm": 0.9177646654507338,
      "language_loss": 0.59609032,
      "learning_rate": 3.995821232642714e-06,
      "loss": 0.61807775,
      "num_input_tokens_seen": 8938420,
      "step": 416,
      "time_per_iteration": 3.312163829803467
    },
    {
      "auxiliary_loss_clip": 0.01293936,
      "auxiliary_loss_mlp": 0.01050276,
      "balance_loss_clip": 1.07725143,
      "balance_loss_mlp": 1.03600669,
      "epoch": 0.05014128539650093,
      "flos": 27928518710400.0,
      "grad_norm": 3.081702941640802,
      "language_loss": 0.82342541,
      "learning_rate": 3.995770752206629e-06,
      "loss": 0.84686756,
      "num_input_tokens_seen": 8959495,
      "step": 417,
      "time_per_iteration": 2.7478768825531006
    },
    {
      "auxiliary_loss_clip": 0.01295686,
      "auxiliary_loss_mlp": 0.01044702,
      "balance_loss_clip": 1.07731879,
      "balance_loss_mlp": 1.02950883,
      "epoch": 0.05026152828714002,
      "flos": 17705576620800.0,
      "grad_norm": 2.947116415213009,
      "language_loss": 0.97330296,
      "learning_rate": 3.995719969013709e-06,
      "loss": 0.9967069,
      "num_input_tokens_seen": 8976675,
      "step": 418,
      "time_per_iteration": 2.6036181449890137
    },
    {
      "auxiliary_loss_clip": 0.01295185,
      "auxiliary_loss_mlp": 0.01050167,
      "balance_loss_clip": 1.07516098,
      "balance_loss_mlp": 1.03502798,
      "epoch": 0.05038177117777912,
      "flos": 19133277477120.0,
      "grad_norm": 3.7563761877767035,
      "language_loss": 0.85910213,
      "learning_rate": 3.995668883071655e-06,
      "loss": 0.88255572,
      "num_input_tokens_seen": 8992900,
      "step": 419,
      "time_per_iteration": 2.6876790523529053
    },
    {
      "auxiliary_loss_clip": 0.01297063,
      "auxiliary_loss_mlp": 0.01046225,
      "balance_loss_clip": 1.07795119,
      "balance_loss_mlp": 1.0314728,
      "epoch": 0.050502014068418206,
      "flos": 20667704618880.0,
      "grad_norm": 2.5868675327289465,
      "language_loss": 0.912099,
      "learning_rate": 3.995617494388219e-06,
      "loss": 0.93553191,
      "num_input_tokens_seen": 9011020,
      "step": 420,
      "time_per_iteration": 2.7568821907043457
    },
    {
      "auxiliary_loss_clip": 0.01291579,
      "auxiliary_loss_mlp": 0.01038577,
      "balance_loss_clip": 1.07109022,
      "balance_loss_mlp": 1.02381289,
      "epoch": 0.050622256959057296,
      "flos": 21361103740800.0,
      "grad_norm": 3.065825895423579,
      "language_loss": 0.806979,
      "learning_rate": 3.995565802971196e-06,
      "loss": 0.83028054,
      "num_input_tokens_seen": 9030995,
      "step": 421,
      "time_per_iteration": 3.5652523040771484
    },
    {
      "auxiliary_loss_clip": 0.0128996,
      "auxiliary_loss_mlp": 0.01049205,
      "balance_loss_clip": 1.07150221,
      "balance_loss_mlp": 1.03518581,
      "epoch": 0.050742499849696386,
      "flos": 27673588909440.0,
      "grad_norm": 1.9908316308049179,
      "language_loss": 0.67475724,
      "learning_rate": 3.995513808828427e-06,
      "loss": 0.69814897,
      "num_input_tokens_seen": 9053790,
      "step": 422,
      "time_per_iteration": 2.6868820190429688
    },
    {
      "auxiliary_loss_clip": 0.01292722,
      "auxiliary_loss_mlp": 0.0104245,
      "balance_loss_clip": 1.07264137,
      "balance_loss_mlp": 1.02789497,
      "epoch": 0.050862742740335476,
      "flos": 19865999013120.0,
      "grad_norm": 2.2408615995850467,
      "language_loss": 0.76636428,
      "learning_rate": 3.9954615119678e-06,
      "loss": 0.78971601,
      "num_input_tokens_seen": 9072345,
      "step": 423,
      "time_per_iteration": 3.677079200744629
    },
    {
      "auxiliary_loss_clip": 0.0128325,
      "auxiliary_loss_mlp": 0.0105085,
      "balance_loss_clip": 1.07210815,
      "balance_loss_mlp": 1.03599644,
      "epoch": 0.050982985630974566,
      "flos": 22085098272000.0,
      "grad_norm": 2.096785365413762,
      "language_loss": 0.80689299,
      "learning_rate": 3.995408912397248e-06,
      "loss": 0.83023393,
      "num_input_tokens_seen": 9090240,
      "step": 424,
      "time_per_iteration": 3.707489252090454
    },
    {
      "auxiliary_loss_clip": 0.01297063,
      "auxiliary_loss_mlp": 0.01049042,
      "balance_loss_clip": 1.07514572,
      "balance_loss_mlp": 1.03351474,
      "epoch": 0.05110322852161366,
      "flos": 20740962407040.0,
      "grad_norm": 6.1113531099106995,
      "language_loss": 0.93314332,
      "learning_rate": 3.99535601012475e-06,
      "loss": 0.95660436,
      "num_input_tokens_seen": 9105570,
      "step": 425,
      "time_per_iteration": 2.655691623687744
    },
    {
      "auxiliary_loss_clip": 0.01290835,
      "auxiliary_loss_mlp": 0.01209465,
      "balance_loss_clip": 1.07162225,
      "balance_loss_mlp": 1.00014794,
      "epoch": 0.05122347141225275,
      "flos": 28547295327360.0,
      "grad_norm": 1.857625845223312,
      "language_loss": 0.75507462,
      "learning_rate": 3.995302805158333e-06,
      "loss": 0.78007764,
      "num_input_tokens_seen": 9128225,
      "step": 426,
      "time_per_iteration": 3.6663241386413574
    },
    {
      "auxiliary_loss_clip": 0.01288063,
      "auxiliary_loss_mlp": 0.01054503,
      "balance_loss_clip": 1.07300806,
      "balance_loss_mlp": 1.03792143,
      "epoch": 0.05134371430289184,
      "flos": 19722679747200.0,
      "grad_norm": 2.0137028729097675,
      "language_loss": 0.8354466,
      "learning_rate": 3.9952492975060665e-06,
      "loss": 0.85887223,
      "num_input_tokens_seen": 9148295,
      "step": 427,
      "time_per_iteration": 2.6938693523406982
    },
    {
      "auxiliary_loss_clip": 0.01292122,
      "auxiliary_loss_mlp": 0.01044868,
      "balance_loss_clip": 1.07482719,
      "balance_loss_mlp": 1.03062284,
      "epoch": 0.05146395719353093,
      "flos": 34458945649920.0,
      "grad_norm": 2.602374547236063,
      "language_loss": 0.8493005,
      "learning_rate": 3.995195487176067e-06,
      "loss": 0.87267047,
      "num_input_tokens_seen": 9168525,
      "step": 428,
      "time_per_iteration": 2.774662971496582
    },
    {
      "auxiliary_loss_clip": 0.01295034,
      "auxiliary_loss_mlp": 0.01048557,
      "balance_loss_clip": 1.0772388,
      "balance_loss_mlp": 1.03371561,
      "epoch": 0.05158420008417002,
      "flos": 21760286561280.0,
      "grad_norm": 2.177354602266061,
      "language_loss": 0.85561442,
      "learning_rate": 3.995141374176499e-06,
      "loss": 0.87905037,
      "num_input_tokens_seen": 9186920,
      "step": 429,
      "time_per_iteration": 2.6314704418182373
    },
    {
      "auxiliary_loss_clip": 0.01191035,
      "auxiliary_loss_mlp": 0.01201348,
      "balance_loss_clip": 1.03692222,
      "balance_loss_mlp": 0.99992239,
      "epoch": 0.05170444297480911,
      "flos": 72553956226560.0,
      "grad_norm": 0.8762429432709465,
      "language_loss": 0.63090503,
      "learning_rate": 3.995086958515572e-06,
      "loss": 0.65482891,
      "num_input_tokens_seen": 9244940,
      "step": 430,
      "time_per_iteration": 3.265130043029785
    },
    {
      "auxiliary_loss_clip": 0.01178099,
      "auxiliary_loss_mlp": 0.01201249,
      "balance_loss_clip": 1.03659534,
      "balance_loss_mlp": 0.9998588,
      "epoch": 0.05182468586544821,
      "flos": 62416159326720.0,
      "grad_norm": 0.8642311675668758,
      "language_loss": 0.59970492,
      "learning_rate": 3.995032240201538e-06,
      "loss": 0.62349838,
      "num_input_tokens_seen": 9307335,
      "step": 431,
      "time_per_iteration": 3.122199058532715
    },
    {
      "auxiliary_loss_clip": 0.01186078,
      "auxiliary_loss_mlp": 0.01006558,
      "balance_loss_clip": 1.03164327,
      "balance_loss_mlp": 0.99864209,
      "epoch": 0.0519449287560873,
      "flos": 41225989432320.0,
      "grad_norm": 0.9372835664595212,
      "language_loss": 0.63147217,
      "learning_rate": 3.9949772192427e-06,
      "loss": 0.65339851,
      "num_input_tokens_seen": 9353960,
      "step": 432,
      "time_per_iteration": 2.9015109539031982
    },
    {
      "auxiliary_loss_clip": 0.01291746,
      "auxiliary_loss_mlp": 0.01047251,
      "balance_loss_clip": 1.07102907,
      "balance_loss_mlp": 1.03206348,
      "epoch": 0.05206517164672639,
      "flos": 17494530261120.0,
      "grad_norm": 1.886832996937148,
      "language_loss": 0.79546416,
      "learning_rate": 3.994921895647405e-06,
      "loss": 0.81885415,
      "num_input_tokens_seen": 9372130,
      "step": 433,
      "time_per_iteration": 2.6531736850738525
    },
    {
      "auxiliary_loss_clip": 0.01172716,
      "auxiliary_loss_mlp": 0.01007328,
      "balance_loss_clip": 1.03363121,
      "balance_loss_mlp": 0.99969876,
      "epoch": 0.05218541453736548,
      "flos": 64002762973440.0,
      "grad_norm": 0.8377672584044363,
      "language_loss": 0.55355901,
      "learning_rate": 3.994866269424043e-06,
      "loss": 0.57535952,
      "num_input_tokens_seen": 9428500,
      "step": 434,
      "time_per_iteration": 3.081937551498413
    },
    {
      "auxiliary_loss_clip": 0.01283822,
      "auxiliary_loss_mlp": 0.01052649,
      "balance_loss_clip": 1.05801821,
      "balance_loss_mlp": 1.03704464,
      "epoch": 0.05230565742800457,
      "flos": 19317319787520.0,
      "grad_norm": 2.5131455769622835,
      "language_loss": 0.78399247,
      "learning_rate": 3.9948103405810545e-06,
      "loss": 0.80735719,
      "num_input_tokens_seen": 9447450,
      "step": 435,
      "time_per_iteration": 2.6920547485351562
    },
    {
      "auxiliary_loss_clip": 0.01278411,
      "auxiliary_loss_mlp": 0.01053332,
      "balance_loss_clip": 1.06691194,
      "balance_loss_mlp": 1.03936672,
      "epoch": 0.05242590031864366,
      "flos": 25298636538240.0,
      "grad_norm": 3.6836707148949395,
      "language_loss": 0.86081076,
      "learning_rate": 3.994754109126923e-06,
      "loss": 0.88412821,
      "num_input_tokens_seen": 9468945,
      "step": 436,
      "time_per_iteration": 2.7867424488067627
    },
    {
      "auxiliary_loss_clip": 0.01285404,
      "auxiliary_loss_mlp": 0.01043568,
      "balance_loss_clip": 1.0661881,
      "balance_loss_mlp": 1.0292275,
      "epoch": 0.052546143209282754,
      "flos": 26211629456640.0,
      "grad_norm": 1.8300142947565239,
      "language_loss": 0.9329747,
      "learning_rate": 3.994697575070181e-06,
      "loss": 0.9562645,
      "num_input_tokens_seen": 9488405,
      "step": 437,
      "time_per_iteration": 2.8184473514556885
    },
    {
      "auxiliary_loss_clip": 0.01295711,
      "auxiliary_loss_mlp": 0.01051244,
      "balance_loss_clip": 1.07751381,
      "balance_loss_mlp": 1.03617597,
      "epoch": 0.052666386099921844,
      "flos": 22158140578560.0,
      "grad_norm": 1.9218040303457866,
      "language_loss": 0.91307729,
      "learning_rate": 3.994640738419402e-06,
      "loss": 0.93654686,
      "num_input_tokens_seen": 9507780,
      "step": 438,
      "time_per_iteration": 2.68652606010437
    },
    {
      "auxiliary_loss_clip": 0.01291144,
      "auxiliary_loss_mlp": 0.01040597,
      "balance_loss_clip": 1.07515383,
      "balance_loss_mlp": 1.02634525,
      "epoch": 0.052786628990560934,
      "flos": 23881817502720.0,
      "grad_norm": 2.8046221129884983,
      "language_loss": 0.80791306,
      "learning_rate": 3.9945835991832075e-06,
      "loss": 0.8312304,
      "num_input_tokens_seen": 9529665,
      "step": 439,
      "time_per_iteration": 2.731982946395874
    },
    {
      "auxiliary_loss_clip": 0.01294234,
      "auxiliary_loss_mlp": 0.01053957,
      "balance_loss_clip": 1.08032,
      "balance_loss_mlp": 1.03977084,
      "epoch": 0.052906871881200024,
      "flos": 24605021934720.0,
      "grad_norm": 2.4904970058604134,
      "language_loss": 0.93014139,
      "learning_rate": 3.994526157370268e-06,
      "loss": 0.95362329,
      "num_input_tokens_seen": 9548280,
      "step": 440,
      "time_per_iteration": 2.645792245864868
    },
    {
      "auxiliary_loss_clip": 0.01184185,
      "auxiliary_loss_mlp": 0.01016362,
      "balance_loss_clip": 1.03205347,
      "balance_loss_mlp": 1.00911379,
      "epoch": 0.053027114771839114,
      "flos": 56461631143680.0,
      "grad_norm": 0.8969310561111847,
      "language_loss": 0.59322268,
      "learning_rate": 3.994468412989296e-06,
      "loss": 0.61522812,
      "num_input_tokens_seen": 9609690,
      "step": 441,
      "time_per_iteration": 3.2741456031799316
    },
    {
      "auxiliary_loss_clip": 0.01266276,
      "auxiliary_loss_mlp": 0.01049979,
      "balance_loss_clip": 1.06651914,
      "balance_loss_mlp": 1.03531623,
      "epoch": 0.053147357662478203,
      "flos": 17311098481920.0,
      "grad_norm": 6.471648002733174,
      "language_loss": 0.92486823,
      "learning_rate": 3.994410366049052e-06,
      "loss": 0.94803071,
      "num_input_tokens_seen": 9627550,
      "step": 442,
      "time_per_iteration": 2.6475627422332764
    },
    {
      "auxiliary_loss_clip": 0.01291165,
      "auxiliary_loss_mlp": 0.01042795,
      "balance_loss_clip": 1.07329619,
      "balance_loss_mlp": 1.02791142,
      "epoch": 0.0532676005531173,
      "flos": 17164977955200.0,
      "grad_norm": 2.640191446490045,
      "language_loss": 0.83202922,
      "learning_rate": 3.994352016558341e-06,
      "loss": 0.85536885,
      "num_input_tokens_seen": 9644855,
      "step": 443,
      "time_per_iteration": 2.6398837566375732
    },
    {
      "auxiliary_loss_clip": 0.01292837,
      "auxiliary_loss_mlp": 0.01049172,
      "balance_loss_clip": 1.07641721,
      "balance_loss_mlp": 1.03434253,
      "epoch": 0.05338784344375639,
      "flos": 27819960831360.0,
      "grad_norm": 2.82478446391051,
      "language_loss": 0.7397635,
      "learning_rate": 3.994293364526014e-06,
      "loss": 0.76318359,
      "num_input_tokens_seen": 9665740,
      "step": 444,
      "time_per_iteration": 2.7211453914642334
    },
    {
      "auxiliary_loss_clip": 0.01282748,
      "auxiliary_loss_mlp": 0.01045722,
      "balance_loss_clip": 1.07309961,
      "balance_loss_mlp": 1.02946162,
      "epoch": 0.05350808633439548,
      "flos": 21507691144320.0,
      "grad_norm": 2.6286364750795665,
      "language_loss": 0.84943986,
      "learning_rate": 3.99423440996097e-06,
      "loss": 0.87272453,
      "num_input_tokens_seen": 9685280,
      "step": 445,
      "time_per_iteration": 2.708656072616577
    },
    {
      "auxiliary_loss_clip": 0.01293772,
      "auxiliary_loss_mlp": 0.01048594,
      "balance_loss_clip": 1.07793963,
      "balance_loss_mlp": 1.03315663,
      "epoch": 0.05362832922503457,
      "flos": 20084299920000.0,
      "grad_norm": 4.322783591134125,
      "language_loss": 0.81498897,
      "learning_rate": 3.994175152872152e-06,
      "loss": 0.83841264,
      "num_input_tokens_seen": 9704365,
      "step": 446,
      "time_per_iteration": 2.6565968990325928
    },
    {
      "auxiliary_loss_clip": 0.01294028,
      "auxiliary_loss_mlp": 0.01039856,
      "balance_loss_clip": 1.074386,
      "balance_loss_mlp": 1.02590799,
      "epoch": 0.05374857211567366,
      "flos": 26137222433280.0,
      "grad_norm": 2.327226074009384,
      "language_loss": 0.78607363,
      "learning_rate": 3.994115593268548e-06,
      "loss": 0.80941248,
      "num_input_tokens_seen": 9724145,
      "step": 447,
      "time_per_iteration": 2.7142155170440674
    },
    {
      "auxiliary_loss_clip": 0.01294341,
      "auxiliary_loss_mlp": 0.01048941,
      "balance_loss_clip": 1.07856202,
      "balance_loss_mlp": 1.03464746,
      "epoch": 0.05386881500631275,
      "flos": 27486817165440.0,
      "grad_norm": 2.4152555727930696,
      "language_loss": 0.82098925,
      "learning_rate": 3.994055731159195e-06,
      "loss": 0.8444221,
      "num_input_tokens_seen": 9741615,
      "step": 448,
      "time_per_iteration": 2.7099621295928955
    },
    {
      "auxiliary_loss_clip": 0.01296877,
      "auxiliary_loss_mlp": 0.01057116,
      "balance_loss_clip": 1.08046806,
      "balance_loss_mlp": 1.04273939,
      "epoch": 0.053989057896951846,
      "flos": 23585087249280.0,
      "grad_norm": 2.2988526144582044,
      "language_loss": 0.86688584,
      "learning_rate": 3.993995566553172e-06,
      "loss": 0.89042574,
      "num_input_tokens_seen": 9760580,
      "step": 449,
      "time_per_iteration": 3.634645700454712
    },
    {
      "auxiliary_loss_clip": 0.01271082,
      "auxiliary_loss_mlp": 0.01048034,
      "balance_loss_clip": 1.06510067,
      "balance_loss_mlp": 1.03358626,
      "epoch": 0.054109300787590936,
      "flos": 25228862369280.0,
      "grad_norm": 1.710891342817181,
      "language_loss": 0.77046204,
      "learning_rate": 3.993935099459607e-06,
      "loss": 0.79365325,
      "num_input_tokens_seen": 9782195,
      "step": 450,
      "time_per_iteration": 3.639688014984131
    },
    {
      "auxiliary_loss_clip": 0.01287605,
      "auxiliary_loss_mlp": 0.0104809,
      "balance_loss_clip": 1.07815814,
      "balance_loss_mlp": 1.03410697,
      "epoch": 0.054229543678230026,
      "flos": 23841525421440.0,
      "grad_norm": 2.0879962037418136,
      "language_loss": 0.73899329,
      "learning_rate": 3.993874329887673e-06,
      "loss": 0.76235026,
      "num_input_tokens_seen": 9800850,
      "step": 451,
      "time_per_iteration": 3.582491636276245
    },
    {
      "auxiliary_loss_clip": 0.01292172,
      "auxiliary_loss_mlp": 0.01056231,
      "balance_loss_clip": 1.07592797,
      "balance_loss_mlp": 1.04129362,
      "epoch": 0.054349786568869116,
      "flos": 16320933192960.0,
      "grad_norm": 5.659905486063858,
      "language_loss": 0.86209178,
      "learning_rate": 3.993813257846589e-06,
      "loss": 0.88557589,
      "num_input_tokens_seen": 9817605,
      "step": 452,
      "time_per_iteration": 2.6086480617523193
    },
    {
      "auxiliary_loss_clip": 0.0129172,
      "auxiliary_loss_mlp": 0.01049289,
      "balance_loss_clip": 1.07792211,
      "balance_loss_mlp": 1.03462005,
      "epoch": 0.054470029459508205,
      "flos": 18660729127680.0,
      "grad_norm": 3.1381239277265833,
      "language_loss": 0.92889744,
      "learning_rate": 3.993751883345619e-06,
      "loss": 0.95230752,
      "num_input_tokens_seen": 9835965,
      "step": 453,
      "time_per_iteration": 3.534494400024414
    },
    {
      "auxiliary_loss_clip": 0.01287878,
      "auxiliary_loss_mlp": 0.01050277,
      "balance_loss_clip": 1.07592773,
      "balance_loss_mlp": 1.0354892,
      "epoch": 0.054590272350147295,
      "flos": 17785298856960.0,
      "grad_norm": 2.3356834925018175,
      "language_loss": 0.87287891,
      "learning_rate": 3.993690206394073e-06,
      "loss": 0.8962605,
      "num_input_tokens_seen": 9852265,
      "step": 454,
      "time_per_iteration": 2.644972562789917
    },
    {
      "auxiliary_loss_clip": 0.01295309,
      "auxiliary_loss_mlp": 0.01049553,
      "balance_loss_clip": 1.07488012,
      "balance_loss_mlp": 1.03530729,
      "epoch": 0.054710515240786385,
      "flos": 17785945301760.0,
      "grad_norm": 4.869404709266602,
      "language_loss": 0.87715602,
      "learning_rate": 3.993628227001307e-06,
      "loss": 0.90060467,
      "num_input_tokens_seen": 9870465,
      "step": 455,
      "time_per_iteration": 2.684591770172119
    },
    {
      "auxiliary_loss_clip": 0.01289712,
      "auxiliary_loss_mlp": 0.01051104,
      "balance_loss_clip": 1.07364726,
      "balance_loss_mlp": 1.03708482,
      "epoch": 0.05483075813142548,
      "flos": 48210900180480.0,
      "grad_norm": 2.1035392062180818,
      "language_loss": 0.71278024,
      "learning_rate": 3.993565945176726e-06,
      "loss": 0.73618841,
      "num_input_tokens_seen": 9891490,
      "step": 456,
      "time_per_iteration": 2.9364194869995117
    },
    {
      "auxiliary_loss_clip": 0.01283558,
      "auxiliary_loss_mlp": 0.01052957,
      "balance_loss_clip": 1.07468987,
      "balance_loss_mlp": 1.03847909,
      "epoch": 0.05495100102206457,
      "flos": 19682244011520.0,
      "grad_norm": 2.1437363248489856,
      "language_loss": 0.84249783,
      "learning_rate": 3.993503360929776e-06,
      "loss": 0.86586297,
      "num_input_tokens_seen": 9910375,
      "step": 457,
      "time_per_iteration": 2.659064769744873
    },
    {
      "auxiliary_loss_clip": 0.01287304,
      "auxiliary_loss_mlp": 0.01047931,
      "balance_loss_clip": 1.06288528,
      "balance_loss_mlp": 1.03248167,
      "epoch": 0.05507124391270366,
      "flos": 26360048453760.0,
      "grad_norm": 2.197897542459757,
      "language_loss": 0.81030202,
      "learning_rate": 3.99344047426995e-06,
      "loss": 0.8336544,
      "num_input_tokens_seen": 9931635,
      "step": 458,
      "time_per_iteration": 2.8105955123901367
    },
    {
      "auxiliary_loss_clip": 0.01293377,
      "auxiliary_loss_mlp": 0.01051279,
      "balance_loss_clip": 1.06962204,
      "balance_loss_mlp": 1.03612781,
      "epoch": 0.05519148680334275,
      "flos": 22601314581120.0,
      "grad_norm": 2.3758577748474323,
      "language_loss": 0.93362761,
      "learning_rate": 3.993377285206789e-06,
      "loss": 0.95707417,
      "num_input_tokens_seen": 9951420,
      "step": 459,
      "time_per_iteration": 2.71272873878479
    },
    {
      "auxiliary_loss_clip": 0.01266226,
      "auxiliary_loss_mlp": 0.01056186,
      "balance_loss_clip": 1.06640053,
      "balance_loss_mlp": 1.04166627,
      "epoch": 0.05531172969398184,
      "flos": 40552519380480.0,
      "grad_norm": 1.710013152950042,
      "language_loss": 0.86426473,
      "learning_rate": 3.99331379374988e-06,
      "loss": 0.88748878,
      "num_input_tokens_seen": 9975025,
      "step": 460,
      "time_per_iteration": 2.865351438522339
    },
    {
      "auxiliary_loss_clip": 0.01293659,
      "auxiliary_loss_mlp": 0.01046151,
      "balance_loss_clip": 1.06913579,
      "balance_loss_mlp": 1.03231084,
      "epoch": 0.05543197258462093,
      "flos": 23477894087040.0,
      "grad_norm": 3.853689733749625,
      "language_loss": 0.79918265,
      "learning_rate": 3.993249999908852e-06,
      "loss": 0.82258081,
      "num_input_tokens_seen": 9995175,
      "step": 461,
      "time_per_iteration": 2.6865110397338867
    },
    {
      "auxiliary_loss_clip": 0.01288624,
      "auxiliary_loss_mlp": 0.01048857,
      "balance_loss_clip": 1.07511592,
      "balance_loss_mlp": 1.03481412,
      "epoch": 0.05555221547526003,
      "flos": 18624603024000.0,
      "grad_norm": 3.5142501150991303,
      "language_loss": 0.87245369,
      "learning_rate": 3.993185903693384e-06,
      "loss": 0.89582849,
      "num_input_tokens_seen": 10011975,
      "step": 462,
      "time_per_iteration": 2.610358238220215
    },
    {
      "auxiliary_loss_clip": 0.01285444,
      "auxiliary_loss_mlp": 0.01040871,
      "balance_loss_clip": 1.07130361,
      "balance_loss_mlp": 1.02702522,
      "epoch": 0.05567245836589912,
      "flos": 23587098410880.0,
      "grad_norm": 2.3819065474749137,
      "language_loss": 0.82106227,
      "learning_rate": 3.9931215051131995e-06,
      "loss": 0.84432542,
      "num_input_tokens_seen": 10032620,
      "step": 463,
      "time_per_iteration": 2.694817304611206
    },
    {
      "auxiliary_loss_clip": 0.01289316,
      "auxiliary_loss_mlp": 0.01043869,
      "balance_loss_clip": 1.06783962,
      "balance_loss_mlp": 1.02985048,
      "epoch": 0.05579270125653821,
      "flos": 27746667129600.0,
      "grad_norm": 2.563450844953798,
      "language_loss": 0.80158097,
      "learning_rate": 3.993056804178068e-06,
      "loss": 0.82491285,
      "num_input_tokens_seen": 10054165,
      "step": 464,
      "time_per_iteration": 2.7699904441833496
    },
    {
      "auxiliary_loss_clip": 0.01282835,
      "auxiliary_loss_mlp": 0.01045453,
      "balance_loss_clip": 1.06802309,
      "balance_loss_mlp": 1.03043282,
      "epoch": 0.0559129441471773,
      "flos": 27014161075200.0,
      "grad_norm": 3.3123033195032083,
      "language_loss": 0.84644198,
      "learning_rate": 3.992991800897803e-06,
      "loss": 0.86972487,
      "num_input_tokens_seen": 10073970,
      "step": 465,
      "time_per_iteration": 2.7533082962036133
    },
    {
      "auxiliary_loss_clip": 0.01285566,
      "auxiliary_loss_mlp": 0.01046075,
      "balance_loss_clip": 1.07460594,
      "balance_loss_mlp": 1.0311439,
      "epoch": 0.05603318703781639,
      "flos": 15229787794560.0,
      "grad_norm": 2.255226525827696,
      "language_loss": 0.8952812,
      "learning_rate": 3.9929264952822665e-06,
      "loss": 0.91859758,
      "num_input_tokens_seen": 10091505,
      "step": 466,
      "time_per_iteration": 2.6055283546447754
    },
    {
      "auxiliary_loss_clip": 0.0128849,
      "auxiliary_loss_mlp": 0.01049986,
      "balance_loss_clip": 1.07112348,
      "balance_loss_mlp": 1.03590703,
      "epoch": 0.05615342992845548,
      "flos": 22266482976000.0,
      "grad_norm": 1.9997691744298527,
      "language_loss": 0.88090789,
      "learning_rate": 3.992860887341366e-06,
      "loss": 0.90429264,
      "num_input_tokens_seen": 10109675,
      "step": 467,
      "time_per_iteration": 2.701089859008789
    },
    {
      "auxiliary_loss_clip": 0.01275228,
      "auxiliary_loss_mlp": 0.01042757,
      "balance_loss_clip": 1.06712651,
      "balance_loss_mlp": 1.02774858,
      "epoch": 0.056273672819094574,
      "flos": 23584979508480.0,
      "grad_norm": 2.0544359632801137,
      "language_loss": 0.81260824,
      "learning_rate": 3.992794977085052e-06,
      "loss": 0.83578807,
      "num_input_tokens_seen": 10127675,
      "step": 468,
      "time_per_iteration": 2.7003376483917236
    },
    {
      "auxiliary_loss_clip": 0.01289685,
      "auxiliary_loss_mlp": 0.01051903,
      "balance_loss_clip": 1.07136822,
      "balance_loss_mlp": 1.03793192,
      "epoch": 0.056393915709733664,
      "flos": 19858708552320.0,
      "grad_norm": 1.897440272879136,
      "language_loss": 0.84832376,
      "learning_rate": 3.992728764523326e-06,
      "loss": 0.87173969,
      "num_input_tokens_seen": 10146620,
      "step": 469,
      "time_per_iteration": 2.7281012535095215
    },
    {
      "auxiliary_loss_clip": 0.01285429,
      "auxiliary_loss_mlp": 0.01045921,
      "balance_loss_clip": 1.07037842,
      "balance_loss_mlp": 1.03168702,
      "epoch": 0.05651415860037275,
      "flos": 22163779013760.0,
      "grad_norm": 4.832790688479102,
      "language_loss": 0.80570197,
      "learning_rate": 3.99266224966623e-06,
      "loss": 0.82901549,
      "num_input_tokens_seen": 10167535,
      "step": 470,
      "time_per_iteration": 2.659118890762329
    },
    {
      "auxiliary_loss_clip": 0.01275955,
      "auxiliary_loss_mlp": 0.010429,
      "balance_loss_clip": 1.07066631,
      "balance_loss_mlp": 1.02804637,
      "epoch": 0.05663440149101184,
      "flos": 19463548055040.0,
      "grad_norm": 1.9763309107789375,
      "language_loss": 0.88163799,
      "learning_rate": 3.992595432523855e-06,
      "loss": 0.90482652,
      "num_input_tokens_seen": 10184825,
      "step": 471,
      "time_per_iteration": 2.7068026065826416
    },
    {
      "auxiliary_loss_clip": 0.01276028,
      "auxiliary_loss_mlp": 0.0104956,
      "balance_loss_clip": 1.06726217,
      "balance_loss_mlp": 1.035761,
      "epoch": 0.05675464438165093,
      "flos": 22670226823680.0,
      "grad_norm": 2.002193977415129,
      "language_loss": 0.85899615,
      "learning_rate": 3.992528313106338e-06,
      "loss": 0.88225204,
      "num_input_tokens_seen": 10203025,
      "step": 472,
      "time_per_iteration": 2.668438196182251
    },
    {
      "auxiliary_loss_clip": 0.01284781,
      "auxiliary_loss_mlp": 0.01209859,
      "balance_loss_clip": 1.07712221,
      "balance_loss_mlp": 1.00016356,
      "epoch": 0.05687488727229002,
      "flos": 16901177495040.0,
      "grad_norm": 4.545679704460693,
      "language_loss": 0.82020223,
      "learning_rate": 3.9924608914238595e-06,
      "loss": 0.84514862,
      "num_input_tokens_seen": 10218020,
      "step": 473,
      "time_per_iteration": 2.6507408618927
    },
    {
      "auxiliary_loss_clip": 0.01287131,
      "auxiliary_loss_mlp": 0.01049167,
      "balance_loss_clip": 1.0757072,
      "balance_loss_mlp": 1.03529668,
      "epoch": 0.05699513016292912,
      "flos": 29168980945920.0,
      "grad_norm": 2.337327353959266,
      "language_loss": 0.83842564,
      "learning_rate": 3.992393167486648e-06,
      "loss": 0.86178863,
      "num_input_tokens_seen": 10237170,
      "step": 474,
      "time_per_iteration": 2.740530014038086
    },
    {
      "auxiliary_loss_clip": 0.01287049,
      "auxiliary_loss_mlp": 0.01049786,
      "balance_loss_clip": 1.07578301,
      "balance_loss_mlp": 1.03487885,
      "epoch": 0.05711537305356821,
      "flos": 18916197632640.0,
      "grad_norm": 2.4646879528355274,
      "language_loss": 0.81018162,
      "learning_rate": 3.992325141304977e-06,
      "loss": 0.83354998,
      "num_input_tokens_seen": 10255125,
      "step": 475,
      "time_per_iteration": 3.464486598968506
    },
    {
      "auxiliary_loss_clip": 0.01273181,
      "auxiliary_loss_mlp": 0.01046895,
      "balance_loss_clip": 1.06748295,
      "balance_loss_mlp": 1.03326344,
      "epoch": 0.0572356159442073,
      "flos": 26758979879040.0,
      "grad_norm": 23.77672951271691,
      "language_loss": 0.8663106,
      "learning_rate": 3.992256812889166e-06,
      "loss": 0.88951135,
      "num_input_tokens_seen": 10271230,
      "step": 476,
      "time_per_iteration": 3.636479377746582
    },
    {
      "auxiliary_loss_clip": 0.01287402,
      "auxiliary_loss_mlp": 0.01046554,
      "balance_loss_clip": 1.07850957,
      "balance_loss_mlp": 1.03265452,
      "epoch": 0.05735585883484639,
      "flos": 35116146840960.0,
      "grad_norm": 5.6566058157687165,
      "language_loss": 0.76672518,
      "learning_rate": 3.992188182249582e-06,
      "loss": 0.79006469,
      "num_input_tokens_seen": 10293125,
      "step": 477,
      "time_per_iteration": 2.688354969024658
    },
    {
      "auxiliary_loss_clip": 0.01285007,
      "auxiliary_loss_mlp": 0.01051757,
      "balance_loss_clip": 1.07489967,
      "balance_loss_mlp": 1.0373745,
      "epoch": 0.05747610172548548,
      "flos": 18734381965440.0,
      "grad_norm": 2.214991752075064,
      "language_loss": 0.90724289,
      "learning_rate": 3.992119249396633e-06,
      "loss": 0.93061054,
      "num_input_tokens_seen": 10311810,
      "step": 478,
      "time_per_iteration": 3.6554410457611084
    },
    {
      "auxiliary_loss_clip": 0.01275458,
      "auxiliary_loss_mlp": 0.01209492,
      "balance_loss_clip": 1.0672754,
      "balance_loss_mlp": 1.00022876,
      "epoch": 0.05759634461612457,
      "flos": 27964752554880.0,
      "grad_norm": 1.9616390069806409,
      "language_loss": 0.81824249,
      "learning_rate": 3.992050014340778e-06,
      "loss": 0.84309202,
      "num_input_tokens_seen": 10332165,
      "step": 479,
      "time_per_iteration": 3.7525362968444824
    },
    {
      "auxiliary_loss_clip": 0.01174201,
      "auxiliary_loss_mlp": 0.01022324,
      "balance_loss_clip": 1.04071939,
      "balance_loss_mlp": 1.01564837,
      "epoch": 0.057716587506763666,
      "flos": 69292009405440.0,
      "grad_norm": 0.840082529476001,
      "language_loss": 0.55019456,
      "learning_rate": 3.99198047709252e-06,
      "loss": 0.57215983,
      "num_input_tokens_seen": 10393685,
      "step": 480,
      "time_per_iteration": 3.2906527519226074
    },
    {
      "auxiliary_loss_clip": 0.01283726,
      "auxiliary_loss_mlp": 0.01050298,
      "balance_loss_clip": 1.06522846,
      "balance_loss_mlp": 1.03582025,
      "epoch": 0.057836830397402755,
      "flos": 25009196745600.0,
      "grad_norm": 1.9266645235825424,
      "language_loss": 0.78457677,
      "learning_rate": 3.991910637662408e-06,
      "loss": 0.80791706,
      "num_input_tokens_seen": 10413975,
      "step": 481,
      "time_per_iteration": 2.7314536571502686
    },
    {
      "auxiliary_loss_clip": 0.01285935,
      "auxiliary_loss_mlp": 0.01039336,
      "balance_loss_clip": 1.07786644,
      "balance_loss_mlp": 1.02484024,
      "epoch": 0.057957073288041845,
      "flos": 25593894334080.0,
      "grad_norm": 2.0953187778479463,
      "language_loss": 0.80809689,
      "learning_rate": 3.9918404960610355e-06,
      "loss": 0.83134961,
      "num_input_tokens_seen": 10433005,
      "step": 482,
      "time_per_iteration": 2.6843056678771973
    },
    {
      "auxiliary_loss_clip": 0.01293245,
      "auxiliary_loss_mlp": 0.01048306,
      "balance_loss_clip": 1.07670021,
      "balance_loss_mlp": 1.03439438,
      "epoch": 0.058077316178680935,
      "flos": 20777411733120.0,
      "grad_norm": 2.303663797946004,
      "language_loss": 0.7713536,
      "learning_rate": 3.991770052299043e-06,
      "loss": 0.79476905,
      "num_input_tokens_seen": 10451235,
      "step": 483,
      "time_per_iteration": 2.6419639587402344
    },
    {
      "auxiliary_loss_clip": 0.01283424,
      "auxiliary_loss_mlp": 0.01042534,
      "balance_loss_clip": 1.06831253,
      "balance_loss_mlp": 1.02965987,
      "epoch": 0.058197559069320025,
      "flos": 18916484941440.0,
      "grad_norm": 4.0414332720832835,
      "language_loss": 0.87792838,
      "learning_rate": 3.991699306387118e-06,
      "loss": 0.9011879,
      "num_input_tokens_seen": 10469705,
      "step": 484,
      "time_per_iteration": 2.6688241958618164
    },
    {
      "auxiliary_loss_clip": 0.0128649,
      "auxiliary_loss_mlp": 0.0105237,
      "balance_loss_clip": 1.07357252,
      "balance_loss_mlp": 1.03853631,
      "epoch": 0.058317801959959115,
      "flos": 24863327614080.0,
      "grad_norm": 1.9740908006174829,
      "language_loss": 0.78039932,
      "learning_rate": 3.991628258335991e-06,
      "loss": 0.80378795,
      "num_input_tokens_seen": 10491910,
      "step": 485,
      "time_per_iteration": 2.6673574447631836
    },
    {
      "auxiliary_loss_clip": 0.01279153,
      "auxiliary_loss_mlp": 0.01046242,
      "balance_loss_clip": 1.06762981,
      "balance_loss_mlp": 1.03222895,
      "epoch": 0.05843804485059821,
      "flos": 23257977068160.0,
      "grad_norm": 3.867885012504226,
      "language_loss": 0.87775415,
      "learning_rate": 3.991556908156442e-06,
      "loss": 0.90100807,
      "num_input_tokens_seen": 10508435,
      "step": 486,
      "time_per_iteration": 2.7164382934570312
    },
    {
      "auxiliary_loss_clip": 0.01291202,
      "auxiliary_loss_mlp": 0.01049353,
      "balance_loss_clip": 1.07178092,
      "balance_loss_mlp": 1.0359478,
      "epoch": 0.0585582877412373,
      "flos": 23150532510720.0,
      "grad_norm": 2.2254090913310915,
      "language_loss": 0.87591302,
      "learning_rate": 3.9914852558592914e-06,
      "loss": 0.89931852,
      "num_input_tokens_seen": 10529485,
      "step": 487,
      "time_per_iteration": 2.6487948894500732
    },
    {
      "auxiliary_loss_clip": 0.01286308,
      "auxiliary_loss_mlp": 0.01048182,
      "balance_loss_clip": 1.07641006,
      "balance_loss_mlp": 1.03380537,
      "epoch": 0.05867853063187639,
      "flos": 23506406507520.0,
      "grad_norm": 3.002531102810409,
      "language_loss": 0.81157351,
      "learning_rate": 3.991413301455413e-06,
      "loss": 0.83491838,
      "num_input_tokens_seen": 10545935,
      "step": 488,
      "time_per_iteration": 2.6419272422790527
    },
    {
      "auxiliary_loss_clip": 0.01271918,
      "auxiliary_loss_mlp": 0.01046779,
      "balance_loss_clip": 1.06930912,
      "balance_loss_mlp": 1.03392875,
      "epoch": 0.05879877352251548,
      "flos": 29495803818240.0,
      "grad_norm": 2.7218717809480157,
      "language_loss": 0.77509952,
      "learning_rate": 3.991341044955719e-06,
      "loss": 0.79828656,
      "num_input_tokens_seen": 10565690,
      "step": 489,
      "time_per_iteration": 2.743152379989624
    },
    {
      "auxiliary_loss_clip": 0.01283263,
      "auxiliary_loss_mlp": 0.01210348,
      "balance_loss_clip": 1.07267404,
      "balance_loss_mlp": 1.00035799,
      "epoch": 0.05891901641315457,
      "flos": 20157485880960.0,
      "grad_norm": 2.2213850745614296,
      "language_loss": 0.81037033,
      "learning_rate": 3.991268486371172e-06,
      "loss": 0.83530641,
      "num_input_tokens_seen": 10584245,
      "step": 490,
      "time_per_iteration": 2.732724905014038
    },
    {
      "auxiliary_loss_clip": 0.01286072,
      "auxiliary_loss_mlp": 0.01049956,
      "balance_loss_clip": 1.07028139,
      "balance_loss_mlp": 1.03368402,
      "epoch": 0.05903925930379366,
      "flos": 24644200694400.0,
      "grad_norm": 2.7474239857237754,
      "language_loss": 0.88019943,
      "learning_rate": 3.991195625712779e-06,
      "loss": 0.90355968,
      "num_input_tokens_seen": 10601210,
      "step": 491,
      "time_per_iteration": 2.687427282333374
    },
    {
      "auxiliary_loss_clip": 0.01285278,
      "auxiliary_loss_mlp": 0.01046501,
      "balance_loss_clip": 1.07849121,
      "balance_loss_mlp": 1.03208923,
      "epoch": 0.05915950219443276,
      "flos": 21250391045760.0,
      "grad_norm": 2.181236254874082,
      "language_loss": 0.8178122,
      "learning_rate": 3.991122462991592e-06,
      "loss": 0.84112996,
      "num_input_tokens_seen": 10620730,
      "step": 492,
      "time_per_iteration": 2.7106854915618896
    },
    {
      "auxiliary_loss_clip": 0.01290915,
      "auxiliary_loss_mlp": 0.01048634,
      "balance_loss_clip": 1.07727444,
      "balance_loss_mlp": 1.03497887,
      "epoch": 0.05927974508507185,
      "flos": 9902727319680.0,
      "grad_norm": 6.611485903963053,
      "language_loss": 0.80987179,
      "learning_rate": 3.991048998218712e-06,
      "loss": 0.83326727,
      "num_input_tokens_seen": 10634035,
      "step": 493,
      "time_per_iteration": 2.5994460582733154
    },
    {
      "auxiliary_loss_clip": 0.01283225,
      "auxiliary_loss_mlp": 0.01047053,
      "balance_loss_clip": 1.07121658,
      "balance_loss_mlp": 1.03317678,
      "epoch": 0.05939998797571094,
      "flos": 18259499232000.0,
      "grad_norm": 2.935535134544963,
      "language_loss": 0.7636888,
      "learning_rate": 3.990975231405281e-06,
      "loss": 0.7869916,
      "num_input_tokens_seen": 10652485,
      "step": 494,
      "time_per_iteration": 2.750419855117798
    },
    {
      "auxiliary_loss_clip": 0.01281787,
      "auxiliary_loss_mlp": 0.01047046,
      "balance_loss_clip": 1.07429767,
      "balance_loss_mlp": 1.03302753,
      "epoch": 0.05952023086635003,
      "flos": 28256598558720.0,
      "grad_norm": 2.0100217660466546,
      "language_loss": 0.78949273,
      "learning_rate": 3.990901162562491e-06,
      "loss": 0.81278104,
      "num_input_tokens_seen": 10673175,
      "step": 495,
      "time_per_iteration": 2.7372043132781982
    },
    {
      "auxiliary_loss_clip": 0.0127862,
      "auxiliary_loss_mlp": 0.01210806,
      "balance_loss_clip": 1.06447494,
      "balance_loss_mlp": 1.00037277,
      "epoch": 0.05964047375698912,
      "flos": 14902498045440.0,
      "grad_norm": 2.276387346753233,
      "language_loss": 0.90382689,
      "learning_rate": 3.9908267917015765e-06,
      "loss": 0.92872119,
      "num_input_tokens_seen": 10691235,
      "step": 496,
      "time_per_iteration": 2.724705219268799
    },
    {
      "auxiliary_loss_clip": 0.01271898,
      "auxiliary_loss_mlp": 0.01060633,
      "balance_loss_clip": 1.06882155,
      "balance_loss_mlp": 1.04616094,
      "epoch": 0.059760716647628206,
      "flos": 23185581206400.0,
      "grad_norm": 2.1423699286804907,
      "language_loss": 0.92846179,
      "learning_rate": 3.990752118833821e-06,
      "loss": 0.95178711,
      "num_input_tokens_seen": 10708675,
      "step": 497,
      "time_per_iteration": 2.6863818168640137
    },
    {
      "auxiliary_loss_clip": 0.01285489,
      "auxiliary_loss_mlp": 0.01043095,
      "balance_loss_clip": 1.07706225,
      "balance_loss_mlp": 1.02909398,
      "epoch": 0.0598809595382673,
      "flos": 22746968231040.0,
      "grad_norm": 1.8013169331515246,
      "language_loss": 0.78104782,
      "learning_rate": 3.990677143970553e-06,
      "loss": 0.80433369,
      "num_input_tokens_seen": 10729485,
      "step": 498,
      "time_per_iteration": 2.778846263885498
    },
    {
      "auxiliary_loss_clip": 0.01282126,
      "auxiliary_loss_mlp": 0.01053826,
      "balance_loss_clip": 1.07250357,
      "balance_loss_mlp": 1.03892529,
      "epoch": 0.06000120242890639,
      "flos": 22127221946880.0,
      "grad_norm": 3.166230454176542,
      "language_loss": 0.80941015,
      "learning_rate": 3.990601867123144e-06,
      "loss": 0.83276975,
      "num_input_tokens_seen": 10749210,
      "step": 499,
      "time_per_iteration": 2.7495970726013184
    },
    {
      "auxiliary_loss_clip": 0.0128672,
      "auxiliary_loss_mlp": 0.01050818,
      "balance_loss_clip": 1.07092035,
      "balance_loss_mlp": 1.03653646,
      "epoch": 0.06012144531954548,
      "flos": 19171773878400.0,
      "grad_norm": 15.126176546584718,
      "language_loss": 0.84570616,
      "learning_rate": 3.990526288303014e-06,
      "loss": 0.8690815,
      "num_input_tokens_seen": 10768000,
      "step": 500,
      "time_per_iteration": 2.6875784397125244
    },
    {
      "auxiliary_loss_clip": 0.01279492,
      "auxiliary_loss_mlp": 0.01209548,
      "balance_loss_clip": 1.06916952,
      "balance_loss_mlp": 1.00041032,
      "epoch": 0.06024168821018457,
      "flos": 22783345729920.0,
      "grad_norm": 1.8059979629672136,
      "language_loss": 0.90872794,
      "learning_rate": 3.9904504075216295e-06,
      "loss": 0.93361837,
      "num_input_tokens_seen": 10788760,
      "step": 501,
      "time_per_iteration": 3.78204345703125
    },
    {
      "auxiliary_loss_clip": 0.01283013,
      "auxiliary_loss_mlp": 0.01054808,
      "balance_loss_clip": 1.06630993,
      "balance_loss_mlp": 1.04013324,
      "epoch": 0.06036193110082366,
      "flos": 18770687637120.0,
      "grad_norm": 2.2239216487083824,
      "language_loss": 0.93967164,
      "learning_rate": 3.990374224790501e-06,
      "loss": 0.96304989,
      "num_input_tokens_seen": 10806965,
      "step": 502,
      "time_per_iteration": 2.6979153156280518
    },
    {
      "auxiliary_loss_clip": 0.0128341,
      "auxiliary_loss_mlp": 0.01053087,
      "balance_loss_clip": 1.0732162,
      "balance_loss_mlp": 1.03874063,
      "epoch": 0.06048217399146275,
      "flos": 17201570935680.0,
      "grad_norm": 4.87062173533515,
      "language_loss": 0.70877826,
      "learning_rate": 3.990297740121185e-06,
      "loss": 0.73214322,
      "num_input_tokens_seen": 10824900,
      "step": 503,
      "time_per_iteration": 3.6366000175476074
    },
    {
      "auxiliary_loss_clip": 0.01281171,
      "auxiliary_loss_mlp": 0.01210464,
      "balance_loss_clip": 1.07238412,
      "balance_loss_mlp": 1.00038028,
      "epoch": 0.06060241688210185,
      "flos": 24024131187840.0,
      "grad_norm": 2.4827937415172605,
      "language_loss": 0.78216606,
      "learning_rate": 3.990220953525284e-06,
      "loss": 0.80708241,
      "num_input_tokens_seen": 10842010,
      "step": 504,
      "time_per_iteration": 2.7147610187530518
    },
    {
      "auxiliary_loss_clip": 0.01270662,
      "auxiliary_loss_mlp": 0.01046092,
      "balance_loss_clip": 1.06603098,
      "balance_loss_mlp": 1.0325917,
      "epoch": 0.06072265977274094,
      "flos": 14611190745600.0,
      "grad_norm": 2.5397972729944835,
      "language_loss": 0.74404424,
      "learning_rate": 3.9901438650144465e-06,
      "loss": 0.76721179,
      "num_input_tokens_seen": 10858260,
      "step": 505,
      "time_per_iteration": 3.580364942550659
    },
    {
      "auxiliary_loss_clip": 0.01272933,
      "auxiliary_loss_mlp": 0.01046453,
      "balance_loss_clip": 1.06955433,
      "balance_loss_mlp": 1.03334045,
      "epoch": 0.06084290266338003,
      "flos": 20558284813440.0,
      "grad_norm": 2.9281954238752803,
      "language_loss": 0.92131323,
      "learning_rate": 3.990066474600367e-06,
      "loss": 0.94450712,
      "num_input_tokens_seen": 10876230,
      "step": 506,
      "time_per_iteration": 3.5681965351104736
    },
    {
      "auxiliary_loss_clip": 0.01267495,
      "auxiliary_loss_mlp": 0.01050518,
      "balance_loss_clip": 1.06591654,
      "balance_loss_mlp": 1.03608131,
      "epoch": 0.06096314555401912,
      "flos": 22309217182080.0,
      "grad_norm": 2.0221270398652345,
      "language_loss": 0.67947298,
      "learning_rate": 3.989988782294786e-06,
      "loss": 0.70265305,
      "num_input_tokens_seen": 10896320,
      "step": 507,
      "time_per_iteration": 2.8146727085113525
    },
    {
      "auxiliary_loss_clip": 0.01266559,
      "auxiliary_loss_mlp": 0.01053655,
      "balance_loss_clip": 1.06499505,
      "balance_loss_mlp": 1.03969526,
      "epoch": 0.06108338844465821,
      "flos": 19131374056320.0,
      "grad_norm": 1.9818377527974085,
      "language_loss": 0.94951737,
      "learning_rate": 3.989910788109489e-06,
      "loss": 0.97271949,
      "num_input_tokens_seen": 10912970,
      "step": 508,
      "time_per_iteration": 2.7564845085144043
    },
    {
      "auxiliary_loss_clip": 0.01277699,
      "auxiliary_loss_mlp": 0.01046542,
      "balance_loss_clip": 1.06520545,
      "balance_loss_mlp": 1.0330112,
      "epoch": 0.0612036313352973,
      "flos": 33584018169600.0,
      "grad_norm": 2.442663502187586,
      "language_loss": 0.74402696,
      "learning_rate": 3.989832492056307e-06,
      "loss": 0.76726937,
      "num_input_tokens_seen": 10933995,
      "step": 509,
      "time_per_iteration": 2.853395462036133
    },
    {
      "auxiliary_loss_clip": 0.01282899,
      "auxiliary_loss_mlp": 0.01049773,
      "balance_loss_clip": 1.07348466,
      "balance_loss_mlp": 1.03538406,
      "epoch": 0.06132387422593639,
      "flos": 27490552179840.0,
      "grad_norm": 6.768242431508346,
      "language_loss": 0.80673337,
      "learning_rate": 3.989753894147119e-06,
      "loss": 0.83006012,
      "num_input_tokens_seen": 10954120,
      "step": 510,
      "time_per_iteration": 2.7394838333129883
    },
    {
      "auxiliary_loss_clip": 0.0127752,
      "auxiliary_loss_mlp": 0.01045889,
      "balance_loss_clip": 1.07503486,
      "balance_loss_mlp": 1.03208458,
      "epoch": 0.061444117116575485,
      "flos": 25885057979520.0,
      "grad_norm": 2.3932496529851566,
      "language_loss": 0.8028329,
      "learning_rate": 3.989674994393846e-06,
      "loss": 0.82606697,
      "num_input_tokens_seen": 10973595,
      "step": 511,
      "time_per_iteration": 2.7229206562042236
    },
    {
      "auxiliary_loss_clip": 0.01279003,
      "auxiliary_loss_mlp": 0.01041435,
      "balance_loss_clip": 1.07389224,
      "balance_loss_mlp": 1.02763104,
      "epoch": 0.061564360007214575,
      "flos": 28512031150080.0,
      "grad_norm": 2.1079283084096465,
      "language_loss": 0.94072658,
      "learning_rate": 3.98959579280846e-06,
      "loss": 0.96393096,
      "num_input_tokens_seen": 10991995,
      "step": 512,
      "time_per_iteration": 2.688432455062866
    },
    {
      "auxiliary_loss_clip": 0.01264612,
      "auxiliary_loss_mlp": 0.0104427,
      "balance_loss_clip": 1.06737435,
      "balance_loss_mlp": 1.03009629,
      "epoch": 0.061684602897853665,
      "flos": 12094355652480.0,
      "grad_norm": 3.914787309230438,
      "language_loss": 0.83312386,
      "learning_rate": 3.989516289402973e-06,
      "loss": 0.85621274,
      "num_input_tokens_seen": 11007625,
      "step": 513,
      "time_per_iteration": 2.6112117767333984
    },
    {
      "auxiliary_loss_clip": 0.01260221,
      "auxiliary_loss_mlp": 0.01044955,
      "balance_loss_clip": 1.05611539,
      "balance_loss_mlp": 1.03075743,
      "epoch": 0.061804845788492754,
      "flos": 19532639865600.0,
      "grad_norm": 2.460807750748541,
      "language_loss": 0.80672908,
      "learning_rate": 3.989436484189447e-06,
      "loss": 0.82978088,
      "num_input_tokens_seen": 11025570,
      "step": 514,
      "time_per_iteration": 2.7478291988372803
    },
    {
      "auxiliary_loss_clip": 0.01281982,
      "auxiliary_loss_mlp": 0.01043923,
      "balance_loss_clip": 1.06982899,
      "balance_loss_mlp": 1.03011882,
      "epoch": 0.061925088679131844,
      "flos": 15341111020800.0,
      "grad_norm": 3.1714322340619456,
      "language_loss": 0.81169224,
      "learning_rate": 3.9893563771799885e-06,
      "loss": 0.83495128,
      "num_input_tokens_seen": 11042045,
      "step": 515,
      "time_per_iteration": 2.644981622695923
    },
    {
      "auxiliary_loss_clip": 0.01279649,
      "auxiliary_loss_mlp": 0.01050174,
      "balance_loss_clip": 1.07340765,
      "balance_loss_mlp": 1.03617895,
      "epoch": 0.062045331569770934,
      "flos": 25919927107200.0,
      "grad_norm": 3.1159318952660673,
      "language_loss": 0.86078805,
      "learning_rate": 3.989275968386749e-06,
      "loss": 0.88408625,
      "num_input_tokens_seen": 11059955,
      "step": 516,
      "time_per_iteration": 2.712254285812378
    },
    {
      "auxiliary_loss_clip": 0.01274734,
      "auxiliary_loss_mlp": 0.01051338,
      "balance_loss_clip": 1.06761408,
      "balance_loss_mlp": 1.03669918,
      "epoch": 0.06216557446041003,
      "flos": 28110621686400.0,
      "grad_norm": 2.594531760043738,
      "language_loss": 0.76642895,
      "learning_rate": 3.989195257821926e-06,
      "loss": 0.78968966,
      "num_input_tokens_seen": 11078440,
      "step": 517,
      "time_per_iteration": 2.709066390991211
    },
    {
      "auxiliary_loss_clip": 0.0127785,
      "auxiliary_loss_mlp": 0.01050209,
      "balance_loss_clip": 1.07156205,
      "balance_loss_mlp": 1.03579688,
      "epoch": 0.06228581735104912,
      "flos": 23478181395840.0,
      "grad_norm": 3.2392293026680132,
      "language_loss": 0.84429348,
      "learning_rate": 3.989114245497765e-06,
      "loss": 0.8675741,
      "num_input_tokens_seen": 11098240,
      "step": 518,
      "time_per_iteration": 2.7683136463165283
    },
    {
      "auxiliary_loss_clip": 0.01279524,
      "auxiliary_loss_mlp": 0.01042694,
      "balance_loss_clip": 1.06672621,
      "balance_loss_mlp": 1.02916348,
      "epoch": 0.06240606024168821,
      "flos": 15195205975680.0,
      "grad_norm": 2.1038051407212275,
      "language_loss": 0.94952965,
      "learning_rate": 3.989032931426554e-06,
      "loss": 0.97275186,
      "num_input_tokens_seen": 11115395,
      "step": 519,
      "time_per_iteration": 2.6177029609680176
    },
    {
      "auxiliary_loss_clip": 0.01272834,
      "auxiliary_loss_mlp": 0.01044048,
      "balance_loss_clip": 1.06848252,
      "balance_loss_mlp": 1.03039908,
      "epoch": 0.06252630313232731,
      "flos": 20631829910400.0,
      "grad_norm": 2.2249517834938124,
      "language_loss": 0.86677825,
      "learning_rate": 3.9889513156206295e-06,
      "loss": 0.88994706,
      "num_input_tokens_seen": 11134835,
      "step": 520,
      "time_per_iteration": 2.6792399883270264
    },
    {
      "auxiliary_loss_clip": 0.01285444,
      "auxiliary_loss_mlp": 0.01048338,
      "balance_loss_clip": 1.06874955,
      "balance_loss_mlp": 1.03361571,
      "epoch": 0.06264654602296639,
      "flos": 20778058177920.0,
      "grad_norm": 3.0682875940573653,
      "language_loss": 0.73649752,
      "learning_rate": 3.988869398092371e-06,
      "loss": 0.75983536,
      "num_input_tokens_seen": 11154745,
      "step": 521,
      "time_per_iteration": 2.653959274291992
    },
    {
      "auxiliary_loss_clip": 0.0128055,
      "auxiliary_loss_mlp": 0.01046639,
      "balance_loss_clip": 1.07021499,
      "balance_loss_mlp": 1.03242314,
      "epoch": 0.06276678891360549,
      "flos": 29605798241280.0,
      "grad_norm": 2.8931836595825144,
      "language_loss": 0.78925145,
      "learning_rate": 3.988787178854206e-06,
      "loss": 0.81252337,
      "num_input_tokens_seen": 11174280,
      "step": 522,
      "time_per_iteration": 2.7957844734191895
    },
    {
      "auxiliary_loss_clip": 0.01278641,
      "auxiliary_loss_mlp": 0.01047484,
      "balance_loss_clip": 1.07414436,
      "balance_loss_mlp": 1.03373909,
      "epoch": 0.06288703180424457,
      "flos": 22126288193280.0,
      "grad_norm": 3.9558614588451984,
      "language_loss": 0.87381971,
      "learning_rate": 3.988704657918608e-06,
      "loss": 0.89708102,
      "num_input_tokens_seen": 11193340,
      "step": 523,
      "time_per_iteration": 2.6757164001464844
    },
    {
      "auxiliary_loss_clip": 0.01276973,
      "auxiliary_loss_mlp": 0.010489,
      "balance_loss_clip": 1.07357502,
      "balance_loss_mlp": 1.03572762,
      "epoch": 0.06300727469488367,
      "flos": 14976689587200.0,
      "grad_norm": 2.5335427813349285,
      "language_loss": 0.79638928,
      "learning_rate": 3.988621835298094e-06,
      "loss": 0.81964803,
      "num_input_tokens_seen": 11210555,
      "step": 524,
      "time_per_iteration": 2.6579248905181885
    },
    {
      "auxiliary_loss_clip": 0.01274939,
      "auxiliary_loss_mlp": 0.01044653,
      "balance_loss_clip": 1.07412648,
      "balance_loss_mlp": 1.03136075,
      "epoch": 0.06312751758552275,
      "flos": 24535391420160.0,
      "grad_norm": 2.1877076324124447,
      "language_loss": 0.91709369,
      "learning_rate": 3.988538711005229e-06,
      "loss": 0.94028968,
      "num_input_tokens_seen": 11230010,
      "step": 525,
      "time_per_iteration": 2.6964292526245117
    },
    {
      "auxiliary_loss_clip": 0.0127265,
      "auxiliary_loss_mlp": 0.01043249,
      "balance_loss_clip": 1.07147408,
      "balance_loss_mlp": 1.03069067,
      "epoch": 0.06324776047616185,
      "flos": 21507008785920.0,
      "grad_norm": 2.4455256868714836,
      "language_loss": 0.88383865,
      "learning_rate": 3.988455285052622e-06,
      "loss": 0.90699768,
      "num_input_tokens_seen": 11246190,
      "step": 526,
      "time_per_iteration": 2.634455680847168
    },
    {
      "auxiliary_loss_clip": 0.01272192,
      "auxiliary_loss_mlp": 0.01048158,
      "balance_loss_clip": 1.07082224,
      "balance_loss_mlp": 1.03455079,
      "epoch": 0.06336800336680094,
      "flos": 21688034353920.0,
      "grad_norm": 1.9979311005514888,
      "language_loss": 0.83821499,
      "learning_rate": 3.98837155745293e-06,
      "loss": 0.86141843,
      "num_input_tokens_seen": 11264230,
      "step": 527,
      "time_per_iteration": 2.7089157104492188
    },
    {
      "auxiliary_loss_clip": 0.01279486,
      "auxiliary_loss_mlp": 0.01045568,
      "balance_loss_clip": 1.07501197,
      "balance_loss_mlp": 1.03145933,
      "epoch": 0.06348824625744003,
      "flos": 19500895221120.0,
      "grad_norm": 2.20972486983456,
      "language_loss": 0.76050866,
      "learning_rate": 3.988287528218854e-06,
      "loss": 0.78375924,
      "num_input_tokens_seen": 11283015,
      "step": 528,
      "time_per_iteration": 3.5662999153137207
    },
    {
      "auxiliary_loss_clip": 0.01274076,
      "auxiliary_loss_mlp": 0.01045124,
      "balance_loss_clip": 1.07289696,
      "balance_loss_mlp": 1.03239846,
      "epoch": 0.06360848914807912,
      "flos": 15481233976320.0,
      "grad_norm": 2.7950498856190418,
      "language_loss": 0.90155524,
      "learning_rate": 3.98820319736314e-06,
      "loss": 0.92474723,
      "num_input_tokens_seen": 11299630,
      "step": 529,
      "time_per_iteration": 2.7003796100616455
    },
    {
      "auxiliary_loss_clip": 0.01275427,
      "auxiliary_loss_mlp": 0.01042455,
      "balance_loss_clip": 1.06214762,
      "balance_loss_mlp": 1.02907383,
      "epoch": 0.0637287320387182,
      "flos": 20593369422720.0,
      "grad_norm": 4.0690328854006745,
      "language_loss": 0.85295892,
      "learning_rate": 3.988118564898582e-06,
      "loss": 0.87613773,
      "num_input_tokens_seen": 11319170,
      "step": 530,
      "time_per_iteration": 3.6810548305511475
    },
    {
      "auxiliary_loss_clip": 0.01271883,
      "auxiliary_loss_mlp": 0.01210487,
      "balance_loss_clip": 1.06861436,
      "balance_loss_mlp": 1.00037539,
      "epoch": 0.0638489749293573,
      "flos": 17412222245760.0,
      "grad_norm": 2.4872048283957886,
      "language_loss": 0.89273512,
      "learning_rate": 3.988033630838019e-06,
      "loss": 0.91755885,
      "num_input_tokens_seen": 11333210,
      "step": 531,
      "time_per_iteration": 2.6574015617370605
    },
    {
      "auxiliary_loss_clip": 0.01277237,
      "auxiliary_loss_mlp": 0.01052243,
      "balance_loss_clip": 1.07143474,
      "balance_loss_mlp": 1.03923762,
      "epoch": 0.0639692178199964,
      "flos": 23807661874560.0,
      "grad_norm": 2.206965513762802,
      "language_loss": 0.87763983,
      "learning_rate": 3.987948395194334e-06,
      "loss": 0.90093464,
      "num_input_tokens_seen": 11355590,
      "step": 532,
      "time_per_iteration": 4.547638893127441
    },
    {
      "auxiliary_loss_clip": 0.01267233,
      "auxiliary_loss_mlp": 0.01050659,
      "balance_loss_clip": 1.06700182,
      "balance_loss_mlp": 1.03765941,
      "epoch": 0.06408946071063548,
      "flos": 18477225521280.0,
      "grad_norm": 6.463996193780529,
      "language_loss": 0.76455766,
      "learning_rate": 3.987862857980458e-06,
      "loss": 0.78773654,
      "num_input_tokens_seen": 11371535,
      "step": 533,
      "time_per_iteration": 2.6865804195404053
    },
    {
      "auxiliary_loss_clip": 0.01277325,
      "auxiliary_loss_mlp": 0.01045197,
      "balance_loss_clip": 1.06635427,
      "balance_loss_mlp": 1.0314405,
      "epoch": 0.06420970360127458,
      "flos": 27162220936320.0,
      "grad_norm": 2.039567070526941,
      "language_loss": 0.77005535,
      "learning_rate": 3.987777019209368e-06,
      "loss": 0.7932806,
      "num_input_tokens_seen": 11392050,
      "step": 534,
      "time_per_iteration": 2.717513084411621
    },
    {
      "auxiliary_loss_clip": 0.0127771,
      "auxiliary_loss_mlp": 0.01040833,
      "balance_loss_clip": 1.07449269,
      "balance_loss_mlp": 1.02738035,
      "epoch": 0.06432994649191366,
      "flos": 23659673840640.0,
      "grad_norm": 2.310843347230377,
      "language_loss": 0.81134033,
      "learning_rate": 3.987690878894084e-06,
      "loss": 0.83452576,
      "num_input_tokens_seen": 11411765,
      "step": 535,
      "time_per_iteration": 2.694498062133789
    },
    {
      "auxiliary_loss_clip": 0.01280551,
      "auxiliary_loss_mlp": 0.01037852,
      "balance_loss_clip": 1.06931388,
      "balance_loss_mlp": 1.02398777,
      "epoch": 0.06445018938255276,
      "flos": 23403953940480.0,
      "grad_norm": 2.8031244095694934,
      "language_loss": 0.85201156,
      "learning_rate": 3.987604437047673e-06,
      "loss": 0.8751955,
      "num_input_tokens_seen": 11431565,
      "step": 536,
      "time_per_iteration": 2.714003562927246
    },
    {
      "auxiliary_loss_clip": 0.01271297,
      "auxiliary_loss_mlp": 0.010398,
      "balance_loss_clip": 1.0693475,
      "balance_loss_mlp": 1.02699113,
      "epoch": 0.06457043227319184,
      "flos": 19646692525440.0,
      "grad_norm": 2.4540093288611735,
      "language_loss": 0.77734578,
      "learning_rate": 3.987517693683251e-06,
      "loss": 0.8004567,
      "num_input_tokens_seen": 11450140,
      "step": 537,
      "time_per_iteration": 2.687366008758545
    },
    {
      "auxiliary_loss_clip": 0.01273088,
      "auxiliary_loss_mlp": 0.01057304,
      "balance_loss_clip": 1.0698601,
      "balance_loss_mlp": 1.04379129,
      "epoch": 0.06469067516383094,
      "flos": 16978744915200.0,
      "grad_norm": 3.5926423006358132,
      "language_loss": 0.96189797,
      "learning_rate": 3.9874306488139745e-06,
      "loss": 0.98520184,
      "num_input_tokens_seen": 11465400,
      "step": 538,
      "time_per_iteration": 2.7254092693328857
    },
    {
      "auxiliary_loss_clip": 0.0127393,
      "auxiliary_loss_mlp": 0.01045367,
      "balance_loss_clip": 1.06770563,
      "balance_loss_mlp": 1.03190255,
      "epoch": 0.06481091805447003,
      "flos": 23296401642240.0,
      "grad_norm": 2.291511024725637,
      "language_loss": 0.88030291,
      "learning_rate": 3.987343302453049e-06,
      "loss": 0.90349591,
      "num_input_tokens_seen": 11486675,
      "step": 539,
      "time_per_iteration": 2.737197160720825
    },
    {
      "auxiliary_loss_clip": 0.01275214,
      "auxiliary_loss_mlp": 0.01045228,
      "balance_loss_clip": 1.07081795,
      "balance_loss_mlp": 1.03221643,
      "epoch": 0.06493116094510912,
      "flos": 29172356824320.0,
      "grad_norm": 1.6958014539667505,
      "language_loss": 0.82380468,
      "learning_rate": 3.987255654613724e-06,
      "loss": 0.84700906,
      "num_input_tokens_seen": 11510440,
      "step": 540,
      "time_per_iteration": 2.7372899055480957
    },
    {
      "auxiliary_loss_clip": 0.01269587,
      "auxiliary_loss_mlp": 0.01044272,
      "balance_loss_clip": 1.06367135,
      "balance_loss_mlp": 1.03099763,
      "epoch": 0.06505140383574821,
      "flos": 19865065259520.0,
      "grad_norm": 2.3080050961197367,
      "language_loss": 0.70285141,
      "learning_rate": 3.987167705309296e-06,
      "loss": 0.72598994,
      "num_input_tokens_seen": 11529715,
      "step": 541,
      "time_per_iteration": 2.720309257507324
    },
    {
      "auxiliary_loss_clip": 0.01275644,
      "auxiliary_loss_mlp": 0.01209148,
      "balance_loss_clip": 1.06958497,
      "balance_loss_mlp": 1.00040698,
      "epoch": 0.0651716467263873,
      "flos": 17924703540480.0,
      "grad_norm": 3.2776706890801237,
      "language_loss": 0.95609105,
      "learning_rate": 3.987079454553108e-06,
      "loss": 0.98093897,
      "num_input_tokens_seen": 11547665,
      "step": 542,
      "time_per_iteration": 2.617680788040161
    },
    {
      "auxiliary_loss_clip": 0.01273514,
      "auxiliary_loss_mlp": 0.01042988,
      "balance_loss_clip": 1.06936836,
      "balance_loss_mlp": 1.03025007,
      "epoch": 0.0652918896170264,
      "flos": 20842840356480.0,
      "grad_norm": 2.553098177509818,
      "language_loss": 0.90970582,
      "learning_rate": 3.986990902358546e-06,
      "loss": 0.93287081,
      "num_input_tokens_seen": 11564605,
      "step": 543,
      "time_per_iteration": 2.6760382652282715
    },
    {
      "auxiliary_loss_clip": 0.01275705,
      "auxiliary_loss_mlp": 0.01045812,
      "balance_loss_clip": 1.06992102,
      "balance_loss_mlp": 1.0328064,
      "epoch": 0.06541213250766549,
      "flos": 21872507627520.0,
      "grad_norm": 2.352618562902267,
      "language_loss": 0.93499261,
      "learning_rate": 3.986902048739045e-06,
      "loss": 0.95820785,
      "num_input_tokens_seen": 11584550,
      "step": 544,
      "time_per_iteration": 2.720672130584717
    },
    {
      "auxiliary_loss_clip": 0.01277045,
      "auxiliary_loss_mlp": 0.01047265,
      "balance_loss_clip": 1.06971645,
      "balance_loss_mlp": 1.03309095,
      "epoch": 0.06553237539830457,
      "flos": 23110743219840.0,
      "grad_norm": 3.480550513745147,
      "language_loss": 0.79857337,
      "learning_rate": 3.986812893708082e-06,
      "loss": 0.82181644,
      "num_input_tokens_seen": 11600740,
      "step": 545,
      "time_per_iteration": 2.7056150436401367
    },
    {
      "auxiliary_loss_clip": 0.01274057,
      "auxiliary_loss_mlp": 0.01049222,
      "balance_loss_clip": 1.06555152,
      "balance_loss_mlp": 1.03530431,
      "epoch": 0.06565261828894367,
      "flos": 17923769786880.0,
      "grad_norm": 2.4361655799127524,
      "language_loss": 0.815449,
      "learning_rate": 3.9867234372791826e-06,
      "loss": 0.83868182,
      "num_input_tokens_seen": 11618695,
      "step": 546,
      "time_per_iteration": 2.6624748706817627
    },
    {
      "auxiliary_loss_clip": 0.01270022,
      "auxiliary_loss_mlp": 0.01041271,
      "balance_loss_clip": 1.06879747,
      "balance_loss_mlp": 1.02816987,
      "epoch": 0.06577286117958275,
      "flos": 22783058421120.0,
      "grad_norm": 1.59906416100744,
      "language_loss": 0.87134677,
      "learning_rate": 3.986633679465918e-06,
      "loss": 0.89445972,
      "num_input_tokens_seen": 11638850,
      "step": 547,
      "time_per_iteration": 2.650940179824829
    },
    {
      "auxiliary_loss_clip": 0.01277306,
      "auxiliary_loss_mlp": 0.0105055,
      "balance_loss_clip": 1.06694293,
      "balance_loss_mlp": 1.03750277,
      "epoch": 0.06589310407022185,
      "flos": 23696194993920.0,
      "grad_norm": 4.51859549860802,
      "language_loss": 0.80501151,
      "learning_rate": 3.986543620281904e-06,
      "loss": 0.8282901,
      "num_input_tokens_seen": 11658500,
      "step": 548,
      "time_per_iteration": 2.817466974258423
    },
    {
      "auxiliary_loss_clip": 0.01255491,
      "auxiliary_loss_mlp": 0.01037079,
      "balance_loss_clip": 1.06328893,
      "balance_loss_mlp": 1.02390027,
      "epoch": 0.06601334696086093,
      "flos": 26864772410880.0,
      "grad_norm": 2.977684437370494,
      "language_loss": 0.91097444,
      "learning_rate": 3.986453259740802e-06,
      "loss": 0.93390024,
      "num_input_tokens_seen": 11676670,
      "step": 549,
      "time_per_iteration": 2.728306531906128
    },
    {
      "auxiliary_loss_clip": 0.0127199,
      "auxiliary_loss_mlp": 0.01047515,
      "balance_loss_clip": 1.07177901,
      "balance_loss_mlp": 1.03418136,
      "epoch": 0.06613358985150003,
      "flos": 12567694101120.0,
      "grad_norm": 3.3514106206671412,
      "language_loss": 0.78831112,
      "learning_rate": 3.986362597856319e-06,
      "loss": 0.81150615,
      "num_input_tokens_seen": 11693170,
      "step": 550,
      "time_per_iteration": 2.7088661193847656
    },
    {
      "auxiliary_loss_clip": 0.01271391,
      "auxiliary_loss_mlp": 0.0121067,
      "balance_loss_clip": 1.06744456,
      "balance_loss_mlp": 1.00043535,
      "epoch": 0.06625383274213913,
      "flos": 18332505624960.0,
      "grad_norm": 2.9313651470394695,
      "language_loss": 0.81586111,
      "learning_rate": 3.986271634642211e-06,
      "loss": 0.84068179,
      "num_input_tokens_seen": 11710150,
      "step": 551,
      "time_per_iteration": 2.685746669769287
    },
    {
      "auxiliary_loss_clip": 0.01271674,
      "auxiliary_loss_mlp": 0.01049767,
      "balance_loss_clip": 1.07336986,
      "balance_loss_mlp": 1.03600383,
      "epoch": 0.06637407563277821,
      "flos": 15375585098880.0,
      "grad_norm": 2.498567543493808,
      "language_loss": 0.81532204,
      "learning_rate": 3.986180370112274e-06,
      "loss": 0.83853638,
      "num_input_tokens_seen": 11726670,
      "step": 552,
      "time_per_iteration": 2.67557692527771
    },
    {
      "auxiliary_loss_clip": 0.01274198,
      "auxiliary_loss_mlp": 0.01210743,
      "balance_loss_clip": 1.0712148,
      "balance_loss_mlp": 1.00040913,
      "epoch": 0.0664943185234173,
      "flos": 24025244509440.0,
      "grad_norm": 2.3308411820532466,
      "language_loss": 0.74726021,
      "learning_rate": 3.986088804280354e-06,
      "loss": 0.77210963,
      "num_input_tokens_seen": 11746400,
      "step": 553,
      "time_per_iteration": 2.686885356903076
    },
    {
      "auxiliary_loss_clip": 0.01275281,
      "auxiliary_loss_mlp": 0.01050127,
      "balance_loss_clip": 1.06929636,
      "balance_loss_mlp": 1.03623867,
      "epoch": 0.06661456141405639,
      "flos": 20957503547520.0,
      "grad_norm": 2.3335449061688682,
      "language_loss": 0.9409377,
      "learning_rate": 3.985996937160342e-06,
      "loss": 0.96419179,
      "num_input_tokens_seen": 11765590,
      "step": 554,
      "time_per_iteration": 2.6815781593322754
    },
    {
      "auxiliary_loss_clip": 0.0126871,
      "auxiliary_loss_mlp": 0.01050771,
      "balance_loss_clip": 1.06893325,
      "balance_loss_mlp": 1.03747916,
      "epoch": 0.06673480430469549,
      "flos": 52223953322880.0,
      "grad_norm": 2.0126786794138285,
      "language_loss": 0.68920064,
      "learning_rate": 3.985904768766173e-06,
      "loss": 0.71239537,
      "num_input_tokens_seen": 11788365,
      "step": 555,
      "time_per_iteration": 3.9384450912475586
    },
    {
      "auxiliary_loss_clip": 0.01278614,
      "auxiliary_loss_mlp": 0.01044545,
      "balance_loss_clip": 1.06766891,
      "balance_loss_mlp": 1.03050256,
      "epoch": 0.06685504719533458,
      "flos": 16217079995520.0,
      "grad_norm": 2.5820929431132553,
      "language_loss": 0.7611568,
      "learning_rate": 3.98581229911183e-06,
      "loss": 0.78438836,
      "num_input_tokens_seen": 11807285,
      "step": 556,
      "time_per_iteration": 2.661391496658325
    },
    {
      "auxiliary_loss_clip": 0.01275426,
      "auxiliary_loss_mlp": 0.01043968,
      "balance_loss_clip": 1.06963074,
      "balance_loss_mlp": 1.02977014,
      "epoch": 0.06697529008597367,
      "flos": 22491535639680.0,
      "grad_norm": 2.3377006395958264,
      "language_loss": 0.92008197,
      "learning_rate": 3.985719528211341e-06,
      "loss": 0.94327593,
      "num_input_tokens_seen": 11826655,
      "step": 557,
      "time_per_iteration": 3.6133620738983154
    },
    {
      "auxiliary_loss_clip": 0.01177432,
      "auxiliary_loss_mlp": 0.01023465,
      "balance_loss_clip": 1.03645968,
      "balance_loss_mlp": 1.01678967,
      "epoch": 0.06709553297661276,
      "flos": 62688216936960.0,
      "grad_norm": 0.8505465213398051,
      "language_loss": 0.63039052,
      "learning_rate": 3.985626456078777e-06,
      "loss": 0.65239948,
      "num_input_tokens_seen": 11891310,
      "step": 558,
      "time_per_iteration": 3.2550485134124756
    },
    {
      "auxiliary_loss_clip": 0.01279563,
      "auxiliary_loss_mlp": 0.01044526,
      "balance_loss_clip": 1.06901526,
      "balance_loss_mlp": 1.03081739,
      "epoch": 0.06721577586725185,
      "flos": 11216590997760.0,
      "grad_norm": 2.6116405228696222,
      "language_loss": 0.86201036,
      "learning_rate": 3.985533082728259e-06,
      "loss": 0.88525128,
      "num_input_tokens_seen": 11906965,
      "step": 559,
      "time_per_iteration": 4.537302732467651
    },
    {
      "auxiliary_loss_clip": 0.01280008,
      "auxiliary_loss_mlp": 0.01038273,
      "balance_loss_clip": 1.07465291,
      "balance_loss_mlp": 1.02412271,
      "epoch": 0.06733601875789094,
      "flos": 25922189664000.0,
      "grad_norm": 2.172246960159877,
      "language_loss": 0.74461246,
      "learning_rate": 3.985439408173951e-06,
      "loss": 0.76779521,
      "num_input_tokens_seen": 11927190,
      "step": 560,
      "time_per_iteration": 2.6342172622680664
    },
    {
      "auxiliary_loss_clip": 0.01276454,
      "auxiliary_loss_mlp": 0.01057759,
      "balance_loss_clip": 1.07506037,
      "balance_loss_mlp": 1.04382324,
      "epoch": 0.06745626164853002,
      "flos": 20813645577600.0,
      "grad_norm": 3.4108553777007193,
      "language_loss": 0.70883536,
      "learning_rate": 3.9853454324300634e-06,
      "loss": 0.7321775,
      "num_input_tokens_seen": 11946400,
      "step": 561,
      "time_per_iteration": 2.6588199138641357
    },
    {
      "auxiliary_loss_clip": 0.01285616,
      "auxiliary_loss_mlp": 0.01041216,
      "balance_loss_clip": 1.06449163,
      "balance_loss_mlp": 1.02645206,
      "epoch": 0.06757650453916912,
      "flos": 19829262378240.0,
      "grad_norm": 2.3755308764968746,
      "language_loss": 0.77906287,
      "learning_rate": 3.985251155510852e-06,
      "loss": 0.80233121,
      "num_input_tokens_seen": 11965430,
      "step": 562,
      "time_per_iteration": 2.7356858253479004
    },
    {
      "auxiliary_loss_clip": 0.01277853,
      "auxiliary_loss_mlp": 0.01043614,
      "balance_loss_clip": 1.07008576,
      "balance_loss_mlp": 1.0293324,
      "epoch": 0.06769674742980822,
      "flos": 25739224761600.0,
      "grad_norm": 1.848468079796401,
      "language_loss": 0.80302513,
      "learning_rate": 3.98515657743062e-06,
      "loss": 0.82623976,
      "num_input_tokens_seen": 11984895,
      "step": 563,
      "time_per_iteration": 2.7503275871276855
    },
    {
      "auxiliary_loss_clip": 0.01269658,
      "auxiliary_loss_mlp": 0.01052568,
      "balance_loss_clip": 1.06537497,
      "balance_loss_mlp": 1.03898418,
      "epoch": 0.0678169903204473,
      "flos": 13074788355840.0,
      "grad_norm": 2.069201019919908,
      "language_loss": 0.77476966,
      "learning_rate": 3.985061698203711e-06,
      "loss": 0.79799187,
      "num_input_tokens_seen": 12002010,
      "step": 564,
      "time_per_iteration": 2.6507761478424072
    },
    {
      "auxiliary_loss_clip": 0.01170243,
      "auxiliary_loss_mlp": 0.01010151,
      "balance_loss_clip": 1.03944969,
      "balance_loss_mlp": 1.00323665,
      "epoch": 0.0679372332110864,
      "flos": 70865830788480.0,
      "grad_norm": 0.8876237497840581,
      "language_loss": 0.63861197,
      "learning_rate": 3.984966517844523e-06,
      "loss": 0.66041589,
      "num_input_tokens_seen": 12057255,
      "step": 565,
      "time_per_iteration": 3.304333448410034
    },
    {
      "auxiliary_loss_clip": 0.01275876,
      "auxiliary_loss_mlp": 0.01050897,
      "balance_loss_clip": 1.07368469,
      "balance_loss_mlp": 1.03712869,
      "epoch": 0.06805747610172548,
      "flos": 28256418990720.0,
      "grad_norm": 5.650871241324609,
      "language_loss": 0.80599058,
      "learning_rate": 3.984871036367492e-06,
      "loss": 0.82925826,
      "num_input_tokens_seen": 12077280,
      "step": 566,
      "time_per_iteration": 2.7782094478607178
    },
    {
      "auxiliary_loss_clip": 0.01272023,
      "auxiliary_loss_mlp": 0.01210002,
      "balance_loss_clip": 1.07155299,
      "balance_loss_mlp": 1.0004909,
      "epoch": 0.06817771899236458,
      "flos": 20120533764480.0,
      "grad_norm": 2.496798928556162,
      "language_loss": 0.83054173,
      "learning_rate": 3.984775253787102e-06,
      "loss": 0.85536206,
      "num_input_tokens_seen": 12095570,
      "step": 567,
      "time_per_iteration": 2.8224568367004395
    },
    {
      "auxiliary_loss_clip": 0.01277285,
      "auxiliary_loss_mlp": 0.01046133,
      "balance_loss_clip": 1.06888676,
      "balance_loss_mlp": 1.03248906,
      "epoch": 0.06829796188300366,
      "flos": 17930629284480.0,
      "grad_norm": 4.108391596331761,
      "language_loss": 0.88073814,
      "learning_rate": 3.984679170117885e-06,
      "loss": 0.90397227,
      "num_input_tokens_seen": 12111775,
      "step": 568,
      "time_per_iteration": 2.7003519535064697
    },
    {
      "auxiliary_loss_clip": 0.01271299,
      "auxiliary_loss_mlp": 0.01045774,
      "balance_loss_clip": 1.07004142,
      "balance_loss_mlp": 1.0308733,
      "epoch": 0.06841820477364276,
      "flos": 14501627285760.0,
      "grad_norm": 2.5428102912587405,
      "language_loss": 0.78081059,
      "learning_rate": 3.984582785374415e-06,
      "loss": 0.80398136,
      "num_input_tokens_seen": 12129215,
      "step": 569,
      "time_per_iteration": 2.6222190856933594
    },
    {
      "auxiliary_loss_clip": 0.01273157,
      "auxiliary_loss_mlp": 0.01210073,
      "balance_loss_clip": 1.07033217,
      "balance_loss_mlp": 1.00036526,
      "epoch": 0.06853844766428185,
      "flos": 21938474954880.0,
      "grad_norm": 2.550413903591803,
      "language_loss": 0.8091749,
      "learning_rate": 3.9844860995713155e-06,
      "loss": 0.83400726,
      "num_input_tokens_seen": 12148755,
      "step": 570,
      "time_per_iteration": 2.735440731048584
    },
    {
      "auxiliary_loss_clip": 0.01275207,
      "auxiliary_loss_mlp": 0.01043979,
      "balance_loss_clip": 1.07728195,
      "balance_loss_mlp": 1.03069341,
      "epoch": 0.06865869055492094,
      "flos": 16800628348800.0,
      "grad_norm": 3.3292059865681836,
      "language_loss": 0.82739019,
      "learning_rate": 3.9843891127232524e-06,
      "loss": 0.850582,
      "num_input_tokens_seen": 12166290,
      "step": 571,
      "time_per_iteration": 2.605729341506958
    },
    {
      "auxiliary_loss_clip": 0.01259332,
      "auxiliary_loss_mlp": 0.01045985,
      "balance_loss_clip": 1.05981946,
      "balance_loss_mlp": 1.03191853,
      "epoch": 0.06877893344556003,
      "flos": 19937281553280.0,
      "grad_norm": 3.3153617340825385,
      "language_loss": 0.67055732,
      "learning_rate": 3.984291824844938e-06,
      "loss": 0.69361049,
      "num_input_tokens_seen": 12181385,
      "step": 572,
      "time_per_iteration": 2.729987144470215
    },
    {
      "auxiliary_loss_clip": 0.01274183,
      "auxiliary_loss_mlp": 0.01047868,
      "balance_loss_clip": 1.07370424,
      "balance_loss_mlp": 1.03367054,
      "epoch": 0.06889917633619912,
      "flos": 23039388852480.0,
      "grad_norm": 2.6002442808061237,
      "language_loss": 0.85195732,
      "learning_rate": 3.984194235951132e-06,
      "loss": 0.8751778,
      "num_input_tokens_seen": 12197530,
      "step": 573,
      "time_per_iteration": 2.5801005363464355
    },
    {
      "auxiliary_loss_clip": 0.01277462,
      "auxiliary_loss_mlp": 0.01057773,
      "balance_loss_clip": 1.07818317,
      "balance_loss_mlp": 1.04463017,
      "epoch": 0.06901941922683821,
      "flos": 20960556203520.0,
      "grad_norm": 3.260548217005575,
      "language_loss": 0.84329641,
      "learning_rate": 3.9840963460566375e-06,
      "loss": 0.86664867,
      "num_input_tokens_seen": 12216310,
      "step": 574,
      "time_per_iteration": 2.6828629970550537
    },
    {
      "auxiliary_loss_clip": 0.01263655,
      "auxiliary_loss_mlp": 0.01048661,
      "balance_loss_clip": 1.06187916,
      "balance_loss_mlp": 1.03514242,
      "epoch": 0.06913966211747731,
      "flos": 24821850384000.0,
      "grad_norm": 2.387959811797951,
      "language_loss": 0.89635241,
      "learning_rate": 3.983998155176305e-06,
      "loss": 0.91947562,
      "num_input_tokens_seen": 12236670,
      "step": 575,
      "time_per_iteration": 2.7549214363098145
    },
    {
      "auxiliary_loss_clip": 0.01165148,
      "auxiliary_loss_mlp": 0.01010197,
      "balance_loss_clip": 1.03658247,
      "balance_loss_mlp": 1.00375926,
      "epoch": 0.06925990500811639,
      "flos": 58367446957440.0,
      "grad_norm": 0.8178157435923102,
      "language_loss": 0.57041216,
      "learning_rate": 3.9838996633250305e-06,
      "loss": 0.59216559,
      "num_input_tokens_seen": 12297185,
      "step": 576,
      "time_per_iteration": 3.159144639968872
    },
    {
      "auxiliary_loss_clip": 0.01274026,
      "auxiliary_loss_mlp": 0.01042369,
      "balance_loss_clip": 1.07078326,
      "balance_loss_mlp": 1.02989411,
      "epoch": 0.06938014789875549,
      "flos": 12749940731520.0,
      "grad_norm": 2.2227645878857345,
      "language_loss": 0.88125211,
      "learning_rate": 3.983800870517753e-06,
      "loss": 0.90441608,
      "num_input_tokens_seen": 12313975,
      "step": 577,
      "time_per_iteration": 2.6776607036590576
    },
    {
      "auxiliary_loss_clip": 0.01271507,
      "auxiliary_loss_mlp": 0.01047436,
      "balance_loss_clip": 1.07581532,
      "balance_loss_mlp": 1.03510344,
      "epoch": 0.06950039078939457,
      "flos": 22820226019200.0,
      "grad_norm": 2.903287313056097,
      "language_loss": 0.77805638,
      "learning_rate": 3.983701776769463e-06,
      "loss": 0.80124581,
      "num_input_tokens_seen": 12331385,
      "step": 578,
      "time_per_iteration": 2.692758083343506
    },
    {
      "auxiliary_loss_clip": 0.01267256,
      "auxiliary_loss_mlp": 0.01044424,
      "balance_loss_clip": 1.07287335,
      "balance_loss_mlp": 1.03118002,
      "epoch": 0.06962063368003367,
      "flos": 21941348042880.0,
      "grad_norm": 1.9305407675739856,
      "language_loss": 0.85689604,
      "learning_rate": 3.9836023820951885e-06,
      "loss": 0.88001287,
      "num_input_tokens_seen": 12350600,
      "step": 579,
      "time_per_iteration": 2.7417335510253906
    },
    {
      "auxiliary_loss_clip": 0.01268506,
      "auxiliary_loss_mlp": 0.01048161,
      "balance_loss_clip": 1.06436634,
      "balance_loss_mlp": 1.03571534,
      "epoch": 0.06974087657067275,
      "flos": 20706021452160.0,
      "grad_norm": 1.9440821021556074,
      "language_loss": 0.68691671,
      "learning_rate": 3.983502686510011e-06,
      "loss": 0.71008337,
      "num_input_tokens_seen": 12371430,
      "step": 580,
      "time_per_iteration": 2.7115767002105713
    },
    {
      "auxiliary_loss_clip": 0.01275014,
      "auxiliary_loss_mlp": 0.01209451,
      "balance_loss_clip": 1.06946409,
      "balance_loss_mlp": 1.00042987,
      "epoch": 0.06986111946131185,
      "flos": 22638230784000.0,
      "grad_norm": 2.189288541424631,
      "language_loss": 0.73259246,
      "learning_rate": 3.9834026900290525e-06,
      "loss": 0.75743711,
      "num_input_tokens_seen": 12390825,
      "step": 581,
      "time_per_iteration": 2.719184398651123
    },
    {
      "auxiliary_loss_clip": 0.01271252,
      "auxiliary_loss_mlp": 0.01045045,
      "balance_loss_clip": 1.07280564,
      "balance_loss_mlp": 1.03207552,
      "epoch": 0.06998136235195095,
      "flos": 26943453152640.0,
      "grad_norm": 2.397888490139693,
      "language_loss": 1.00096262,
      "learning_rate": 3.983302392667482e-06,
      "loss": 1.0241257,
      "num_input_tokens_seen": 12411670,
      "step": 582,
      "time_per_iteration": 3.628464698791504
    },
    {
      "auxiliary_loss_clip": 0.01272452,
      "auxiliary_loss_mlp": 0.01040065,
      "balance_loss_clip": 1.07414067,
      "balance_loss_mlp": 1.02735758,
      "epoch": 0.07010160524259003,
      "flos": 22492505306880.0,
      "grad_norm": 1.9248793687100865,
      "language_loss": 0.93498075,
      "learning_rate": 3.983201794440517e-06,
      "loss": 0.95810592,
      "num_input_tokens_seen": 12431245,
      "step": 583,
      "time_per_iteration": 3.6075894832611084
    },
    {
      "auxiliary_loss_clip": 0.0126278,
      "auxiliary_loss_mlp": 0.01038967,
      "balance_loss_clip": 1.0689404,
      "balance_loss_mlp": 1.02613986,
      "epoch": 0.07022184813322913,
      "flos": 18332541538560.0,
      "grad_norm": 1.8745089789644076,
      "language_loss": 0.67449772,
      "learning_rate": 3.9831008953634165e-06,
      "loss": 0.69751519,
      "num_input_tokens_seen": 12450535,
      "step": 584,
      "time_per_iteration": 2.6891961097717285
    },
    {
      "auxiliary_loss_clip": 0.01254386,
      "auxiliary_loss_mlp": 0.01042542,
      "balance_loss_clip": 1.05922914,
      "balance_loss_mlp": 1.02900004,
      "epoch": 0.07034209102386821,
      "flos": 24675550289280.0,
      "grad_norm": 5.04534371838688,
      "language_loss": 0.81279409,
      "learning_rate": 3.9829996954514864e-06,
      "loss": 0.83576334,
      "num_input_tokens_seen": 12469675,
      "step": 585,
      "time_per_iteration": 3.6306636333465576
    },
    {
      "auxiliary_loss_clip": 0.01260825,
      "auxiliary_loss_mlp": 0.01044624,
      "balance_loss_clip": 1.06909513,
      "balance_loss_mlp": 1.03114116,
      "epoch": 0.0704623339145073,
      "flos": 25995878415360.0,
      "grad_norm": 2.128700369014741,
      "language_loss": 0.84232831,
      "learning_rate": 3.982898194720079e-06,
      "loss": 0.86538285,
      "num_input_tokens_seen": 12490405,
      "step": 586,
      "time_per_iteration": 3.617377758026123
    },
    {
      "auxiliary_loss_clip": 0.01269027,
      "auxiliary_loss_mlp": 0.01209855,
      "balance_loss_clip": 1.0720855,
      "balance_loss_mlp": 1.00041234,
      "epoch": 0.0705825768051464,
      "flos": 25338318088320.0,
      "grad_norm": 2.3622966872481355,
      "language_loss": 0.82465965,
      "learning_rate": 3.982796393184592e-06,
      "loss": 0.84944844,
      "num_input_tokens_seen": 12509485,
      "step": 587,
      "time_per_iteration": 2.7676773071289062
    },
    {
      "auxiliary_loss_clip": 0.01162526,
      "auxiliary_loss_mlp": 0.01014015,
      "balance_loss_clip": 1.03220975,
      "balance_loss_mlp": 1.00729167,
      "epoch": 0.07070281969578548,
      "flos": 66047552507520.0,
      "grad_norm": 0.7971420749305271,
      "language_loss": 0.62621242,
      "learning_rate": 3.98269429086047e-06,
      "loss": 0.64797783,
      "num_input_tokens_seen": 12567325,
      "step": 588,
      "time_per_iteration": 3.072357654571533
    },
    {
      "auxiliary_loss_clip": 0.01260445,
      "auxiliary_loss_mlp": 0.01048016,
      "balance_loss_clip": 1.06880546,
      "balance_loss_mlp": 1.03423548,
      "epoch": 0.07082306258642458,
      "flos": 23653568528640.0,
      "grad_norm": 2.804140286910665,
      "language_loss": 0.86680871,
      "learning_rate": 3.982591887763199e-06,
      "loss": 0.88989335,
      "num_input_tokens_seen": 12584785,
      "step": 589,
      "time_per_iteration": 2.7329330444335938
    },
    {
      "auxiliary_loss_clip": 0.01249206,
      "auxiliary_loss_mlp": 0.01039347,
      "balance_loss_clip": 1.05809617,
      "balance_loss_mlp": 1.02587628,
      "epoch": 0.07094330547706366,
      "flos": 13880049408000.0,
      "grad_norm": 2.15993419640961,
      "language_loss": 0.81794167,
      "learning_rate": 3.982489183908316e-06,
      "loss": 0.84082723,
      "num_input_tokens_seen": 12601205,
      "step": 590,
      "time_per_iteration": 2.6577351093292236
    },
    {
      "auxiliary_loss_clip": 0.01244444,
      "auxiliary_loss_mlp": 0.01040898,
      "balance_loss_clip": 1.05175388,
      "balance_loss_mlp": 1.02913237,
      "epoch": 0.07106354836770276,
      "flos": 24645098534400.0,
      "grad_norm": 1.8549851291191108,
      "language_loss": 0.84349972,
      "learning_rate": 3.982386179311399e-06,
      "loss": 0.86635315,
      "num_input_tokens_seen": 12621725,
      "step": 591,
      "time_per_iteration": 2.777859687805176
    },
    {
      "auxiliary_loss_clip": 0.01274961,
      "auxiliary_loss_mlp": 0.01047248,
      "balance_loss_clip": 1.07328844,
      "balance_loss_mlp": 1.0325439,
      "epoch": 0.07118379125834184,
      "flos": 16217223649920.0,
      "grad_norm": 2.355171532374396,
      "language_loss": 0.87829912,
      "learning_rate": 3.982282873988075e-06,
      "loss": 0.90152127,
      "num_input_tokens_seen": 12639600,
      "step": 592,
      "time_per_iteration": 2.6579761505126953
    },
    {
      "auxiliary_loss_clip": 0.01270248,
      "auxiliary_loss_mlp": 0.01037861,
      "balance_loss_clip": 1.06921482,
      "balance_loss_mlp": 1.02584493,
      "epoch": 0.07130403414898094,
      "flos": 19719986227200.0,
      "grad_norm": 1.739398563098819,
      "language_loss": 0.87028348,
      "learning_rate": 3.982179267954016e-06,
      "loss": 0.89336461,
      "num_input_tokens_seen": 12660030,
      "step": 593,
      "time_per_iteration": 2.732783079147339
    },
    {
      "auxiliary_loss_clip": 0.012679,
      "auxiliary_loss_mlp": 0.01037819,
      "balance_loss_clip": 1.07101178,
      "balance_loss_mlp": 1.02469361,
      "epoch": 0.07142427703962004,
      "flos": 21871933009920.0,
      "grad_norm": 2.835104832726978,
      "language_loss": 0.96018612,
      "learning_rate": 3.982075361224937e-06,
      "loss": 0.98324323,
      "num_input_tokens_seen": 12678395,
      "step": 594,
      "time_per_iteration": 2.7168116569519043
    },
    {
      "auxiliary_loss_clip": 0.0126643,
      "auxiliary_loss_mlp": 0.01208957,
      "balance_loss_clip": 1.0722152,
      "balance_loss_mlp": 1.00044298,
      "epoch": 0.07154451993025912,
      "flos": 18296595002880.0,
      "grad_norm": 1.9882370130094646,
      "language_loss": 0.87950689,
      "learning_rate": 3.981971153816602e-06,
      "loss": 0.90426075,
      "num_input_tokens_seen": 12696000,
      "step": 595,
      "time_per_iteration": 2.6228725910186768
    },
    {
      "auxiliary_loss_clip": 0.01269012,
      "auxiliary_loss_mlp": 0.01041958,
      "balance_loss_clip": 1.07538843,
      "balance_loss_mlp": 1.02996516,
      "epoch": 0.07166476282089822,
      "flos": 22160690444160.0,
      "grad_norm": 1.8519903056753886,
      "language_loss": 0.96180326,
      "learning_rate": 3.981866645744819e-06,
      "loss": 0.98491299,
      "num_input_tokens_seen": 12716715,
      "step": 596,
      "time_per_iteration": 2.6784873008728027
    },
    {
      "auxiliary_loss_clip": 0.01271957,
      "auxiliary_loss_mlp": 0.01209863,
      "balance_loss_clip": 1.074633,
      "balance_loss_mlp": 1.00035143,
      "epoch": 0.0717850057115373,
      "flos": 14136343925760.0,
      "grad_norm": 2.3658306166048155,
      "language_loss": 0.81311387,
      "learning_rate": 3.9817618370254416e-06,
      "loss": 0.83793211,
      "num_input_tokens_seen": 12733370,
      "step": 597,
      "time_per_iteration": 2.5972659587860107
    },
    {
      "auxiliary_loss_clip": 0.01271385,
      "auxiliary_loss_mlp": 0.01048569,
      "balance_loss_clip": 1.07470107,
      "balance_loss_mlp": 1.03534269,
      "epoch": 0.0719052486021764,
      "flos": 30917794412160.0,
      "grad_norm": 2.4366703470740605,
      "language_loss": 0.87640232,
      "learning_rate": 3.9816567276743684e-06,
      "loss": 0.89960182,
      "num_input_tokens_seen": 12753235,
      "step": 598,
      "time_per_iteration": 2.743959665298462
    },
    {
      "auxiliary_loss_clip": 0.0126384,
      "auxiliary_loss_mlp": 0.01039894,
      "balance_loss_clip": 1.06785583,
      "balance_loss_mlp": 1.0267756,
      "epoch": 0.0720254914928155,
      "flos": 21287019939840.0,
      "grad_norm": 1.952783562443112,
      "language_loss": 0.77981162,
      "learning_rate": 3.9815513177075466e-06,
      "loss": 0.80284894,
      "num_input_tokens_seen": 12772020,
      "step": 599,
      "time_per_iteration": 2.69875431060791
    },
    {
      "auxiliary_loss_clip": 0.01257361,
      "auxiliary_loss_mlp": 0.01042325,
      "balance_loss_clip": 1.06797349,
      "balance_loss_mlp": 1.0310359,
      "epoch": 0.07214573438345458,
      "flos": 27819170732160.0,
      "grad_norm": 1.6730296530890083,
      "language_loss": 0.70399725,
      "learning_rate": 3.9814456071409646e-06,
      "loss": 0.72699416,
      "num_input_tokens_seen": 12792555,
      "step": 600,
      "time_per_iteration": 2.715888738632202
    },
    {
      "auxiliary_loss_clip": 0.01274254,
      "auxiliary_loss_mlp": 0.01051999,
      "balance_loss_clip": 1.06538486,
      "balance_loss_mlp": 1.0382061,
      "epoch": 0.07226597727409367,
      "flos": 25483576688640.0,
      "grad_norm": 2.686024034303169,
      "language_loss": 0.85403132,
      "learning_rate": 3.981339595990659e-06,
      "loss": 0.87729383,
      "num_input_tokens_seen": 12811085,
      "step": 601,
      "time_per_iteration": 2.699216604232788
    },
    {
      "auxiliary_loss_clip": 0.01267025,
      "auxiliary_loss_mlp": 0.01047066,
      "balance_loss_clip": 1.07058597,
      "balance_loss_mlp": 1.03348231,
      "epoch": 0.07238622016473276,
      "flos": 23513840622720.0,
      "grad_norm": 2.4123594980480134,
      "language_loss": 0.81412536,
      "learning_rate": 3.981233284272713e-06,
      "loss": 0.83726621,
      "num_input_tokens_seen": 12830830,
      "step": 602,
      "time_per_iteration": 2.708949327468872
    },
    {
      "auxiliary_loss_clip": 0.01269643,
      "auxiliary_loss_mlp": 0.01041515,
      "balance_loss_clip": 1.06574869,
      "balance_loss_mlp": 1.03010678,
      "epoch": 0.07250646305537185,
      "flos": 25453519983360.0,
      "grad_norm": 1.8046463091668845,
      "language_loss": 0.90025026,
      "learning_rate": 3.981126672003253e-06,
      "loss": 0.9233619,
      "num_input_tokens_seen": 12853505,
      "step": 603,
      "time_per_iteration": 2.736710548400879
    },
    {
      "auxiliary_loss_clip": 0.0127262,
      "auxiliary_loss_mlp": 0.01043591,
      "balance_loss_clip": 1.06519246,
      "balance_loss_mlp": 1.03141391,
      "epoch": 0.07262670594601094,
      "flos": 27155038216320.0,
      "grad_norm": 2.9287115022611214,
      "language_loss": 0.78122723,
      "learning_rate": 3.981019759198451e-06,
      "loss": 0.80438936,
      "num_input_tokens_seen": 12872455,
      "step": 604,
      "time_per_iteration": 2.760751485824585
    },
    {
      "auxiliary_loss_clip": 0.01268431,
      "auxiliary_loss_mlp": 0.01043557,
      "balance_loss_clip": 1.06786203,
      "balance_loss_mlp": 1.03055143,
      "epoch": 0.07274694883665003,
      "flos": 26651607148800.0,
      "grad_norm": 2.506143742211367,
      "language_loss": 0.846084,
      "learning_rate": 3.980912545874528e-06,
      "loss": 0.86920381,
      "num_input_tokens_seen": 12892620,
      "step": 605,
      "time_per_iteration": 2.7869534492492676
    },
    {
      "auxiliary_loss_clip": 0.01264218,
      "auxiliary_loss_mlp": 0.01209486,
      "balance_loss_clip": 1.06990314,
      "balance_loss_mlp": 1.00041199,
      "epoch": 0.07286719172728913,
      "flos": 29862344154240.0,
      "grad_norm": 2.048188742012201,
      "language_loss": 0.85710305,
      "learning_rate": 3.980805032047746e-06,
      "loss": 0.88183999,
      "num_input_tokens_seen": 12914090,
      "step": 606,
      "time_per_iteration": 2.7915267944335938
    },
    {
      "auxiliary_loss_clip": 0.01263065,
      "auxiliary_loss_mlp": 0.01041952,
      "balance_loss_clip": 1.06502724,
      "balance_loss_mlp": 1.02726543,
      "epoch": 0.07298743461792821,
      "flos": 17382057799680.0,
      "grad_norm": 2.096294812955162,
      "language_loss": 0.81187052,
      "learning_rate": 3.980697217734415e-06,
      "loss": 0.8349207,
      "num_input_tokens_seen": 12931830,
      "step": 607,
      "time_per_iteration": 2.6450319290161133
    },
    {
      "auxiliary_loss_clip": 0.01273502,
      "auxiliary_loss_mlp": 0.01209218,
      "balance_loss_clip": 1.06625295,
      "balance_loss_mlp": 1.00043845,
      "epoch": 0.07310767750856731,
      "flos": 19498201701120.0,
      "grad_norm": 2.3612590057161698,
      "language_loss": 0.91739303,
      "learning_rate": 3.980589102950891e-06,
      "loss": 0.94222021,
      "num_input_tokens_seen": 12949995,
      "step": 608,
      "time_per_iteration": 3.64668869972229
    },
    {
      "auxiliary_loss_clip": 0.01271318,
      "auxiliary_loss_mlp": 0.0104171,
      "balance_loss_clip": 1.0726285,
      "balance_loss_mlp": 1.02820981,
      "epoch": 0.07322792039920639,
      "flos": 29168693637120.0,
      "grad_norm": 3.6599366362439265,
      "language_loss": 0.76113075,
      "learning_rate": 3.9804806877135755e-06,
      "loss": 0.78426105,
      "num_input_tokens_seen": 12968040,
      "step": 609,
      "time_per_iteration": 2.777009963989258
    },
    {
      "auxiliary_loss_clip": 0.01274407,
      "auxiliary_loss_mlp": 0.012101,
      "balance_loss_clip": 1.07055664,
      "balance_loss_mlp": 1.00056076,
      "epoch": 0.07334816328984549,
      "flos": 23477822259840.0,
      "grad_norm": 2.3847922495040192,
      "language_loss": 0.86265993,
      "learning_rate": 3.980371972038915e-06,
      "loss": 0.88750494,
      "num_input_tokens_seen": 12988530,
      "step": 610,
      "time_per_iteration": 3.7794058322906494
    },
    {
      "auxiliary_loss_clip": 0.01271902,
      "auxiliary_loss_mlp": 0.01047293,
      "balance_loss_clip": 1.07489109,
      "balance_loss_mlp": 1.03395343,
      "epoch": 0.07346840618048459,
      "flos": 22962467877120.0,
      "grad_norm": 1.8674905173519405,
      "language_loss": 0.84322363,
      "learning_rate": 3.980262955943399e-06,
      "loss": 0.86641562,
      "num_input_tokens_seen": 13008195,
      "step": 611,
      "time_per_iteration": 2.6862616539001465
    },
    {
      "auxiliary_loss_clip": 0.0126604,
      "auxiliary_loss_mlp": 0.01045187,
      "balance_loss_clip": 1.07232606,
      "balance_loss_mlp": 1.03313518,
      "epoch": 0.07358864907112367,
      "flos": 17673903803520.0,
      "grad_norm": 2.9023121746872524,
      "language_loss": 0.86777997,
      "learning_rate": 3.980153639443569e-06,
      "loss": 0.89089227,
      "num_input_tokens_seen": 13024180,
      "step": 612,
      "time_per_iteration": 3.600294351577759
    },
    {
      "auxiliary_loss_clip": 0.01276096,
      "auxiliary_loss_mlp": 0.01046223,
      "balance_loss_clip": 1.07178926,
      "balance_loss_mlp": 1.03314579,
      "epoch": 0.07370889196176277,
      "flos": 24097029840000.0,
      "grad_norm": 1.9331789475192467,
      "language_loss": 0.79989284,
      "learning_rate": 3.980044022556005e-06,
      "loss": 0.82311606,
      "num_input_tokens_seen": 13043865,
      "step": 613,
      "time_per_iteration": 3.6325883865356445
    },
    {
      "auxiliary_loss_clip": 0.01271474,
      "auxiliary_loss_mlp": 0.0105227,
      "balance_loss_clip": 1.07259607,
      "balance_loss_mlp": 1.03963971,
      "epoch": 0.07382913485240185,
      "flos": 25885919905920.0,
      "grad_norm": 2.2212592558176474,
      "language_loss": 0.72788292,
      "learning_rate": 3.9799341052973375e-06,
      "loss": 0.75112033,
      "num_input_tokens_seen": 13063700,
      "step": 614,
      "time_per_iteration": 2.6694443225860596
    },
    {
      "auxiliary_loss_clip": 0.0127253,
      "auxiliary_loss_mlp": 0.01041381,
      "balance_loss_clip": 1.07403278,
      "balance_loss_mlp": 1.02744532,
      "epoch": 0.07394937774304094,
      "flos": 16873850223360.0,
      "grad_norm": 2.7344972354102297,
      "language_loss": 0.74991643,
      "learning_rate": 3.979823887684241e-06,
      "loss": 0.77305555,
      "num_input_tokens_seen": 13082640,
      "step": 615,
      "time_per_iteration": 2.700326919555664
    },
    {
      "auxiliary_loss_clip": 0.01270861,
      "auxiliary_loss_mlp": 0.0104758,
      "balance_loss_clip": 1.07582712,
      "balance_loss_mlp": 1.03415108,
      "epoch": 0.07406962063368003,
      "flos": 20703471586560.0,
      "grad_norm": 6.781313337105927,
      "language_loss": 0.84456241,
      "learning_rate": 3.979713369733434e-06,
      "loss": 0.86774683,
      "num_input_tokens_seen": 13100505,
      "step": 616,
      "time_per_iteration": 2.635222911834717
    },
    {
      "auxiliary_loss_clip": 0.0126652,
      "auxiliary_loss_mlp": 0.0105191,
      "balance_loss_clip": 1.07185864,
      "balance_loss_mlp": 1.03860617,
      "epoch": 0.07418986352431912,
      "flos": 21430985650560.0,
      "grad_norm": 2.0428539366761544,
      "language_loss": 0.85070074,
      "learning_rate": 3.979602551461683e-06,
      "loss": 0.87388504,
      "num_input_tokens_seen": 13121285,
      "step": 617,
      "time_per_iteration": 2.7001700401306152
    },
    {
      "auxiliary_loss_clip": 0.0126457,
      "auxiliary_loss_mlp": 0.01038521,
      "balance_loss_clip": 1.06757796,
      "balance_loss_mlp": 1.02565265,
      "epoch": 0.07431010641495822,
      "flos": 12021133777920.0,
      "grad_norm": 2.7125166527768765,
      "language_loss": 0.92250043,
      "learning_rate": 3.979491432885799e-06,
      "loss": 0.94553137,
      "num_input_tokens_seen": 13137550,
      "step": 618,
      "time_per_iteration": 2.6162519454956055
    },
    {
      "auxiliary_loss_clip": 0.0124892,
      "auxiliary_loss_mlp": 0.01209264,
      "balance_loss_clip": 1.06044531,
      "balance_loss_mlp": 1.00050068,
      "epoch": 0.0744303493055973,
      "flos": 20957575374720.0,
      "grad_norm": 2.6975398616460105,
      "language_loss": 0.82884109,
      "learning_rate": 3.97938001402264e-06,
      "loss": 0.85342294,
      "num_input_tokens_seen": 13156675,
      "step": 619,
      "time_per_iteration": 2.7454073429107666
    },
    {
      "auxiliary_loss_clip": 0.0126242,
      "auxiliary_loss_mlp": 0.01042353,
      "balance_loss_clip": 1.06589508,
      "balance_loss_mlp": 1.02960908,
      "epoch": 0.0745505921962364,
      "flos": 16253134272000.0,
      "grad_norm": 7.878369636180548,
      "language_loss": 0.80183035,
      "learning_rate": 3.979268294889105e-06,
      "loss": 0.8248781,
      "num_input_tokens_seen": 13172225,
      "step": 620,
      "time_per_iteration": 2.6766834259033203
    },
    {
      "auxiliary_loss_clip": 0.01270249,
      "auxiliary_loss_mlp": 0.01043534,
      "balance_loss_clip": 1.07454491,
      "balance_loss_mlp": 1.03058195,
      "epoch": 0.07467083508687548,
      "flos": 50944635550080.0,
      "grad_norm": 1.9742793589832937,
      "language_loss": 0.74295229,
      "learning_rate": 3.979156275502143e-06,
      "loss": 0.7660901,
      "num_input_tokens_seen": 13195885,
      "step": 621,
      "time_per_iteration": 2.990414619445801
    },
    {
      "auxiliary_loss_clip": 0.01270568,
      "auxiliary_loss_mlp": 0.0104886,
      "balance_loss_clip": 1.06565261,
      "balance_loss_mlp": 1.03519821,
      "epoch": 0.07479107797751458,
      "flos": 17529686697600.0,
      "grad_norm": 2.641688880312645,
      "language_loss": 0.91583228,
      "learning_rate": 3.979043955878749e-06,
      "loss": 0.93902653,
      "num_input_tokens_seen": 13213730,
      "step": 622,
      "time_per_iteration": 2.7630767822265625
    },
    {
      "auxiliary_loss_clip": 0.01267566,
      "auxiliary_loss_mlp": 0.01042932,
      "balance_loss_clip": 1.06931615,
      "balance_loss_mlp": 1.03052878,
      "epoch": 0.07491132086815366,
      "flos": 23473943591040.0,
      "grad_norm": 21.03307109029356,
      "language_loss": 0.83093321,
      "learning_rate": 3.978931336035959e-06,
      "loss": 0.85403818,
      "num_input_tokens_seen": 13232540,
      "step": 623,
      "time_per_iteration": 2.7361857891082764
    },
    {
      "auxiliary_loss_clip": 0.01269462,
      "auxiliary_loss_mlp": 0.01050066,
      "balance_loss_clip": 1.07250595,
      "balance_loss_mlp": 1.03666747,
      "epoch": 0.07503156375879276,
      "flos": 20157557708160.0,
      "grad_norm": 3.914992515296537,
      "language_loss": 0.81832045,
      "learning_rate": 3.9788184159908595e-06,
      "loss": 0.84151572,
      "num_input_tokens_seen": 13249670,
      "step": 624,
      "time_per_iteration": 2.7266411781311035
    },
    {
      "auxiliary_loss_clip": 0.01264339,
      "auxiliary_loss_mlp": 0.01048505,
      "balance_loss_clip": 1.06876194,
      "balance_loss_mlp": 1.03590453,
      "epoch": 0.07515180664943186,
      "flos": 15115519653120.0,
      "grad_norm": 4.00884058267607,
      "language_loss": 0.82673532,
      "learning_rate": 3.97870519576058e-06,
      "loss": 0.84986377,
      "num_input_tokens_seen": 13266095,
      "step": 625,
      "time_per_iteration": 2.6454074382781982
    },
    {
      "auxiliary_loss_clip": 0.01269172,
      "auxiliary_loss_mlp": 0.0121011,
      "balance_loss_clip": 1.06755519,
      "balance_loss_mlp": 1.00041437,
      "epoch": 0.07527204954007094,
      "flos": 21287702298240.0,
      "grad_norm": 2.8956055453991714,
      "language_loss": 0.81131506,
      "learning_rate": 3.978591675362295e-06,
      "loss": 0.83610791,
      "num_input_tokens_seen": 13284810,
      "step": 626,
      "time_per_iteration": 2.7594528198242188
    },
    {
      "auxiliary_loss_clip": 0.01267707,
      "auxiliary_loss_mlp": 0.01042308,
      "balance_loss_clip": 1.06939244,
      "balance_loss_mlp": 1.02960646,
      "epoch": 0.07539229243071004,
      "flos": 21324187537920.0,
      "grad_norm": 2.0792393948363292,
      "language_loss": 0.87633944,
      "learning_rate": 3.978477854813226e-06,
      "loss": 0.89943957,
      "num_input_tokens_seen": 13304150,
      "step": 627,
      "time_per_iteration": 2.6954076290130615
    },
    {
      "auxiliary_loss_clip": 0.01270384,
      "auxiliary_loss_mlp": 0.01048056,
      "balance_loss_clip": 1.0714004,
      "balance_loss_mlp": 1.03588462,
      "epoch": 0.07551253532134912,
      "flos": 13042540920960.0,
      "grad_norm": 11.508346794780161,
      "language_loss": 0.82644093,
      "learning_rate": 3.97836373413064e-06,
      "loss": 0.84962535,
      "num_input_tokens_seen": 13322205,
      "step": 628,
      "time_per_iteration": 2.726210355758667
    },
    {
      "auxiliary_loss_clip": 0.0126633,
      "auxiliary_loss_mlp": 0.01042143,
      "balance_loss_clip": 1.07109725,
      "balance_loss_mlp": 1.02900052,
      "epoch": 0.07563277821198822,
      "flos": 19208761908480.0,
      "grad_norm": 2.904763087889035,
      "language_loss": 0.74949354,
      "learning_rate": 3.978249313331848e-06,
      "loss": 0.7725783,
      "num_input_tokens_seen": 13340435,
      "step": 629,
      "time_per_iteration": 2.610400438308716
    },
    {
      "auxiliary_loss_clip": 0.01276226,
      "auxiliary_loss_mlp": 0.01209682,
      "balance_loss_clip": 1.07175541,
      "balance_loss_mlp": 1.00040507,
      "epoch": 0.07575302110262731,
      "flos": 19537200892800.0,
      "grad_norm": 2.873757259101161,
      "language_loss": 0.62736839,
      "learning_rate": 3.978134592434208e-06,
      "loss": 0.65222752,
      "num_input_tokens_seen": 13358185,
      "step": 630,
      "time_per_iteration": 2.764695882797241
    },
    {
      "auxiliary_loss_clip": 0.01168262,
      "auxiliary_loss_mlp": 0.01009252,
      "balance_loss_clip": 1.03293872,
      "balance_loss_mlp": 1.0029099,
      "epoch": 0.0758732639932664,
      "flos": 67961808017280.0,
      "grad_norm": 1.0015046235953973,
      "language_loss": 0.59403068,
      "learning_rate": 3.978019571455123e-06,
      "loss": 0.6158058,
      "num_input_tokens_seen": 13410130,
      "step": 631,
      "time_per_iteration": 3.3177919387817383
    },
    {
      "auxiliary_loss_clip": 0.01265745,
      "auxiliary_loss_mlp": 0.01040022,
      "balance_loss_clip": 1.07313657,
      "balance_loss_mlp": 1.02858973,
      "epoch": 0.07599350688390549,
      "flos": 18989204025600.0,
      "grad_norm": 10.342657031483522,
      "language_loss": 0.83966261,
      "learning_rate": 3.977904250412042e-06,
      "loss": 0.86272031,
      "num_input_tokens_seen": 13429085,
      "step": 632,
      "time_per_iteration": 2.6720752716064453
    },
    {
      "auxiliary_loss_clip": 0.01274259,
      "auxiliary_loss_mlp": 0.01045338,
      "balance_loss_clip": 1.07171488,
      "balance_loss_mlp": 1.0331012,
      "epoch": 0.07611374977454458,
      "flos": 21069006341760.0,
      "grad_norm": 2.2995311306133566,
      "language_loss": 0.85753834,
      "learning_rate": 3.97778862932246e-06,
      "loss": 0.88073432,
      "num_input_tokens_seen": 13446250,
      "step": 633,
      "time_per_iteration": 2.8082354068756104
    },
    {
      "auxiliary_loss_clip": 0.01248161,
      "auxiliary_loss_mlp": 0.01039852,
      "balance_loss_clip": 1.04860318,
      "balance_loss_mlp": 1.02732348,
      "epoch": 0.07623399266518367,
      "flos": 18514536773760.0,
      "grad_norm": 2.2015679254750466,
      "language_loss": 0.9411428,
      "learning_rate": 3.9776727082039144e-06,
      "loss": 0.96402293,
      "num_input_tokens_seen": 13463220,
      "step": 634,
      "time_per_iteration": 3.753690004348755
    },
    {
      "auxiliary_loss_clip": 0.01156729,
      "auxiliary_loss_mlp": 0.0101165,
      "balance_loss_clip": 1.03211546,
      "balance_loss_mlp": 1.00602329,
      "epoch": 0.07635423555582276,
      "flos": 44663036077440.0,
      "grad_norm": 0.8177033316840165,
      "language_loss": 0.55467105,
      "learning_rate": 3.977556487073991e-06,
      "loss": 0.57635486,
      "num_input_tokens_seen": 13517775,
      "step": 635,
      "time_per_iteration": 3.0438337326049805
    },
    {
      "auxiliary_loss_clip": 0.01258699,
      "auxiliary_loss_mlp": 0.01045348,
      "balance_loss_clip": 1.06316721,
      "balance_loss_mlp": 1.03352225,
      "epoch": 0.07647447844646185,
      "flos": 21761148487680.0,
      "grad_norm": 1.8480509279570068,
      "language_loss": 0.8178007,
      "learning_rate": 3.97743996595032e-06,
      "loss": 0.84084117,
      "num_input_tokens_seen": 13537815,
      "step": 636,
      "time_per_iteration": 2.810472011566162
    },
    {
      "auxiliary_loss_clip": 0.01264728,
      "auxiliary_loss_mlp": 0.01046018,
      "balance_loss_clip": 1.07155359,
      "balance_loss_mlp": 1.03272033,
      "epoch": 0.07659472133710095,
      "flos": 23806799948160.0,
      "grad_norm": 1.611937627903467,
      "language_loss": 0.81826067,
      "learning_rate": 3.9773231448505804e-06,
      "loss": 0.84136808,
      "num_input_tokens_seen": 13559605,
      "step": 637,
      "time_per_iteration": 3.6516857147216797
    },
    {
      "auxiliary_loss_clip": 0.01265131,
      "auxiliary_loss_mlp": 0.01210148,
      "balance_loss_clip": 1.06913912,
      "balance_loss_mlp": 1.00044715,
      "epoch": 0.07671496422774003,
      "flos": 21469984842240.0,
      "grad_norm": 2.305218376384661,
      "language_loss": 0.78116721,
      "learning_rate": 3.977206023792491e-06,
      "loss": 0.80592,
      "num_input_tokens_seen": 13579495,
      "step": 638,
      "time_per_iteration": 2.776841878890991
    },
    {
      "auxiliary_loss_clip": 0.01268887,
      "auxiliary_loss_mlp": 0.01053113,
      "balance_loss_clip": 1.07459867,
      "balance_loss_mlp": 1.04086447,
      "epoch": 0.07683520711837913,
      "flos": 16980971558400.0,
      "grad_norm": 2.898517773853587,
      "language_loss": 0.8134141,
      "learning_rate": 3.97708860279382e-06,
      "loss": 0.83663404,
      "num_input_tokens_seen": 13597605,
      "step": 639,
      "time_per_iteration": 4.558034658432007
    },
    {
      "auxiliary_loss_clip": 0.01264592,
      "auxiliary_loss_mlp": 0.01044805,
      "balance_loss_clip": 1.06412935,
      "balance_loss_mlp": 1.03169751,
      "epoch": 0.07695545000901821,
      "flos": 23476744851840.0,
      "grad_norm": 3.315110790933045,
      "language_loss": 0.78176856,
      "learning_rate": 3.97697088187238e-06,
      "loss": 0.8048625,
      "num_input_tokens_seen": 13618120,
      "step": 640,
      "time_per_iteration": 2.689657688140869
    },
    {
      "auxiliary_loss_clip": 0.01262801,
      "auxiliary_loss_mlp": 0.01046238,
      "balance_loss_clip": 1.06961489,
      "balance_loss_mlp": 1.03401268,
      "epoch": 0.07707569289965731,
      "flos": 17634258167040.0,
      "grad_norm": 2.5504978061126393,
      "language_loss": 0.92028522,
      "learning_rate": 3.976852861046029e-06,
      "loss": 0.94337559,
      "num_input_tokens_seen": 13634735,
      "step": 641,
      "time_per_iteration": 2.741819381713867
    },
    {
      "auxiliary_loss_clip": 0.01267107,
      "auxiliary_loss_mlp": 0.01039555,
      "balance_loss_clip": 1.06364059,
      "balance_loss_mlp": 1.02705002,
      "epoch": 0.0771959357902964,
      "flos": 25775674087680.0,
      "grad_norm": 1.67720280143457,
      "language_loss": 0.80074668,
      "learning_rate": 3.97673454033267e-06,
      "loss": 0.82381332,
      "num_input_tokens_seen": 13656835,
      "step": 642,
      "time_per_iteration": 2.7825987339019775
    },
    {
      "auxiliary_loss_clip": 0.0126495,
      "auxiliary_loss_mlp": 0.01047033,
      "balance_loss_clip": 1.06666017,
      "balance_loss_mlp": 1.03396785,
      "epoch": 0.07731617868093549,
      "flos": 19828651847040.0,
      "grad_norm": 3.827640674491623,
      "language_loss": 0.83060145,
      "learning_rate": 3.976615919750254e-06,
      "loss": 0.85372126,
      "num_input_tokens_seen": 13674535,
      "step": 643,
      "time_per_iteration": 2.7658944129943848
    },
    {
      "auxiliary_loss_clip": 0.01262871,
      "auxiliary_loss_mlp": 0.01046523,
      "balance_loss_clip": 1.06964767,
      "balance_loss_mlp": 1.03294468,
      "epoch": 0.07743642157157458,
      "flos": 21324654414720.0,
      "grad_norm": 2.32488490446037,
      "language_loss": 0.86942393,
      "learning_rate": 3.976496999316775e-06,
      "loss": 0.89251786,
      "num_input_tokens_seen": 13693290,
      "step": 644,
      "time_per_iteration": 2.681000232696533
    },
    {
      "auxiliary_loss_clip": 0.01265717,
      "auxiliary_loss_mlp": 0.01048614,
      "balance_loss_clip": 1.07322621,
      "balance_loss_mlp": 1.03532219,
      "epoch": 0.07755666446221367,
      "flos": 19969133938560.0,
      "grad_norm": 3.3474943678579216,
      "language_loss": 0.84067667,
      "learning_rate": 3.976377779050271e-06,
      "loss": 0.86381996,
      "num_input_tokens_seen": 13711420,
      "step": 645,
      "time_per_iteration": 2.7433247566223145
    },
    {
      "auxiliary_loss_clip": 0.01257299,
      "auxiliary_loss_mlp": 0.01045338,
      "balance_loss_clip": 1.06821084,
      "balance_loss_mlp": 1.03295779,
      "epoch": 0.07767690735285276,
      "flos": 23623224514560.0,
      "grad_norm": 2.581018992039056,
      "language_loss": 0.8443234,
      "learning_rate": 3.976258258968831e-06,
      "loss": 0.86734974,
      "num_input_tokens_seen": 13729965,
      "step": 646,
      "time_per_iteration": 2.7128746509552
    },
    {
      "auxiliary_loss_clip": 0.01263706,
      "auxiliary_loss_mlp": 0.01047724,
      "balance_loss_clip": 1.06824684,
      "balance_loss_mlp": 1.03586221,
      "epoch": 0.07779715024349185,
      "flos": 22236246702720.0,
      "grad_norm": 2.769311620033194,
      "language_loss": 0.74041396,
      "learning_rate": 3.976138439090583e-06,
      "loss": 0.76352823,
      "num_input_tokens_seen": 13748045,
      "step": 647,
      "time_per_iteration": 2.656135082244873
    },
    {
      "auxiliary_loss_clip": 0.01270641,
      "auxiliary_loss_mlp": 0.010411,
      "balance_loss_clip": 1.0694474,
      "balance_loss_mlp": 1.02792728,
      "epoch": 0.07791739313413094,
      "flos": 20955097336320.0,
      "grad_norm": 2.2312916813359363,
      "language_loss": 0.84983003,
      "learning_rate": 3.976018319433706e-06,
      "loss": 0.87294745,
      "num_input_tokens_seen": 13765590,
      "step": 648,
      "time_per_iteration": 2.6591737270355225
    },
    {
      "auxiliary_loss_clip": 0.01260305,
      "auxiliary_loss_mlp": 0.01044379,
      "balance_loss_clip": 1.06720662,
      "balance_loss_mlp": 1.03196287,
      "epoch": 0.07803763602477004,
      "flos": 19312327797120.0,
      "grad_norm": 3.010928663001196,
      "language_loss": 0.91197133,
      "learning_rate": 3.9758979000164205e-06,
      "loss": 0.93501824,
      "num_input_tokens_seen": 13782410,
      "step": 649,
      "time_per_iteration": 2.512425184249878
    },
    {
      "auxiliary_loss_clip": 0.01271954,
      "auxiliary_loss_mlp": 0.0103497,
      "balance_loss_clip": 1.0679574,
      "balance_loss_mlp": 1.0212791,
      "epoch": 0.07815787891540912,
      "flos": 22710806213760.0,
      "grad_norm": 9.391331199283416,
      "language_loss": 0.71988666,
      "learning_rate": 3.975777180856995e-06,
      "loss": 0.74295592,
      "num_input_tokens_seen": 13801530,
      "step": 650,
      "time_per_iteration": 2.597712993621826
    },
    {
      "auxiliary_loss_clip": 0.01269155,
      "auxiliary_loss_mlp": 0.01048568,
      "balance_loss_clip": 1.07235694,
      "balance_loss_mlp": 1.03501403,
      "epoch": 0.07827812180604822,
      "flos": 22711129436160.0,
      "grad_norm": 3.2629968774439426,
      "language_loss": 0.86626601,
      "learning_rate": 3.975656161973742e-06,
      "loss": 0.88944322,
      "num_input_tokens_seen": 13820615,
      "step": 651,
      "time_per_iteration": 2.7019386291503906
    },
    {
      "auxiliary_loss_clip": 0.012666,
      "auxiliary_loss_mlp": 0.01042882,
      "balance_loss_clip": 1.07022083,
      "balance_loss_mlp": 1.02917314,
      "epoch": 0.0783983646966873,
      "flos": 21725597001600.0,
      "grad_norm": 2.787314816467026,
      "language_loss": 0.89125943,
      "learning_rate": 3.9755348433850194e-06,
      "loss": 0.91435421,
      "num_input_tokens_seen": 13835955,
      "step": 652,
      "time_per_iteration": 2.649022340774536
    },
    {
      "auxiliary_loss_clip": 0.0115674,
      "auxiliary_loss_mlp": 0.01029404,
      "balance_loss_clip": 1.02823973,
      "balance_loss_mlp": 1.02346754,
      "epoch": 0.0785186075873264,
      "flos": 60640877537280.0,
      "grad_norm": 0.9679011207092333,
      "language_loss": 0.63588738,
      "learning_rate": 3.975413225109232e-06,
      "loss": 0.65774882,
      "num_input_tokens_seen": 13896505,
      "step": 653,
      "time_per_iteration": 3.2341201305389404
    },
    {
      "auxiliary_loss_clip": 0.01262054,
      "auxiliary_loss_mlp": 0.01042596,
      "balance_loss_clip": 1.06700492,
      "balance_loss_mlp": 1.02935791,
      "epoch": 0.0786388504779655,
      "flos": 23877902920320.0,
      "grad_norm": 3.2842056691194155,
      "language_loss": 0.93571109,
      "learning_rate": 3.975291307164829e-06,
      "loss": 0.95875764,
      "num_input_tokens_seen": 13915150,
      "step": 654,
      "time_per_iteration": 2.653759002685547
    },
    {
      "auxiliary_loss_clip": 0.01251502,
      "auxiliary_loss_mlp": 0.01040874,
      "balance_loss_clip": 1.06192636,
      "balance_loss_mlp": 1.02914381,
      "epoch": 0.07875909336860458,
      "flos": 15158684822400.0,
      "grad_norm": 2.1259318821715056,
      "language_loss": 0.85319066,
      "learning_rate": 3.975169089570306e-06,
      "loss": 0.87611443,
      "num_input_tokens_seen": 13933525,
      "step": 655,
      "time_per_iteration": 2.630927085876465
    },
    {
      "auxiliary_loss_clip": 0.01248675,
      "auxiliary_loss_mlp": 0.01040172,
      "balance_loss_clip": 1.06670868,
      "balance_loss_mlp": 1.02730358,
      "epoch": 0.07887933625924368,
      "flos": 22236857233920.0,
      "grad_norm": 2.198383174551144,
      "language_loss": 0.91546547,
      "learning_rate": 3.975046572344202e-06,
      "loss": 0.93835396,
      "num_input_tokens_seen": 13949985,
      "step": 656,
      "time_per_iteration": 2.676954746246338
    },
    {
      "auxiliary_loss_clip": 0.01259929,
      "auxiliary_loss_mlp": 0.01047719,
      "balance_loss_clip": 1.06214309,
      "balance_loss_mlp": 1.03479695,
      "epoch": 0.07899957914988276,
      "flos": 20777734955520.0,
      "grad_norm": 1.9172450817868867,
      "language_loss": 0.7090553,
      "learning_rate": 3.974923755505103e-06,
      "loss": 0.73213184,
      "num_input_tokens_seen": 13969215,
      "step": 657,
      "time_per_iteration": 2.7179858684539795
    },
    {
      "auxiliary_loss_clip": 0.01254391,
      "auxiliary_loss_mlp": 0.01040723,
      "balance_loss_clip": 1.06291819,
      "balance_loss_mlp": 1.02828956,
      "epoch": 0.07911982204052186,
      "flos": 23003047267200.0,
      "grad_norm": 1.749631270962018,
      "language_loss": 0.9094252,
      "learning_rate": 3.974800639071641e-06,
      "loss": 0.93237638,
      "num_input_tokens_seen": 13989935,
      "step": 658,
      "time_per_iteration": 2.650843620300293
    },
    {
      "auxiliary_loss_clip": 0.01254143,
      "auxiliary_loss_mlp": 0.01209239,
      "balance_loss_clip": 1.05891776,
      "balance_loss_mlp": 1.00033796,
      "epoch": 0.07924006493116094,
      "flos": 23111389664640.0,
      "grad_norm": 2.577836328921302,
      "language_loss": 1.00301969,
      "learning_rate": 3.974677223062492e-06,
      "loss": 1.02765346,
      "num_input_tokens_seen": 14007150,
      "step": 659,
      "time_per_iteration": 2.7253925800323486
    },
    {
      "auxiliary_loss_clip": 0.01263313,
      "auxiliary_loss_mlp": 0.01039974,
      "balance_loss_clip": 1.06840217,
      "balance_loss_mlp": 1.02742767,
      "epoch": 0.07936030782180004,
      "flos": 16472153450880.0,
      "grad_norm": 2.8304307144281875,
      "language_loss": 0.74556279,
      "learning_rate": 3.974553507496378e-06,
      "loss": 0.7685957,
      "num_input_tokens_seen": 14025725,
      "step": 660,
      "time_per_iteration": 2.64214825630188
    },
    {
      "auxiliary_loss_clip": 0.01255957,
      "auxiliary_loss_mlp": 0.01042775,
      "balance_loss_clip": 1.06793427,
      "balance_loss_mlp": 1.0281359,
      "epoch": 0.07948055071243913,
      "flos": 23733290764800.0,
      "grad_norm": 2.5052567116690394,
      "language_loss": 0.89427042,
      "learning_rate": 3.974429492392068e-06,
      "loss": 0.91725779,
      "num_input_tokens_seen": 14045750,
      "step": 661,
      "time_per_iteration": 3.6446070671081543
    },
    {
      "auxiliary_loss_clip": 0.01259957,
      "auxiliary_loss_mlp": 0.01209288,
      "balance_loss_clip": 1.07081676,
      "balance_loss_mlp": 1.00031734,
      "epoch": 0.07960079360307822,
      "flos": 19573326996480.0,
      "grad_norm": 3.168593220348475,
      "language_loss": 0.90742803,
      "learning_rate": 3.974305177768373e-06,
      "loss": 0.93212044,
      "num_input_tokens_seen": 14063960,
      "step": 662,
      "time_per_iteration": 2.5738513469696045
    },
    {
      "auxiliary_loss_clip": 0.01254467,
      "auxiliary_loss_mlp": 0.0104636,
      "balance_loss_clip": 1.06365776,
      "balance_loss_mlp": 1.03346109,
      "epoch": 0.07972103649371731,
      "flos": 23513409659520.0,
      "grad_norm": 2.2046274146821814,
      "language_loss": 0.86833388,
      "learning_rate": 3.974180563644152e-06,
      "loss": 0.8913421,
      "num_input_tokens_seen": 14082525,
      "step": 663,
      "time_per_iteration": 2.6710894107818604
    },
    {
      "auxiliary_loss_clip": 0.01266843,
      "auxiliary_loss_mlp": 0.0104404,
      "balance_loss_clip": 1.06705511,
      "balance_loss_mlp": 1.0314393,
      "epoch": 0.0798412793843564,
      "flos": 16726867770240.0,
      "grad_norm": 2.223241051455668,
      "language_loss": 0.89413965,
      "learning_rate": 3.97405565003831e-06,
      "loss": 0.91724849,
      "num_input_tokens_seen": 14098610,
      "step": 664,
      "time_per_iteration": 3.6965057849884033
    },
    {
      "auxiliary_loss_clip": 0.0126348,
      "auxiliary_loss_mlp": 0.01036209,
      "balance_loss_clip": 1.06416535,
      "balance_loss_mlp": 1.02352476,
      "epoch": 0.07996152227499549,
      "flos": 18223337214720.0,
      "grad_norm": 2.598059827125602,
      "language_loss": 0.78233856,
      "learning_rate": 3.973930436969794e-06,
      "loss": 0.8053354,
      "num_input_tokens_seen": 14117065,
      "step": 665,
      "time_per_iteration": 2.8125619888305664
    },
    {
      "auxiliary_loss_clip": 0.01254056,
      "auxiliary_loss_mlp": 0.01043941,
      "balance_loss_clip": 1.06366038,
      "balance_loss_mlp": 1.03106058,
      "epoch": 0.08008176516563459,
      "flos": 20594877793920.0,
      "grad_norm": 3.85692687707478,
      "language_loss": 0.85980344,
      "learning_rate": 3.973804924457602e-06,
      "loss": 0.88278341,
      "num_input_tokens_seen": 14135145,
      "step": 666,
      "time_per_iteration": 4.600157976150513
    },
    {
      "auxiliary_loss_clip": 0.01255255,
      "auxiliary_loss_mlp": 0.0105152,
      "balance_loss_clip": 1.06569266,
      "balance_loss_mlp": 1.03920603,
      "epoch": 0.08020200805627367,
      "flos": 31834306863360.0,
      "grad_norm": 1.8081951133796588,
      "language_loss": 0.85607022,
      "learning_rate": 3.973679112520771e-06,
      "loss": 0.87913799,
      "num_input_tokens_seen": 14156860,
      "step": 667,
      "time_per_iteration": 2.8474862575531006
    },
    {
      "auxiliary_loss_clip": 0.01252683,
      "auxiliary_loss_mlp": 0.01037906,
      "balance_loss_clip": 1.06077337,
      "balance_loss_mlp": 1.02529955,
      "epoch": 0.08032225094691277,
      "flos": 17783503176960.0,
      "grad_norm": 2.217814700393783,
      "language_loss": 0.99049962,
      "learning_rate": 3.973553001178389e-06,
      "loss": 1.01340556,
      "num_input_tokens_seen": 14174365,
      "step": 668,
      "time_per_iteration": 2.6552743911743164
    },
    {
      "auxiliary_loss_clip": 0.01266427,
      "auxiliary_loss_mlp": 0.01038261,
      "balance_loss_clip": 1.06732678,
      "balance_loss_mlp": 1.02583313,
      "epoch": 0.08044249383755185,
      "flos": 24061693835520.0,
      "grad_norm": 3.4942029529386947,
      "language_loss": 0.75652349,
      "learning_rate": 3.973426590449585e-06,
      "loss": 0.77957034,
      "num_input_tokens_seen": 14192320,
      "step": 669,
      "time_per_iteration": 2.7323033809661865
    },
    {
      "auxiliary_loss_clip": 0.01265594,
      "auxiliary_loss_mlp": 0.01036981,
      "balance_loss_clip": 1.06406462,
      "balance_loss_mlp": 1.02474976,
      "epoch": 0.08056273672819095,
      "flos": 18223624523520.0,
      "grad_norm": 2.695464369753556,
      "language_loss": 0.75224799,
      "learning_rate": 3.9732998803535364e-06,
      "loss": 0.77527368,
      "num_input_tokens_seen": 14210380,
      "step": 670,
      "time_per_iteration": 2.6946473121643066
    },
    {
      "auxiliary_loss_clip": 0.01262358,
      "auxiliary_loss_mlp": 0.0104285,
      "balance_loss_clip": 1.06982875,
      "balance_loss_mlp": 1.03028488,
      "epoch": 0.08068297961883003,
      "flos": 19676856971520.0,
      "grad_norm": 2.8522170477850524,
      "language_loss": 0.85700166,
      "learning_rate": 3.973172870909465e-06,
      "loss": 0.88005376,
      "num_input_tokens_seen": 14225145,
      "step": 671,
      "time_per_iteration": 2.6332039833068848
    },
    {
      "auxiliary_loss_clip": 0.0126943,
      "auxiliary_loss_mlp": 0.01042095,
      "balance_loss_clip": 1.06649947,
      "balance_loss_mlp": 1.02885675,
      "epoch": 0.08080322250946913,
      "flos": 23148736830720.0,
      "grad_norm": 2.507717689698909,
      "language_loss": 0.80716467,
      "learning_rate": 3.973045562136638e-06,
      "loss": 0.83027995,
      "num_input_tokens_seen": 14241960,
      "step": 672,
      "time_per_iteration": 2.6300888061523438
    },
    {
      "auxiliary_loss_clip": 0.01268163,
      "auxiliary_loss_mlp": 0.01041123,
      "balance_loss_clip": 1.07072258,
      "balance_loss_mlp": 1.02867186,
      "epoch": 0.08092346540010822,
      "flos": 21763626526080.0,
      "grad_norm": 2.045710342363124,
      "language_loss": 0.91363358,
      "learning_rate": 3.972917954054368e-06,
      "loss": 0.93672645,
      "num_input_tokens_seen": 14260515,
      "step": 673,
      "time_per_iteration": 2.6678380966186523
    },
    {
      "auxiliary_loss_clip": 0.01262208,
      "auxiliary_loss_mlp": 0.01046557,
      "balance_loss_clip": 1.07047331,
      "balance_loss_mlp": 1.03209054,
      "epoch": 0.08104370829074731,
      "flos": 21032485188480.0,
      "grad_norm": 2.485337964823104,
      "language_loss": 0.81928027,
      "learning_rate": 3.972790046682013e-06,
      "loss": 0.84236789,
      "num_input_tokens_seen": 14279190,
      "step": 674,
      "time_per_iteration": 2.672597646713257
    },
    {
      "auxiliary_loss_clip": 0.01261162,
      "auxiliary_loss_mlp": 0.01040364,
      "balance_loss_clip": 1.05968559,
      "balance_loss_mlp": 1.0279901,
      "epoch": 0.0811639511813864,
      "flos": 20083186598400.0,
      "grad_norm": 1.8956202447747945,
      "language_loss": 0.78962827,
      "learning_rate": 3.972661840038977e-06,
      "loss": 0.81264353,
      "num_input_tokens_seen": 14299480,
      "step": 675,
      "time_per_iteration": 2.720322608947754
    },
    {
      "auxiliary_loss_clip": 0.0126634,
      "auxiliary_loss_mlp": 0.0103708,
      "balance_loss_clip": 1.07188022,
      "balance_loss_mlp": 1.02503991,
      "epoch": 0.08128419407202549,
      "flos": 16836718538880.0,
      "grad_norm": 2.1721050830821507,
      "language_loss": 0.83455718,
      "learning_rate": 3.972533334144707e-06,
      "loss": 0.85759139,
      "num_input_tokens_seen": 14316405,
      "step": 676,
      "time_per_iteration": 2.658395767211914
    },
    {
      "auxiliary_loss_clip": 0.01267162,
      "auxiliary_loss_mlp": 0.01040003,
      "balance_loss_clip": 1.06770253,
      "balance_loss_mlp": 1.02724206,
      "epoch": 0.08140443696266458,
      "flos": 23769273214080.0,
      "grad_norm": 2.0062254121837104,
      "language_loss": 0.78819788,
      "learning_rate": 3.972404529018699e-06,
      "loss": 0.81126952,
      "num_input_tokens_seen": 14336265,
      "step": 677,
      "time_per_iteration": 2.638706922531128
    },
    {
      "auxiliary_loss_clip": 0.01257119,
      "auxiliary_loss_mlp": 0.0103632,
      "balance_loss_clip": 1.06195331,
      "balance_loss_mlp": 1.02433372,
      "epoch": 0.08152467985330367,
      "flos": 24390132819840.0,
      "grad_norm": 2.104442598137455,
      "language_loss": 0.85636449,
      "learning_rate": 3.972275424680493e-06,
      "loss": 0.87929893,
      "num_input_tokens_seen": 14356375,
      "step": 678,
      "time_per_iteration": 2.740975856781006
    },
    {
      "auxiliary_loss_clip": 0.01260099,
      "auxiliary_loss_mlp": 0.01036972,
      "balance_loss_clip": 1.06936455,
      "balance_loss_mlp": 1.02500939,
      "epoch": 0.08164492274394276,
      "flos": 19317750750720.0,
      "grad_norm": 2.0060505229184007,
      "language_loss": 0.92215323,
      "learning_rate": 3.972146021149673e-06,
      "loss": 0.94512391,
      "num_input_tokens_seen": 14374650,
      "step": 679,
      "time_per_iteration": 2.613952159881592
    },
    {
      "auxiliary_loss_clip": 0.01262294,
      "auxiliary_loss_mlp": 0.01038295,
      "balance_loss_clip": 1.06504381,
      "balance_loss_mlp": 1.02696419,
      "epoch": 0.08176516563458186,
      "flos": 14830461319680.0,
      "grad_norm": 2.5842256681243225,
      "language_loss": 0.78499019,
      "learning_rate": 3.972016318445868e-06,
      "loss": 0.80799603,
      "num_input_tokens_seen": 14392650,
      "step": 680,
      "time_per_iteration": 2.7111198902130127
    },
    {
      "auxiliary_loss_clip": 0.01260993,
      "auxiliary_loss_mlp": 0.01042213,
      "balance_loss_clip": 1.06851542,
      "balance_loss_mlp": 1.02999997,
      "epoch": 0.08188540852522094,
      "flos": 22602320161920.0,
      "grad_norm": 3.6843389940268905,
      "language_loss": 0.92744595,
      "learning_rate": 3.971886316588757e-06,
      "loss": 0.95047796,
      "num_input_tokens_seen": 14413155,
      "step": 681,
      "time_per_iteration": 2.65543532371521
    },
    {
      "auxiliary_loss_clip": 0.01255298,
      "auxiliary_loss_mlp": 0.01045275,
      "balance_loss_clip": 1.06644583,
      "balance_loss_mlp": 1.03218591,
      "epoch": 0.08200565141586004,
      "flos": 19463727623040.0,
      "grad_norm": 2.8498066187197852,
      "language_loss": 0.73330504,
      "learning_rate": 3.9717560155980595e-06,
      "loss": 0.75631082,
      "num_input_tokens_seen": 14428805,
      "step": 682,
      "time_per_iteration": 2.680370807647705
    },
    {
      "auxiliary_loss_clip": 0.01260766,
      "auxiliary_loss_mlp": 0.01040569,
      "balance_loss_clip": 1.06820428,
      "balance_loss_mlp": 1.02820086,
      "epoch": 0.08212589430649912,
      "flos": 20594662312320.0,
      "grad_norm": 2.6311561648968853,
      "language_loss": 0.91983402,
      "learning_rate": 3.971625415493542e-06,
      "loss": 0.94284737,
      "num_input_tokens_seen": 14447125,
      "step": 683,
      "time_per_iteration": 2.575098991394043
    },
    {
      "auxiliary_loss_clip": 0.01256901,
      "auxiliary_loss_mlp": 0.01040355,
      "balance_loss_clip": 1.0634253,
      "balance_loss_mlp": 1.02760553,
      "epoch": 0.08224613719713822,
      "flos": 25953611086080.0,
      "grad_norm": 2.367458261124159,
      "language_loss": 0.87801445,
      "learning_rate": 3.971494516295017e-06,
      "loss": 0.90098703,
      "num_input_tokens_seen": 14466575,
      "step": 684,
      "time_per_iteration": 2.7452914714813232
    },
    {
      "auxiliary_loss_clip": 0.01264734,
      "auxiliary_loss_mlp": 0.01042592,
      "balance_loss_clip": 1.06270194,
      "balance_loss_mlp": 1.02983069,
      "epoch": 0.08236638008777732,
      "flos": 23768734510080.0,
      "grad_norm": 5.1450324472314835,
      "language_loss": 0.85661393,
      "learning_rate": 3.971363318022341e-06,
      "loss": 0.87968719,
      "num_input_tokens_seen": 14487915,
      "step": 685,
      "time_per_iteration": 2.675476312637329
    },
    {
      "auxiliary_loss_clip": 0.01262919,
      "auxiliary_loss_mlp": 0.01043819,
      "balance_loss_clip": 1.06215453,
      "balance_loss_mlp": 1.03106952,
      "epoch": 0.0824866229784164,
      "flos": 38799144887040.0,
      "grad_norm": 1.8423344342933694,
      "language_loss": 0.68462634,
      "learning_rate": 3.971231820695417e-06,
      "loss": 0.7076937,
      "num_input_tokens_seen": 14511530,
      "step": 686,
      "time_per_iteration": 2.7902963161468506
    },
    {
      "auxiliary_loss_clip": 0.01268173,
      "auxiliary_loss_mlp": 0.01041698,
      "balance_loss_clip": 1.06756639,
      "balance_loss_mlp": 1.02898467,
      "epoch": 0.0826068658690555,
      "flos": 23107762391040.0,
      "grad_norm": 2.460282613495514,
      "language_loss": 0.81565547,
      "learning_rate": 3.971100024334193e-06,
      "loss": 0.8387543,
      "num_input_tokens_seen": 14529050,
      "step": 687,
      "time_per_iteration": 3.6382291316986084
    },
    {
      "auxiliary_loss_clip": 0.01245924,
      "auxiliary_loss_mlp": 0.01040572,
      "balance_loss_clip": 1.05791402,
      "balance_loss_mlp": 1.02900231,
      "epoch": 0.08272710875969458,
      "flos": 21136374299520.0,
      "grad_norm": 2.405092149827521,
      "language_loss": 0.86434913,
      "learning_rate": 3.970967928958663e-06,
      "loss": 0.88721406,
      "num_input_tokens_seen": 14546165,
      "step": 688,
      "time_per_iteration": 2.6526312828063965
    },
    {
      "auxiliary_loss_clip": 0.01265644,
      "auxiliary_loss_mlp": 0.01044156,
      "balance_loss_clip": 1.0620811,
      "balance_loss_mlp": 1.03197312,
      "epoch": 0.08284735165033368,
      "flos": 19063000517760.0,
      "grad_norm": 1.9001470397493556,
      "language_loss": 0.83622652,
      "learning_rate": 3.970835534588865e-06,
      "loss": 0.85932451,
      "num_input_tokens_seen": 14563660,
      "step": 689,
      "time_per_iteration": 2.6898910999298096
    },
    {
      "auxiliary_loss_clip": 0.01263813,
      "auxiliary_loss_mlp": 0.01046804,
      "balance_loss_clip": 1.06931925,
      "balance_loss_mlp": 1.0348537,
      "epoch": 0.08296759454097276,
      "flos": 16727442387840.0,
      "grad_norm": 1.890536467334341,
      "language_loss": 0.85517597,
      "learning_rate": 3.970702841244883e-06,
      "loss": 0.87828213,
      "num_input_tokens_seen": 14581980,
      "step": 690,
      "time_per_iteration": 3.5846118927001953
    },
    {
      "auxiliary_loss_clip": 0.01263701,
      "auxiliary_loss_mlp": 0.01044625,
      "balance_loss_clip": 1.06862319,
      "balance_loss_mlp": 1.03234684,
      "epoch": 0.08308783743161186,
      "flos": 18004928567040.0,
      "grad_norm": 1.768933287076289,
      "language_loss": 0.82556248,
      "learning_rate": 3.970569848946847e-06,
      "loss": 0.84864575,
      "num_input_tokens_seen": 14601795,
      "step": 691,
      "time_per_iteration": 2.608332395553589
    },
    {
      "auxiliary_loss_clip": 0.01245494,
      "auxiliary_loss_mlp": 0.01038494,
      "balance_loss_clip": 1.06469524,
      "balance_loss_mlp": 1.02676952,
      "epoch": 0.08320808032225095,
      "flos": 15079788599040.0,
      "grad_norm": 2.553388777489516,
      "language_loss": 0.82916403,
      "learning_rate": 3.970436557714932e-06,
      "loss": 0.85200393,
      "num_input_tokens_seen": 14618315,
      "step": 692,
      "time_per_iteration": 2.5370287895202637
    },
    {
      "auxiliary_loss_clip": 0.01254549,
      "auxiliary_loss_mlp": 0.01039349,
      "balance_loss_clip": 1.06270504,
      "balance_loss_mlp": 1.02676606,
      "epoch": 0.08332832321289003,
      "flos": 22383085501440.0,
      "grad_norm": 2.986283949937453,
      "language_loss": 0.86558807,
      "learning_rate": 3.970302967569358e-06,
      "loss": 0.88852704,
      "num_input_tokens_seen": 14636905,
      "step": 693,
      "time_per_iteration": 4.46932315826416
    },
    {
      "auxiliary_loss_clip": 0.01262459,
      "auxiliary_loss_mlp": 0.0104177,
      "balance_loss_clip": 1.07196319,
      "balance_loss_mlp": 1.02946162,
      "epoch": 0.08344856610352913,
      "flos": 24717386655360.0,
      "grad_norm": 2.546819052482461,
      "language_loss": 0.68395221,
      "learning_rate": 3.9701690785303896e-06,
      "loss": 0.70699453,
      "num_input_tokens_seen": 14656100,
      "step": 694,
      "time_per_iteration": 2.6650307178497314
    },
    {
      "auxiliary_loss_clip": 0.01265601,
      "auxiliary_loss_mlp": 0.01038905,
      "balance_loss_clip": 1.06795955,
      "balance_loss_mlp": 1.0267992,
      "epoch": 0.08356880899416821,
      "flos": 25370206387200.0,
      "grad_norm": 2.3482468160763705,
      "language_loss": 0.8854546,
      "learning_rate": 3.970034890618339e-06,
      "loss": 0.90849966,
      "num_input_tokens_seen": 14675790,
      "step": 695,
      "time_per_iteration": 2.6556224822998047
    },
    {
      "auxiliary_loss_clip": 0.0124511,
      "auxiliary_loss_mlp": 0.01036911,
      "balance_loss_clip": 1.06436229,
      "balance_loss_mlp": 1.02512741,
      "epoch": 0.08368905188480731,
      "flos": 24353072962560.0,
      "grad_norm": 2.2550107131056336,
      "language_loss": 0.8797431,
      "learning_rate": 3.969900403853562e-06,
      "loss": 0.90256333,
      "num_input_tokens_seen": 14694830,
      "step": 696,
      "time_per_iteration": 2.6818087100982666
    },
    {
      "auxiliary_loss_clip": 0.01262935,
      "auxiliary_loss_mlp": 0.01054575,
      "balance_loss_clip": 1.0710237,
      "balance_loss_mlp": 1.04182518,
      "epoch": 0.08380929477544641,
      "flos": 18037319656320.0,
      "grad_norm": 1.7073111527194007,
      "language_loss": 0.77954721,
      "learning_rate": 3.96976561825646e-06,
      "loss": 0.80272233,
      "num_input_tokens_seen": 14711920,
      "step": 697,
      "time_per_iteration": 2.5562925338745117
    },
    {
      "auxiliary_loss_clip": 0.01265584,
      "auxiliary_loss_mlp": 0.01036195,
      "balance_loss_clip": 1.06362295,
      "balance_loss_mlp": 1.0244236,
      "epoch": 0.08392953766608549,
      "flos": 26286287875200.0,
      "grad_norm": 1.981144889205683,
      "language_loss": 0.86865681,
      "learning_rate": 3.969630533847479e-06,
      "loss": 0.89167458,
      "num_input_tokens_seen": 14730880,
      "step": 698,
      "time_per_iteration": 2.783606767654419
    },
    {
      "auxiliary_loss_clip": 0.01259263,
      "auxiliary_loss_mlp": 0.01035753,
      "balance_loss_clip": 1.06601548,
      "balance_loss_mlp": 1.02383816,
      "epoch": 0.08404978055672459,
      "flos": 22492146170880.0,
      "grad_norm": 2.3982555301941963,
      "language_loss": 0.84174323,
      "learning_rate": 3.969495150647113e-06,
      "loss": 0.8646934,
      "num_input_tokens_seen": 14749050,
      "step": 699,
      "time_per_iteration": 2.6676478385925293
    },
    {
      "auxiliary_loss_clip": 0.01259999,
      "auxiliary_loss_mlp": 0.0104067,
      "balance_loss_clip": 1.06598186,
      "balance_loss_mlp": 1.02852273,
      "epoch": 0.08417002344736367,
      "flos": 24826878288000.0,
      "grad_norm": 1.8606379635284538,
      "language_loss": 0.76354694,
      "learning_rate": 3.969359468675899e-06,
      "loss": 0.78655362,
      "num_input_tokens_seen": 14769180,
      "step": 700,
      "time_per_iteration": 2.6998519897460938
    },
    {
      "auxiliary_loss_clip": 0.01256209,
      "auxiliary_loss_mlp": 0.01038177,
      "balance_loss_clip": 1.06675673,
      "balance_loss_mlp": 1.02575576,
      "epoch": 0.08429026633800277,
      "flos": 16945922862720.0,
      "grad_norm": 2.4482488990248155,
      "language_loss": 0.89335608,
      "learning_rate": 3.969223487954418e-06,
      "loss": 0.91629994,
      "num_input_tokens_seen": 14786640,
      "step": 701,
      "time_per_iteration": 2.6423346996307373
    },
    {
      "auxiliary_loss_clip": 0.01262434,
      "auxiliary_loss_mlp": 0.01041319,
      "balance_loss_clip": 1.06524575,
      "balance_loss_mlp": 1.02958918,
      "epoch": 0.08441050922864185,
      "flos": 23841920471040.0,
      "grad_norm": 2.1141427699342783,
      "language_loss": 0.82868421,
      "learning_rate": 3.969087208503301e-06,
      "loss": 0.85172176,
      "num_input_tokens_seen": 14806720,
      "step": 702,
      "time_per_iteration": 2.7178232669830322
    },
    {
      "auxiliary_loss_clip": 0.01249463,
      "auxiliary_loss_mlp": 0.01039788,
      "balance_loss_clip": 1.06427777,
      "balance_loss_mlp": 1.02793288,
      "epoch": 0.08453075211928095,
      "flos": 25520205582720.0,
      "grad_norm": 4.620638138718304,
      "language_loss": 0.84571368,
      "learning_rate": 3.968950630343219e-06,
      "loss": 0.86860621,
      "num_input_tokens_seen": 14823705,
      "step": 703,
      "time_per_iteration": 2.7920448780059814
    },
    {
      "auxiliary_loss_clip": 0.01257832,
      "auxiliary_loss_mlp": 0.01042119,
      "balance_loss_clip": 1.06288421,
      "balance_loss_mlp": 1.03033543,
      "epoch": 0.08465099500992004,
      "flos": 19532496211200.0,
      "grad_norm": 2.153035074613488,
      "language_loss": 0.9364509,
      "learning_rate": 3.968813753494892e-06,
      "loss": 0.95945042,
      "num_input_tokens_seen": 14841865,
      "step": 704,
      "time_per_iteration": 2.6283156871795654
    },
    {
      "auxiliary_loss_clip": 0.01249514,
      "auxiliary_loss_mlp": 0.01209983,
      "balance_loss_clip": 1.05821276,
      "balance_loss_mlp": 1.00064373,
      "epoch": 0.08477123790055913,
      "flos": 29351299403520.0,
      "grad_norm": 2.7783467105191946,
      "language_loss": 0.75619739,
      "learning_rate": 3.968676577979084e-06,
      "loss": 0.7807923,
      "num_input_tokens_seen": 14861415,
      "step": 705,
      "time_per_iteration": 2.8134500980377197
    },
    {
      "auxiliary_loss_clip": 0.01253251,
      "auxiliary_loss_mlp": 0.01044384,
      "balance_loss_clip": 1.05764413,
      "balance_loss_mlp": 1.03201008,
      "epoch": 0.08489148079119822,
      "flos": 18624495283200.0,
      "grad_norm": 2.118652038616322,
      "language_loss": 0.78441083,
      "learning_rate": 3.968539103816605e-06,
      "loss": 0.80738717,
      "num_input_tokens_seen": 14879215,
      "step": 706,
      "time_per_iteration": 2.6628406047821045
    },
    {
      "auxiliary_loss_clip": 0.01259816,
      "auxiliary_loss_mlp": 0.01209475,
      "balance_loss_clip": 1.06869841,
      "balance_loss_mlp": 1.00064409,
      "epoch": 0.0850117236818373,
      "flos": 23471393725440.0,
      "grad_norm": 2.082196257974558,
      "language_loss": 0.89492738,
      "learning_rate": 3.9684013310283085e-06,
      "loss": 0.91962034,
      "num_input_tokens_seen": 14897900,
      "step": 707,
      "time_per_iteration": 2.724179744720459
    },
    {
      "auxiliary_loss_clip": 0.01256699,
      "auxiliary_loss_mlp": 0.01043026,
      "balance_loss_clip": 1.06819832,
      "balance_loss_mlp": 1.03097975,
      "epoch": 0.0851319665724764,
      "flos": 40625058896640.0,
      "grad_norm": 2.1187360041865086,
      "language_loss": 0.63685071,
      "learning_rate": 3.9682632596350956e-06,
      "loss": 0.65984803,
      "num_input_tokens_seen": 14919065,
      "step": 708,
      "time_per_iteration": 2.817925453186035
    },
    {
      "auxiliary_loss_clip": 0.01256669,
      "auxiliary_loss_mlp": 0.01035376,
      "balance_loss_clip": 1.068156,
      "balance_loss_mlp": 1.02276397,
      "epoch": 0.0852522094631155,
      "flos": 15879554870400.0,
      "grad_norm": 2.115404253978879,
      "language_loss": 0.78661883,
      "learning_rate": 3.968124889657911e-06,
      "loss": 0.80953932,
      "num_input_tokens_seen": 14934165,
      "step": 709,
      "time_per_iteration": 2.6272177696228027
    },
    {
      "auxiliary_loss_clip": 0.01257408,
      "auxiliary_loss_mlp": 0.01039191,
      "balance_loss_clip": 1.06046629,
      "balance_loss_mlp": 1.02834272,
      "epoch": 0.08537245235375458,
      "flos": 14567091822720.0,
      "grad_norm": 2.2051645032199962,
      "language_loss": 0.90809357,
      "learning_rate": 3.967986221117746e-06,
      "loss": 0.93105954,
      "num_input_tokens_seen": 14950105,
      "step": 710,
      "time_per_iteration": 2.622807025909424
    },
    {
      "auxiliary_loss_clip": 0.01269373,
      "auxiliary_loss_mlp": 0.01035447,
      "balance_loss_clip": 1.06095266,
      "balance_loss_mlp": 1.02395523,
      "epoch": 0.08549269524439368,
      "flos": 26468929555200.0,
      "grad_norm": 2.86613626302618,
      "language_loss": 0.8688004,
      "learning_rate": 3.967847254035635e-06,
      "loss": 0.89184856,
      "num_input_tokens_seen": 14969490,
      "step": 711,
      "time_per_iteration": 2.844862222671509
    },
    {
      "auxiliary_loss_clip": 0.0125735,
      "auxiliary_loss_mlp": 0.01038817,
      "balance_loss_clip": 1.06063163,
      "balance_loss_mlp": 1.02700341,
      "epoch": 0.08561293813503276,
      "flos": 13590214565760.0,
      "grad_norm": 3.822481461432657,
      "language_loss": 0.86364549,
      "learning_rate": 3.967707988432661e-06,
      "loss": 0.88660711,
      "num_input_tokens_seen": 14987195,
      "step": 712,
      "time_per_iteration": 2.6690523624420166
    },
    {
      "auxiliary_loss_clip": 0.01259412,
      "auxiliary_loss_mlp": 0.01041012,
      "balance_loss_clip": 1.06790113,
      "balance_loss_mlp": 1.02828062,
      "epoch": 0.08573318102567186,
      "flos": 26943524979840.0,
      "grad_norm": 2.5118511027312245,
      "language_loss": 0.87692475,
      "learning_rate": 3.967568424329949e-06,
      "loss": 0.89992905,
      "num_input_tokens_seen": 15007620,
      "step": 713,
      "time_per_iteration": 2.624325752258301
    },
    {
      "auxiliary_loss_clip": 0.01189043,
      "auxiliary_loss_mlp": 0.0105321,
      "balance_loss_clip": 1.04796791,
      "balance_loss_mlp": 1.04739273,
      "epoch": 0.08585342391631094,
      "flos": 67302739319040.0,
      "grad_norm": 0.8402384321666237,
      "language_loss": 0.55523849,
      "learning_rate": 3.967428561748671e-06,
      "loss": 0.57766104,
      "num_input_tokens_seen": 15075590,
      "step": 714,
      "time_per_iteration": 4.280876636505127
    },
    {
      "auxiliary_loss_clip": 0.01255928,
      "auxiliary_loss_mlp": 0.01040444,
      "balance_loss_clip": 1.0586648,
      "balance_loss_mlp": 1.02773654,
      "epoch": 0.08597366680695004,
      "flos": 22456594684800.0,
      "grad_norm": 1.9994723480477485,
      "language_loss": 0.87893194,
      "learning_rate": 3.967288400710045e-06,
      "loss": 0.90189564,
      "num_input_tokens_seen": 15095055,
      "step": 715,
      "time_per_iteration": 2.7049560546875
    },
    {
      "auxiliary_loss_clip": 0.01254843,
      "auxiliary_loss_mlp": 0.01040225,
      "balance_loss_clip": 1.06605697,
      "balance_loss_mlp": 1.02785158,
      "epoch": 0.08609390969758914,
      "flos": 23550505430400.0,
      "grad_norm": 2.3525091667435736,
      "language_loss": 0.88533223,
      "learning_rate": 3.9671479412353335e-06,
      "loss": 0.90828288,
      "num_input_tokens_seen": 15113520,
      "step": 716,
      "time_per_iteration": 2.683218002319336
    },
    {
      "auxiliary_loss_clip": 0.01259476,
      "auxiliary_loss_mlp": 0.01038764,
      "balance_loss_clip": 1.06846416,
      "balance_loss_mlp": 1.02681899,
      "epoch": 0.08621415258822822,
      "flos": 25885848078720.0,
      "grad_norm": 2.36744221177744,
      "language_loss": 0.742302,
      "learning_rate": 3.967007183345843e-06,
      "loss": 0.76528442,
      "num_input_tokens_seen": 15133375,
      "step": 717,
      "time_per_iteration": 3.6235134601593018
    },
    {
      "auxiliary_loss_clip": 0.01250118,
      "auxiliary_loss_mlp": 0.01039247,
      "balance_loss_clip": 1.06373012,
      "balance_loss_mlp": 1.02741563,
      "epoch": 0.08633439547886732,
      "flos": 13589568120960.0,
      "grad_norm": 3.881521057747085,
      "language_loss": 0.89895612,
      "learning_rate": 3.966866127062927e-06,
      "loss": 0.92184973,
      "num_input_tokens_seen": 15150500,
      "step": 718,
      "time_per_iteration": 2.57582950592041
    },
    {
      "auxiliary_loss_clip": 0.01181425,
      "auxiliary_loss_mlp": 0.01004099,
      "balance_loss_clip": 1.05170977,
      "balance_loss_mlp": 0.99806654,
      "epoch": 0.0864546383695064,
      "flos": 57767342434560.0,
      "grad_norm": 0.8718550618891445,
      "language_loss": 0.62746167,
      "learning_rate": 3.966724772407982e-06,
      "loss": 0.64931691,
      "num_input_tokens_seen": 15208015,
      "step": 719,
      "time_per_iteration": 4.8871190547943115
    },
    {
      "auxiliary_loss_clip": 0.01252914,
      "auxiliary_loss_mlp": 0.01038613,
      "balance_loss_clip": 1.06190968,
      "balance_loss_mlp": 1.02686536,
      "epoch": 0.0865748812601455,
      "flos": 20046952753920.0,
      "grad_norm": 2.2376238853356485,
      "language_loss": 0.88778198,
      "learning_rate": 3.966583119402454e-06,
      "loss": 0.91069728,
      "num_input_tokens_seen": 15224780,
      "step": 720,
      "time_per_iteration": 2.681212902069092
    },
    {
      "auxiliary_loss_clip": 0.01252468,
      "auxiliary_loss_mlp": 0.01209328,
      "balance_loss_clip": 1.06431532,
      "balance_loss_mlp": 1.00060296,
      "epoch": 0.08669512415078459,
      "flos": 35262446935680.0,
      "grad_norm": 1.7378030019667359,
      "language_loss": 0.82345295,
      "learning_rate": 3.9664411680678305e-06,
      "loss": 0.84807098,
      "num_input_tokens_seen": 15246535,
      "step": 721,
      "time_per_iteration": 2.7736151218414307
    },
    {
      "auxiliary_loss_clip": 0.01187948,
      "auxiliary_loss_mlp": 0.01001766,
      "balance_loss_clip": 1.04326749,
      "balance_loss_mlp": 0.9952814,
      "epoch": 0.08681536704142367,
      "flos": 65654870048640.0,
      "grad_norm": 0.8457721330985329,
      "language_loss": 0.61377329,
      "learning_rate": 3.966298918425644e-06,
      "loss": 0.63567048,
      "num_input_tokens_seen": 15304025,
      "step": 722,
      "time_per_iteration": 3.0822010040283203
    },
    {
      "auxiliary_loss_clip": 0.01259045,
      "auxiliary_loss_mlp": 0.01040345,
      "balance_loss_clip": 1.06415927,
      "balance_loss_mlp": 1.02794766,
      "epoch": 0.08693560993206277,
      "flos": 34529940881280.0,
      "grad_norm": 2.373200109558202,
      "language_loss": 0.82587993,
      "learning_rate": 3.966156370497476e-06,
      "loss": 0.84887385,
      "num_input_tokens_seen": 15327635,
      "step": 723,
      "time_per_iteration": 2.722946882247925
    },
    {
      "auxiliary_loss_clip": 0.01260539,
      "auxiliary_loss_mlp": 0.01037094,
      "balance_loss_clip": 1.06451941,
      "balance_loss_mlp": 1.02544737,
      "epoch": 0.08705585282270185,
      "flos": 23149419189120.0,
      "grad_norm": 1.967670888901323,
      "language_loss": 0.886186,
      "learning_rate": 3.96601352430495e-06,
      "loss": 0.90916234,
      "num_input_tokens_seen": 15347405,
      "step": 724,
      "time_per_iteration": 2.6404428482055664
    },
    {
      "auxiliary_loss_clip": 0.01258764,
      "auxiliary_loss_mlp": 0.01044966,
      "balance_loss_clip": 1.0669663,
      "balance_loss_mlp": 1.03257442,
      "epoch": 0.08717609571334095,
      "flos": 29497599498240.0,
      "grad_norm": 1.8201100729034718,
      "language_loss": 0.83292115,
      "learning_rate": 3.965870379869735e-06,
      "loss": 0.85595846,
      "num_input_tokens_seen": 15369450,
      "step": 725,
      "time_per_iteration": 2.7405149936676025
    },
    {
      "auxiliary_loss_clip": 0.01255319,
      "auxiliary_loss_mlp": 0.01035602,
      "balance_loss_clip": 1.06194282,
      "balance_loss_mlp": 1.02396739,
      "epoch": 0.08729633860398003,
      "flos": 20667489137280.0,
      "grad_norm": 2.236510423989846,
      "language_loss": 0.86915499,
      "learning_rate": 3.965726937213547e-06,
      "loss": 0.89206421,
      "num_input_tokens_seen": 15388085,
      "step": 726,
      "time_per_iteration": 2.6442370414733887
    },
    {
      "auxiliary_loss_clip": 0.01254519,
      "auxiliary_loss_mlp": 0.01043003,
      "balance_loss_clip": 1.06197715,
      "balance_loss_mlp": 1.03074801,
      "epoch": 0.08741658149461913,
      "flos": 18369493655040.0,
      "grad_norm": 3.7500220719078676,
      "language_loss": 0.81202388,
      "learning_rate": 3.965583196358144e-06,
      "loss": 0.83499908,
      "num_input_tokens_seen": 15407120,
      "step": 727,
      "time_per_iteration": 2.645047664642334
    },
    {
      "auxiliary_loss_clip": 0.01259955,
      "auxiliary_loss_mlp": 0.01037172,
      "balance_loss_clip": 1.06823123,
      "balance_loss_mlp": 1.02380848,
      "epoch": 0.08753682438525823,
      "flos": 18729677283840.0,
      "grad_norm": 2.4537190489177068,
      "language_loss": 0.74416602,
      "learning_rate": 3.965439157325335e-06,
      "loss": 0.76713717,
      "num_input_tokens_seen": 15424485,
      "step": 728,
      "time_per_iteration": 2.594214677810669
    },
    {
      "auxiliary_loss_clip": 0.01250699,
      "auxiliary_loss_mlp": 0.01035531,
      "balance_loss_clip": 1.0579735,
      "balance_loss_mlp": 1.02222705,
      "epoch": 0.08765706727589731,
      "flos": 27776113303680.0,
      "grad_norm": 2.7453048078883895,
      "language_loss": 0.76146913,
      "learning_rate": 3.965294820136968e-06,
      "loss": 0.78433144,
      "num_input_tokens_seen": 15446285,
      "step": 729,
      "time_per_iteration": 2.774019479751587
    },
    {
      "auxiliary_loss_clip": 0.0126025,
      "auxiliary_loss_mlp": 0.01035261,
      "balance_loss_clip": 1.06436729,
      "balance_loss_mlp": 1.02344728,
      "epoch": 0.08777731016653641,
      "flos": 24389127239040.0,
      "grad_norm": 2.233008841968927,
      "language_loss": 0.87382126,
      "learning_rate": 3.965150184814938e-06,
      "loss": 0.89677638,
      "num_input_tokens_seen": 15465770,
      "step": 730,
      "time_per_iteration": 2.725895881652832
    },
    {
      "auxiliary_loss_clip": 0.0124778,
      "auxiliary_loss_mlp": 0.01040578,
      "balance_loss_clip": 1.0629549,
      "balance_loss_mlp": 1.02850199,
      "epoch": 0.08789755305717549,
      "flos": 21981855605760.0,
      "grad_norm": 2.5838191129822676,
      "language_loss": 0.76551747,
      "learning_rate": 3.965005251381189e-06,
      "loss": 0.78840101,
      "num_input_tokens_seen": 15483705,
      "step": 731,
      "time_per_iteration": 2.71455979347229
    },
    {
      "auxiliary_loss_clip": 0.01180986,
      "auxiliary_loss_mlp": 0.0101008,
      "balance_loss_clip": 1.0460099,
      "balance_loss_mlp": 1.00421453,
      "epoch": 0.08801779594781459,
      "flos": 58360120583040.0,
      "grad_norm": 0.912617120496266,
      "language_loss": 0.64626479,
      "learning_rate": 3.964860019857705e-06,
      "loss": 0.66817546,
      "num_input_tokens_seen": 15548620,
      "step": 732,
      "time_per_iteration": 3.169369697570801
    },
    {
      "auxiliary_loss_clip": 0.01258145,
      "auxiliary_loss_mlp": 0.01037047,
      "balance_loss_clip": 1.06988657,
      "balance_loss_mlp": 1.0259428,
      "epoch": 0.08813803883845367,
      "flos": 23294785530240.0,
      "grad_norm": 1.8959042804011235,
      "language_loss": 0.84234375,
      "learning_rate": 3.964714490266518e-06,
      "loss": 0.86529565,
      "num_input_tokens_seen": 15569265,
      "step": 733,
      "time_per_iteration": 2.668318510055542
    },
    {
      "auxiliary_loss_clip": 0.01175773,
      "auxiliary_loss_mlp": 0.01020326,
      "balance_loss_clip": 1.04271019,
      "balance_loss_mlp": 1.01427066,
      "epoch": 0.08825828172909277,
      "flos": 63424924882560.0,
      "grad_norm": 0.9139204548247766,
      "language_loss": 0.6449554,
      "learning_rate": 3.964568662629706e-06,
      "loss": 0.66691637,
      "num_input_tokens_seen": 15630570,
      "step": 734,
      "time_per_iteration": 3.0840070247650146
    },
    {
      "auxiliary_loss_clip": 0.01250566,
      "auxiliary_loss_mlp": 0.01036066,
      "balance_loss_clip": 1.06183684,
      "balance_loss_mlp": 1.0248481,
      "epoch": 0.08837852461973186,
      "flos": 26720986268160.0,
      "grad_norm": 2.7780874350283584,
      "language_loss": 0.84408402,
      "learning_rate": 3.9644225369693895e-06,
      "loss": 0.86695033,
      "num_input_tokens_seen": 15650870,
      "step": 735,
      "time_per_iteration": 2.692121982574463
    },
    {
      "auxiliary_loss_clip": 0.01253851,
      "auxiliary_loss_mlp": 0.01035053,
      "balance_loss_clip": 1.06723118,
      "balance_loss_mlp": 1.02352548,
      "epoch": 0.08849876751037095,
      "flos": 27265427688960.0,
      "grad_norm": 2.1407653992967606,
      "language_loss": 0.8675735,
      "learning_rate": 3.964276113307735e-06,
      "loss": 0.89046252,
      "num_input_tokens_seen": 15670835,
      "step": 736,
      "time_per_iteration": 2.690622329711914
    },
    {
      "auxiliary_loss_clip": 0.01256476,
      "auxiliary_loss_mlp": 0.01047623,
      "balance_loss_clip": 1.06225276,
      "balance_loss_mlp": 1.03511834,
      "epoch": 0.08861901040101004,
      "flos": 19828759587840.0,
      "grad_norm": 3.2762114442388324,
      "language_loss": 0.80816191,
      "learning_rate": 3.9641293916669574e-06,
      "loss": 0.83120286,
      "num_input_tokens_seen": 15689795,
      "step": 737,
      "time_per_iteration": 2.7158989906311035
    },
    {
      "auxiliary_loss_clip": 0.01250864,
      "auxiliary_loss_mlp": 0.01031988,
      "balance_loss_clip": 1.05983853,
      "balance_loss_mlp": 1.01956618,
      "epoch": 0.08873925329164913,
      "flos": 23658704173440.0,
      "grad_norm": 1.7453223187882976,
      "language_loss": 0.82743204,
      "learning_rate": 3.9639823720693115e-06,
      "loss": 0.85026062,
      "num_input_tokens_seen": 15711650,
      "step": 738,
      "time_per_iteration": 2.7467546463012695
    },
    {
      "auxiliary_loss_clip": 0.01161821,
      "auxiliary_loss_mlp": 0.01015399,
      "balance_loss_clip": 1.0342108,
      "balance_loss_mlp": 1.01013017,
      "epoch": 0.08885949618228822,
      "flos": 71831541893760.0,
      "grad_norm": 0.8341379804899419,
      "language_loss": 0.59972274,
      "learning_rate": 3.963835054537102e-06,
      "loss": 0.62149489,
      "num_input_tokens_seen": 15780615,
      "step": 739,
      "time_per_iteration": 3.2618517875671387
    },
    {
      "auxiliary_loss_clip": 0.01251418,
      "auxiliary_loss_mlp": 0.0104387,
      "balance_loss_clip": 1.05943203,
      "balance_loss_mlp": 1.03225362,
      "epoch": 0.08897973907292732,
      "flos": 22346169298560.0,
      "grad_norm": 2.574633185426983,
      "language_loss": 0.60849869,
      "learning_rate": 3.963687439092676e-06,
      "loss": 0.63145161,
      "num_input_tokens_seen": 15801300,
      "step": 740,
      "time_per_iteration": 2.690434694290161
    },
    {
      "auxiliary_loss_clip": 0.01252633,
      "auxiliary_loss_mlp": 0.01039976,
      "balance_loss_clip": 1.06375921,
      "balance_loss_mlp": 1.02841222,
      "epoch": 0.0890999819635664,
      "flos": 21251827589760.0,
      "grad_norm": 2.0592877215272716,
      "language_loss": 0.80362785,
      "learning_rate": 3.963539525758427e-06,
      "loss": 0.82655394,
      "num_input_tokens_seen": 15820860,
      "step": 741,
      "time_per_iteration": 3.6150834560394287
    },
    {
      "auxiliary_loss_clip": 0.01256757,
      "auxiliary_loss_mlp": 0.01035126,
      "balance_loss_clip": 1.06380463,
      "balance_loss_mlp": 1.02266836,
      "epoch": 0.0892202248542055,
      "flos": 25370888745600.0,
      "grad_norm": 2.1662347692003516,
      "language_loss": 0.67480421,
      "learning_rate": 3.9633913145567925e-06,
      "loss": 0.69772303,
      "num_input_tokens_seen": 15841350,
      "step": 742,
      "time_per_iteration": 2.740525722503662
    },
    {
      "auxiliary_loss_clip": 0.01254687,
      "auxiliary_loss_mlp": 0.01034149,
      "balance_loss_clip": 1.06496716,
      "balance_loss_mlp": 1.02326536,
      "epoch": 0.08934046774484458,
      "flos": 24457895827200.0,
      "grad_norm": 3.8472480214622498,
      "language_loss": 0.81675231,
      "learning_rate": 3.9632428055102575e-06,
      "loss": 0.83964074,
      "num_input_tokens_seen": 15861360,
      "step": 743,
      "time_per_iteration": 3.6529111862182617
    },
    {
      "auxiliary_loss_clip": 0.01256008,
      "auxiliary_loss_mlp": 0.01042148,
      "balance_loss_clip": 1.06671691,
      "balance_loss_mlp": 1.02942276,
      "epoch": 0.08946071063548368,
      "flos": 35772773414400.0,
      "grad_norm": 2.3066835209052896,
      "language_loss": 0.66884172,
      "learning_rate": 3.9630939986413495e-06,
      "loss": 0.69182336,
      "num_input_tokens_seen": 15883160,
      "step": 744,
      "time_per_iteration": 2.7590813636779785
    },
    {
      "auxiliary_loss_clip": 0.01242219,
      "auxiliary_loss_mlp": 0.01041318,
      "balance_loss_clip": 1.05802822,
      "balance_loss_mlp": 1.02941537,
      "epoch": 0.08958095352612276,
      "flos": 14356584167040.0,
      "grad_norm": 2.177084589608245,
      "language_loss": 0.77935541,
      "learning_rate": 3.962944893972643e-06,
      "loss": 0.80219078,
      "num_input_tokens_seen": 15901610,
      "step": 745,
      "time_per_iteration": 2.664335012435913
    },
    {
      "auxiliary_loss_clip": 0.0125089,
      "auxiliary_loss_mlp": 0.01037489,
      "balance_loss_clip": 1.0610013,
      "balance_loss_mlp": 1.02590728,
      "epoch": 0.08970119641676186,
      "flos": 17853277345920.0,
      "grad_norm": 3.87770511674048,
      "language_loss": 0.90591651,
      "learning_rate": 3.962795491526756e-06,
      "loss": 0.92880034,
      "num_input_tokens_seen": 15918770,
      "step": 746,
      "time_per_iteration": 3.53220272064209
    },
    {
      "auxiliary_loss_clip": 0.012577,
      "auxiliary_loss_mlp": 0.01043039,
      "balance_loss_clip": 1.06870866,
      "balance_loss_mlp": 1.03011656,
      "epoch": 0.08982143930740095,
      "flos": 20811670329600.0,
      "grad_norm": 2.6590117127378994,
      "language_loss": 0.89630347,
      "learning_rate": 3.962645791326354e-06,
      "loss": 0.91931081,
      "num_input_tokens_seen": 15938025,
      "step": 747,
      "time_per_iteration": 2.628387689590454
    },
    {
      "auxiliary_loss_clip": 0.0124993,
      "auxiliary_loss_mlp": 0.01034716,
      "balance_loss_clip": 1.06409669,
      "balance_loss_mlp": 1.0240109,
      "epoch": 0.08994168219804004,
      "flos": 24097712198400.0,
      "grad_norm": 2.358421008423701,
      "language_loss": 0.83195519,
      "learning_rate": 3.962495793394146e-06,
      "loss": 0.85480165,
      "num_input_tokens_seen": 15957215,
      "step": 748,
      "time_per_iteration": 2.7861223220825195
    },
    {
      "auxiliary_loss_clip": 0.01173195,
      "auxiliary_loss_mlp": 0.01023844,
      "balance_loss_clip": 1.04713225,
      "balance_loss_mlp": 1.01859903,
      "epoch": 0.09006192508867913,
      "flos": 57188893812480.0,
      "grad_norm": 0.7494030481279381,
      "language_loss": 0.61214089,
      "learning_rate": 3.9623454977528864e-06,
      "loss": 0.63411129,
      "num_input_tokens_seen": 16015870,
      "step": 749,
      "time_per_iteration": 3.0094785690307617
    },
    {
      "auxiliary_loss_clip": 0.01256974,
      "auxiliary_loss_mlp": 0.01045096,
      "balance_loss_clip": 1.05971956,
      "balance_loss_mlp": 1.03331792,
      "epoch": 0.09018216797931822,
      "flos": 20487505063680.0,
      "grad_norm": 4.4843323200764,
      "language_loss": 0.8496027,
      "learning_rate": 3.962194904425375e-06,
      "loss": 0.87262332,
      "num_input_tokens_seen": 16036500,
      "step": 750,
      "time_per_iteration": 2.743330478668213
    },
    {
      "auxiliary_loss_clip": 0.01246503,
      "auxiliary_loss_mlp": 0.01037446,
      "balance_loss_clip": 1.06217611,
      "balance_loss_mlp": 1.02558494,
      "epoch": 0.09030241086995731,
      "flos": 22638123043200.0,
      "grad_norm": 2.421105902466187,
      "language_loss": 0.68383133,
      "learning_rate": 3.9620440134344566e-06,
      "loss": 0.70667088,
      "num_input_tokens_seen": 16054655,
      "step": 751,
      "time_per_iteration": 2.6839568614959717
    },
    {
      "auxiliary_loss_clip": 0.01250909,
      "auxiliary_loss_mlp": 0.01044178,
      "balance_loss_clip": 1.05989575,
      "balance_loss_mlp": 1.03185773,
      "epoch": 0.09042265376059641,
      "flos": 21871502046720.0,
      "grad_norm": 2.3112260511968827,
      "language_loss": 0.82148933,
      "learning_rate": 3.9618928248030215e-06,
      "loss": 0.84444022,
      "num_input_tokens_seen": 16074165,
      "step": 752,
      "time_per_iteration": 2.7276132106781006
    },
    {
      "auxiliary_loss_clip": 0.01249996,
      "auxiliary_loss_mlp": 0.01041387,
      "balance_loss_clip": 1.06451285,
      "balance_loss_mlp": 1.02998471,
      "epoch": 0.0905428966512355,
      "flos": 24316192673280.0,
      "grad_norm": 2.2851508714751745,
      "language_loss": 0.83237123,
      "learning_rate": 3.961741338554005e-06,
      "loss": 0.85528505,
      "num_input_tokens_seen": 16092505,
      "step": 753,
      "time_per_iteration": 2.695803642272949
    },
    {
      "auxiliary_loss_clip": 0.01260569,
      "auxiliary_loss_mlp": 0.01043503,
      "balance_loss_clip": 1.06518447,
      "balance_loss_mlp": 1.03114152,
      "epoch": 0.09066313954187459,
      "flos": 35845061535360.0,
      "grad_norm": 2.3895573397773067,
      "language_loss": 0.76205093,
      "learning_rate": 3.9615895547103865e-06,
      "loss": 0.78509164,
      "num_input_tokens_seen": 16116150,
      "step": 754,
      "time_per_iteration": 2.747941017150879
    },
    {
      "auxiliary_loss_clip": 0.0124908,
      "auxiliary_loss_mlp": 0.0104211,
      "balance_loss_clip": 1.05844069,
      "balance_loss_mlp": 1.03015292,
      "epoch": 0.09078338243251367,
      "flos": 29168729550720.0,
      "grad_norm": 4.640197127138329,
      "language_loss": 0.77697456,
      "learning_rate": 3.961437473295193e-06,
      "loss": 0.79988647,
      "num_input_tokens_seen": 16136295,
      "step": 755,
      "time_per_iteration": 2.7262630462646484
    },
    {
      "auxiliary_loss_clip": 0.01240781,
      "auxiliary_loss_mlp": 0.01036329,
      "balance_loss_clip": 1.05170465,
      "balance_loss_mlp": 1.02495635,
      "epoch": 0.09090362532315277,
      "flos": 21907699977600.0,
      "grad_norm": 2.601296191536863,
      "language_loss": 0.72274482,
      "learning_rate": 3.961285094331495e-06,
      "loss": 0.74551594,
      "num_input_tokens_seen": 16154210,
      "step": 756,
      "time_per_iteration": 2.679595947265625
    },
    {
      "auxiliary_loss_clip": 0.01247774,
      "auxiliary_loss_mlp": 0.0103603,
      "balance_loss_clip": 1.06353068,
      "balance_loss_mlp": 1.02506828,
      "epoch": 0.09102386821379185,
      "flos": 27344503480320.0,
      "grad_norm": 1.884162558534486,
      "language_loss": 0.85917914,
      "learning_rate": 3.961132417842406e-06,
      "loss": 0.88201714,
      "num_input_tokens_seen": 16173995,
      "step": 757,
      "time_per_iteration": 2.75276255607605
    },
    {
      "auxiliary_loss_clip": 0.01243051,
      "auxiliary_loss_mlp": 0.0104302,
      "balance_loss_clip": 1.06083345,
      "balance_loss_mlp": 1.03203523,
      "epoch": 0.09114411110443095,
      "flos": 20813501923200.0,
      "grad_norm": 2.754182723508025,
      "language_loss": 0.75260037,
      "learning_rate": 3.960979443851089e-06,
      "loss": 0.77546108,
      "num_input_tokens_seen": 16191020,
      "step": 758,
      "time_per_iteration": 2.6367452144622803
    },
    {
      "auxiliary_loss_clip": 0.01250715,
      "auxiliary_loss_mlp": 0.01039113,
      "balance_loss_clip": 1.06127501,
      "balance_loss_mlp": 1.02679276,
      "epoch": 0.09126435399507005,
      "flos": 26145949438080.0,
      "grad_norm": 2.5488040922681567,
      "language_loss": 0.79064012,
      "learning_rate": 3.96082617238075e-06,
      "loss": 0.81353837,
      "num_input_tokens_seen": 16213645,
      "step": 759,
      "time_per_iteration": 2.82546329498291
    },
    {
      "auxiliary_loss_clip": 0.01250694,
      "auxiliary_loss_mlp": 0.01036712,
      "balance_loss_clip": 1.06048369,
      "balance_loss_mlp": 1.02586997,
      "epoch": 0.09138459688570913,
      "flos": 24388911757440.0,
      "grad_norm": 2.6201239383094763,
      "language_loss": 0.79993987,
      "learning_rate": 3.960672603454639e-06,
      "loss": 0.82281399,
      "num_input_tokens_seen": 16233625,
      "step": 760,
      "time_per_iteration": 2.6865224838256836
    },
    {
      "auxiliary_loss_clip": 0.01247015,
      "auxiliary_loss_mlp": 0.01038525,
      "balance_loss_clip": 1.06395245,
      "balance_loss_mlp": 1.02639592,
      "epoch": 0.09150483977634823,
      "flos": 21032664756480.0,
      "grad_norm": 3.202002899159235,
      "language_loss": 0.7716096,
      "learning_rate": 3.960518737096054e-06,
      "loss": 0.79446507,
      "num_input_tokens_seen": 16253255,
      "step": 761,
      "time_per_iteration": 2.670940637588501
    },
    {
      "auxiliary_loss_clip": 0.01252782,
      "auxiliary_loss_mlp": 0.01032769,
      "balance_loss_clip": 1.06479847,
      "balance_loss_mlp": 1.02158666,
      "epoch": 0.09162508266698731,
      "flos": 22856998567680.0,
      "grad_norm": 3.157555549217015,
      "language_loss": 0.72554028,
      "learning_rate": 3.960364573328334e-06,
      "loss": 0.7483958,
      "num_input_tokens_seen": 16272580,
      "step": 762,
      "time_per_iteration": 2.643110513687134
    },
    {
      "auxiliary_loss_clip": 0.01252715,
      "auxiliary_loss_mlp": 0.01034727,
      "balance_loss_clip": 1.0587709,
      "balance_loss_mlp": 1.02247798,
      "epoch": 0.0917453255576264,
      "flos": 21724411852800.0,
      "grad_norm": 2.189413367729454,
      "language_loss": 0.88520575,
      "learning_rate": 3.9602101121748675e-06,
      "loss": 0.90808016,
      "num_input_tokens_seen": 16293075,
      "step": 763,
      "time_per_iteration": 2.697286367416382
    },
    {
      "auxiliary_loss_clip": 0.01251861,
      "auxiliary_loss_mlp": 0.01035217,
      "balance_loss_clip": 1.06364346,
      "balance_loss_mlp": 1.02466655,
      "epoch": 0.0918655684482655,
      "flos": 14609215497600.0,
      "grad_norm": 2.192188750823869,
      "language_loss": 0.72736192,
      "learning_rate": 3.960055353659085e-06,
      "loss": 0.75023264,
      "num_input_tokens_seen": 16310185,
      "step": 764,
      "time_per_iteration": 2.6064956188201904
    },
    {
      "auxiliary_loss_clip": 0.0125866,
      "auxiliary_loss_mlp": 0.01031901,
      "balance_loss_clip": 1.06326854,
      "balance_loss_mlp": 1.02079058,
      "epoch": 0.09198581133890459,
      "flos": 23435016226560.0,
      "grad_norm": 1.809858124185328,
      "language_loss": 0.83746153,
      "learning_rate": 3.959900297804465e-06,
      "loss": 0.86036718,
      "num_input_tokens_seen": 16330355,
      "step": 765,
      "time_per_iteration": 2.6899614334106445
    },
    {
      "auxiliary_loss_clip": 0.01240687,
      "auxiliary_loss_mlp": 0.01036697,
      "balance_loss_clip": 1.0613935,
      "balance_loss_mlp": 1.02565765,
      "epoch": 0.09210605422954368,
      "flos": 16795887753600.0,
      "grad_norm": 2.909560590493134,
      "language_loss": 0.77357662,
      "learning_rate": 3.9597449446345276e-06,
      "loss": 0.79635048,
      "num_input_tokens_seen": 16347600,
      "step": 766,
      "time_per_iteration": 2.6756904125213623
    },
    {
      "auxiliary_loss_clip": 0.01239092,
      "auxiliary_loss_mlp": 0.01032957,
      "balance_loss_clip": 1.05683184,
      "balance_loss_mlp": 1.02216291,
      "epoch": 0.09222629712018277,
      "flos": 22674249146880.0,
      "grad_norm": 2.4752767033121366,
      "language_loss": 0.83693582,
      "learning_rate": 3.95958929417284e-06,
      "loss": 0.85965633,
      "num_input_tokens_seen": 16365755,
      "step": 767,
      "time_per_iteration": 2.703176498413086
    },
    {
      "auxiliary_loss_clip": 0.01181708,
      "auxiliary_loss_mlp": 0.01032399,
      "balance_loss_clip": 1.05022728,
      "balance_loss_mlp": 1.02796459,
      "epoch": 0.09234654001082186,
      "flos": 69976756327680.0,
      "grad_norm": 0.7387563284567377,
      "language_loss": 0.58811772,
      "learning_rate": 3.9594333464430145e-06,
      "loss": 0.61025882,
      "num_input_tokens_seen": 16435245,
      "step": 768,
      "time_per_iteration": 4.226280689239502
    },
    {
      "auxiliary_loss_clip": 0.01243953,
      "auxiliary_loss_mlp": 0.01035549,
      "balance_loss_clip": 1.05058503,
      "balance_loss_mlp": 1.02532077,
      "epoch": 0.09246678290146094,
      "flos": 20011437181440.0,
      "grad_norm": 2.580955849927949,
      "language_loss": 0.88320577,
      "learning_rate": 3.959277101468709e-06,
      "loss": 0.90600085,
      "num_input_tokens_seen": 16454795,
      "step": 769,
      "time_per_iteration": 2.8077681064605713
    },
    {
      "auxiliary_loss_clip": 0.01236451,
      "auxiliary_loss_mlp": 0.01041038,
      "balance_loss_clip": 1.05901241,
      "balance_loss_mlp": 1.03011298,
      "epoch": 0.09258702579210004,
      "flos": 17747448900480.0,
      "grad_norm": 2.918036509629474,
      "language_loss": 0.78498328,
      "learning_rate": 3.959120559273624e-06,
      "loss": 0.80775809,
      "num_input_tokens_seen": 16472580,
      "step": 770,
      "time_per_iteration": 3.6233911514282227
    },
    {
      "auxiliary_loss_clip": 0.0123442,
      "auxiliary_loss_mlp": 0.01035972,
      "balance_loss_clip": 1.0591073,
      "balance_loss_mlp": 1.02488017,
      "epoch": 0.09270726868273914,
      "flos": 20886544229760.0,
      "grad_norm": 1.9407790521114334,
      "language_loss": 0.83510959,
      "learning_rate": 3.958963719881509e-06,
      "loss": 0.8578136,
      "num_input_tokens_seen": 16490670,
      "step": 771,
      "time_per_iteration": 2.70627498626709
    },
    {
      "auxiliary_loss_clip": 0.01251729,
      "auxiliary_loss_mlp": 0.01035193,
      "balance_loss_clip": 1.06612968,
      "balance_loss_mlp": 1.0230689,
      "epoch": 0.09282751157337822,
      "flos": 17015697031680.0,
      "grad_norm": 3.032887096579489,
      "language_loss": 0.94284189,
      "learning_rate": 3.958806583316154e-06,
      "loss": 0.96571112,
      "num_input_tokens_seen": 16508640,
      "step": 772,
      "time_per_iteration": 3.574585199356079
    },
    {
      "auxiliary_loss_clip": 0.01250587,
      "auxiliary_loss_mlp": 0.01030992,
      "balance_loss_clip": 1.06769967,
      "balance_loss_mlp": 1.02063322,
      "epoch": 0.09294775446401732,
      "flos": 32523647748480.0,
      "grad_norm": 1.9571989650995238,
      "language_loss": 0.7855233,
      "learning_rate": 3.9586491496013985e-06,
      "loss": 0.808339,
      "num_input_tokens_seen": 16531035,
      "step": 773,
      "time_per_iteration": 3.6072301864624023
    },
    {
      "auxiliary_loss_clip": 0.01255443,
      "auxiliary_loss_mlp": 0.01044443,
      "balance_loss_clip": 1.06483662,
      "balance_loss_mlp": 1.03354776,
      "epoch": 0.0930679973546564,
      "flos": 18259750627200.0,
      "grad_norm": 2.6189171466927337,
      "language_loss": 0.83067757,
      "learning_rate": 3.958491418761124e-06,
      "loss": 0.85367644,
      "num_input_tokens_seen": 16548605,
      "step": 774,
      "time_per_iteration": 2.7423248291015625
    },
    {
      "auxiliary_loss_clip": 0.01251463,
      "auxiliary_loss_mlp": 0.01034147,
      "balance_loss_clip": 1.05905485,
      "balance_loss_mlp": 1.02340627,
      "epoch": 0.0931882402452955,
      "flos": 21099745405440.0,
      "grad_norm": 2.448768978601711,
      "language_loss": 0.72805429,
      "learning_rate": 3.958333390819258e-06,
      "loss": 0.75091034,
      "num_input_tokens_seen": 16565535,
      "step": 775,
      "time_per_iteration": 2.6603505611419678
    },
    {
      "auxiliary_loss_clip": 0.01248765,
      "auxiliary_loss_mlp": 0.01039225,
      "balance_loss_clip": 1.06525588,
      "balance_loss_mlp": 1.02858555,
      "epoch": 0.0933084831359346,
      "flos": 24207275658240.0,
      "grad_norm": 2.245904656861001,
      "language_loss": 0.80387682,
      "learning_rate": 3.9581750657997754e-06,
      "loss": 0.82675672,
      "num_input_tokens_seen": 16584900,
      "step": 776,
      "time_per_iteration": 2.703869342803955
    },
    {
      "auxiliary_loss_clip": 0.01248673,
      "auxiliary_loss_mlp": 0.01037957,
      "balance_loss_clip": 1.05807745,
      "balance_loss_mlp": 1.02746618,
      "epoch": 0.09342872602657368,
      "flos": 25480272637440.0,
      "grad_norm": 2.0587904748441117,
      "language_loss": 0.8957659,
      "learning_rate": 3.95801644372669e-06,
      "loss": 0.91863215,
      "num_input_tokens_seen": 16604805,
      "step": 777,
      "time_per_iteration": 2.7564332485198975
    },
    {
      "auxiliary_loss_clip": 0.01257361,
      "auxiliary_loss_mlp": 0.01042396,
      "balance_loss_clip": 1.05961418,
      "balance_loss_mlp": 1.03142297,
      "epoch": 0.09354896891721277,
      "flos": 23149060053120.0,
      "grad_norm": 2.273116158258394,
      "language_loss": 0.8448559,
      "learning_rate": 3.957857524624068e-06,
      "loss": 0.86785346,
      "num_input_tokens_seen": 16623685,
      "step": 778,
      "time_per_iteration": 2.737070083618164
    },
    {
      "auxiliary_loss_clip": 0.01250662,
      "auxiliary_loss_mlp": 0.01042632,
      "balance_loss_clip": 1.06144011,
      "balance_loss_mlp": 1.03220093,
      "epoch": 0.09366921180785186,
      "flos": 24279563779200.0,
      "grad_norm": 1.7123062649515006,
      "language_loss": 0.89750522,
      "learning_rate": 3.957698308516016e-06,
      "loss": 0.92043811,
      "num_input_tokens_seen": 16644985,
      "step": 779,
      "time_per_iteration": 2.7531890869140625
    },
    {
      "auxiliary_loss_clip": 0.01246395,
      "auxiliary_loss_mlp": 0.01208423,
      "balance_loss_clip": 1.06489503,
      "balance_loss_mlp": 1.00071251,
      "epoch": 0.09378945469849095,
      "flos": 18730036419840.0,
      "grad_norm": 3.0368405575545907,
      "language_loss": 0.82539415,
      "learning_rate": 3.957538795426688e-06,
      "loss": 0.84994233,
      "num_input_tokens_seen": 16662410,
      "step": 780,
      "time_per_iteration": 2.6286730766296387
    },
    {
      "auxiliary_loss_clip": 0.01251375,
      "auxiliary_loss_mlp": 0.01042061,
      "balance_loss_clip": 1.060287,
      "balance_loss_mlp": 1.03010428,
      "epoch": 0.09390969758913004,
      "flos": 23218834222080.0,
      "grad_norm": 2.8426382699172166,
      "language_loss": 0.77514356,
      "learning_rate": 3.9573789853802804e-06,
      "loss": 0.79807794,
      "num_input_tokens_seen": 16680885,
      "step": 781,
      "time_per_iteration": 2.690441131591797
    },
    {
      "auxiliary_loss_clip": 0.01250279,
      "auxiliary_loss_mlp": 0.01208176,
      "balance_loss_clip": 1.06229734,
      "balance_loss_mlp": 1.00070548,
      "epoch": 0.09402994047976913,
      "flos": 19646728439040.0,
      "grad_norm": 2.8194868587579456,
      "language_loss": 0.74649751,
      "learning_rate": 3.957218878401037e-06,
      "loss": 0.7710821,
      "num_input_tokens_seen": 16699375,
      "step": 782,
      "time_per_iteration": 2.6680727005004883
    },
    {
      "auxiliary_loss_clip": 0.0124955,
      "auxiliary_loss_mlp": 0.01041274,
      "balance_loss_clip": 1.06594646,
      "balance_loss_mlp": 1.02984142,
      "epoch": 0.09415018337040823,
      "flos": 29420463041280.0,
      "grad_norm": 2.290294985091577,
      "language_loss": 0.89434719,
      "learning_rate": 3.957058474513246e-06,
      "loss": 0.9172554,
      "num_input_tokens_seen": 16719230,
      "step": 783,
      "time_per_iteration": 2.6991286277770996
    },
    {
      "auxiliary_loss_clip": 0.01245518,
      "auxiliary_loss_mlp": 0.0103851,
      "balance_loss_clip": 1.06349945,
      "balance_loss_mlp": 1.02861595,
      "epoch": 0.09427042626104731,
      "flos": 24572092141440.0,
      "grad_norm": 1.9770765846678593,
      "language_loss": 0.78853297,
      "learning_rate": 3.956897773741241e-06,
      "loss": 0.81137323,
      "num_input_tokens_seen": 16738220,
      "step": 784,
      "time_per_iteration": 2.686946392059326
    },
    {
      "auxiliary_loss_clip": 0.01238977,
      "auxiliary_loss_mlp": 0.01043756,
      "balance_loss_clip": 1.06002688,
      "balance_loss_mlp": 1.03227067,
      "epoch": 0.09439066915168641,
      "flos": 26359581576960.0,
      "grad_norm": 1.8808940661864892,
      "language_loss": 0.71803916,
      "learning_rate": 3.956736776109398e-06,
      "loss": 0.74086654,
      "num_input_tokens_seen": 16759395,
      "step": 785,
      "time_per_iteration": 2.721327304840088
    },
    {
      "auxiliary_loss_clip": 0.01239688,
      "auxiliary_loss_mlp": 0.01208826,
      "balance_loss_clip": 1.05935729,
      "balance_loss_mlp": 1.00078952,
      "epoch": 0.09451091204232549,
      "flos": 19427278296960.0,
      "grad_norm": 2.0506171307272396,
      "language_loss": 0.83302522,
      "learning_rate": 3.956575481642143e-06,
      "loss": 0.85751045,
      "num_input_tokens_seen": 16778285,
      "step": 786,
      "time_per_iteration": 2.610990524291992
    },
    {
      "auxiliary_loss_clip": 0.012407,
      "auxiliary_loss_mlp": 0.01032381,
      "balance_loss_clip": 1.05088782,
      "balance_loss_mlp": 1.02153313,
      "epoch": 0.09463115493296459,
      "flos": 25368051571200.0,
      "grad_norm": 4.03881573591731,
      "language_loss": 0.75053966,
      "learning_rate": 3.956413890363943e-06,
      "loss": 0.77327049,
      "num_input_tokens_seen": 16795265,
      "step": 787,
      "time_per_iteration": 2.7363622188568115
    },
    {
      "auxiliary_loss_clip": 0.01243949,
      "auxiliary_loss_mlp": 0.01035345,
      "balance_loss_clip": 1.06045926,
      "balance_loss_mlp": 1.02520061,
      "epoch": 0.09475139782360369,
      "flos": 10123254869760.0,
      "grad_norm": 2.062954591318327,
      "language_loss": 0.82023925,
      "learning_rate": 3.956252002299312e-06,
      "loss": 0.84303224,
      "num_input_tokens_seen": 16811165,
      "step": 788,
      "time_per_iteration": 2.621439218521118
    },
    {
      "auxiliary_loss_clip": 0.0124225,
      "auxiliary_loss_mlp": 0.01025697,
      "balance_loss_clip": 1.06066799,
      "balance_loss_mlp": 1.01503921,
      "epoch": 0.09487164071424277,
      "flos": 17231088936960.0,
      "grad_norm": 2.373193464864576,
      "language_loss": 0.9090355,
      "learning_rate": 3.956089817472807e-06,
      "loss": 0.93171495,
      "num_input_tokens_seen": 16828470,
      "step": 789,
      "time_per_iteration": 2.6359047889709473
    },
    {
      "auxiliary_loss_clip": 0.01247378,
      "auxiliary_loss_mlp": 0.0103848,
      "balance_loss_clip": 1.06374931,
      "balance_loss_mlp": 1.02816856,
      "epoch": 0.09499188360488187,
      "flos": 30849564528000.0,
      "grad_norm": 5.950812243738825,
      "language_loss": 0.85648715,
      "learning_rate": 3.955927335909032e-06,
      "loss": 0.87934577,
      "num_input_tokens_seen": 16851680,
      "step": 790,
      "time_per_iteration": 2.7267038822174072
    },
    {
      "auxiliary_loss_clip": 0.01249067,
      "auxiliary_loss_mlp": 0.01040782,
      "balance_loss_clip": 1.06246924,
      "balance_loss_mlp": 1.03004766,
      "epoch": 0.09511212649552095,
      "flos": 29351694453120.0,
      "grad_norm": 2.6420142971174485,
      "language_loss": 0.7594344,
      "learning_rate": 3.955764557632634e-06,
      "loss": 0.7823329,
      "num_input_tokens_seen": 16871490,
      "step": 791,
      "time_per_iteration": 2.8421366214752197
    },
    {
      "auxiliary_loss_clip": 0.01242054,
      "auxiliary_loss_mlp": 0.01031583,
      "balance_loss_clip": 1.05909514,
      "balance_loss_mlp": 1.02111602,
      "epoch": 0.09523236938616005,
      "flos": 10378687461120.0,
      "grad_norm": 2.6347142204876555,
      "language_loss": 0.94666016,
      "learning_rate": 3.955601482668309e-06,
      "loss": 0.96939653,
      "num_input_tokens_seen": 16889350,
      "step": 792,
      "time_per_iteration": 2.7404191493988037
    },
    {
      "auxiliary_loss_clip": 0.01244179,
      "auxiliary_loss_mlp": 0.01036053,
      "balance_loss_clip": 1.0521251,
      "balance_loss_mlp": 1.02388191,
      "epoch": 0.09535261227679913,
      "flos": 19061815368960.0,
      "grad_norm": 2.529608082400478,
      "language_loss": 0.88911396,
      "learning_rate": 3.955438111040794e-06,
      "loss": 0.91191626,
      "num_input_tokens_seen": 16907625,
      "step": 793,
      "time_per_iteration": 2.665135622024536
    },
    {
      "auxiliary_loss_clip": 0.01244384,
      "auxiliary_loss_mlp": 0.01034854,
      "balance_loss_clip": 1.05372691,
      "balance_loss_mlp": 1.0244348,
      "epoch": 0.09547285516743823,
      "flos": 20922993555840.0,
      "grad_norm": 2.0647924313322887,
      "language_loss": 0.80476701,
      "learning_rate": 3.955274442774873e-06,
      "loss": 0.82755935,
      "num_input_tokens_seen": 16926205,
      "step": 794,
      "time_per_iteration": 3.7187790870666504
    },
    {
      "auxiliary_loss_clip": 0.01244658,
      "auxiliary_loss_mlp": 0.01037057,
      "balance_loss_clip": 1.06088114,
      "balance_loss_mlp": 1.02579165,
      "epoch": 0.09559309805807732,
      "flos": 30154405639680.0,
      "grad_norm": 2.707123102691156,
      "language_loss": 0.71068978,
      "learning_rate": 3.9551104778953725e-06,
      "loss": 0.73350692,
      "num_input_tokens_seen": 16946500,
      "step": 795,
      "time_per_iteration": 2.76977276802063
    },
    {
      "auxiliary_loss_clip": 0.01247278,
      "auxiliary_loss_mlp": 0.01028776,
      "balance_loss_clip": 1.05617416,
      "balance_loss_mlp": 1.01845276,
      "epoch": 0.0957133409487164,
      "flos": 21066743784960.0,
      "grad_norm": 2.4020289790225906,
      "language_loss": 0.85716128,
      "learning_rate": 3.954946216427167e-06,
      "loss": 0.87992185,
      "num_input_tokens_seen": 16966960,
      "step": 796,
      "time_per_iteration": 2.7731268405914307
    },
    {
      "auxiliary_loss_clip": 0.01166911,
      "auxiliary_loss_mlp": 0.01005977,
      "balance_loss_clip": 1.04008079,
      "balance_loss_mlp": 1.00094688,
      "epoch": 0.0958335838393555,
      "flos": 71297979315840.0,
      "grad_norm": 0.8753993646328773,
      "language_loss": 0.61528456,
      "learning_rate": 3.954781658395176e-06,
      "loss": 0.63701344,
      "num_input_tokens_seen": 17023215,
      "step": 797,
      "time_per_iteration": 4.058869123458862
    },
    {
      "auxiliary_loss_clip": 0.01251395,
      "auxiliary_loss_mlp": 0.01033023,
      "balance_loss_clip": 1.06075311,
      "balance_loss_mlp": 1.02169776,
      "epoch": 0.09595382672999458,
      "flos": 21872974504320.0,
      "grad_norm": 2.470407026214291,
      "language_loss": 0.92403901,
      "learning_rate": 3.95461680382436e-06,
      "loss": 0.94688326,
      "num_input_tokens_seen": 17042140,
      "step": 798,
      "time_per_iteration": 2.679316759109497
    },
    {
      "auxiliary_loss_clip": 0.01248044,
      "auxiliary_loss_mlp": 0.01036525,
      "balance_loss_clip": 1.06313241,
      "balance_loss_mlp": 1.02518845,
      "epoch": 0.09607406962063368,
      "flos": 18695562341760.0,
      "grad_norm": 2.706865787072605,
      "language_loss": 0.86364955,
      "learning_rate": 3.9544516527397295e-06,
      "loss": 0.88649523,
      "num_input_tokens_seen": 17058490,
      "step": 799,
      "time_per_iteration": 4.483951568603516
    },
    {
      "auxiliary_loss_clip": 0.01231199,
      "auxiliary_loss_mlp": 0.01033524,
      "balance_loss_clip": 1.05823898,
      "balance_loss_mlp": 1.0230751,
      "epoch": 0.09619431251127276,
      "flos": 22568456615040.0,
      "grad_norm": 2.217113692905276,
      "language_loss": 0.80694914,
      "learning_rate": 3.954286205166338e-06,
      "loss": 0.8295964,
      "num_input_tokens_seen": 17079655,
      "step": 800,
      "time_per_iteration": 2.7520227432250977
    },
    {
      "auxiliary_loss_clip": 0.01251813,
      "auxiliary_loss_mlp": 0.01035528,
      "balance_loss_clip": 1.06658018,
      "balance_loss_mlp": 1.02384591,
      "epoch": 0.09631455540191186,
      "flos": 14246230608000.0,
      "grad_norm": 2.141709959015025,
      "language_loss": 0.83882463,
      "learning_rate": 3.954120461129282e-06,
      "loss": 0.86169803,
      "num_input_tokens_seen": 17097065,
      "step": 801,
      "time_per_iteration": 2.629415273666382
    },
    {
      "auxiliary_loss_clip": 0.01247454,
      "auxiliary_loss_mlp": 0.01038213,
      "balance_loss_clip": 1.06689906,
      "balance_loss_mlp": 1.02774656,
      "epoch": 0.09643479829255096,
      "flos": 20740387789440.0,
      "grad_norm": 2.6887988586189238,
      "language_loss": 0.83762366,
      "learning_rate": 3.953954420653706e-06,
      "loss": 0.86048037,
      "num_input_tokens_seen": 17114090,
      "step": 802,
      "time_per_iteration": 2.6647555828094482
    },
    {
      "auxiliary_loss_clip": 0.01244088,
      "auxiliary_loss_mlp": 0.01035805,
      "balance_loss_clip": 1.06356514,
      "balance_loss_mlp": 1.02555943,
      "epoch": 0.09655504118319004,
      "flos": 24420476833920.0,
      "grad_norm": 2.8372356764106774,
      "language_loss": 0.87976348,
      "learning_rate": 3.953788083764798e-06,
      "loss": 0.90256244,
      "num_input_tokens_seen": 17133325,
      "step": 803,
      "time_per_iteration": 2.6208879947662354
    },
    {
      "auxiliary_loss_clip": 0.01248191,
      "auxiliary_loss_mlp": 0.01038213,
      "balance_loss_clip": 1.05709028,
      "balance_loss_mlp": 1.02749562,
      "epoch": 0.09667528407382914,
      "flos": 18441961344000.0,
      "grad_norm": 2.8196397480711273,
      "language_loss": 0.92088479,
      "learning_rate": 3.953621450487792e-06,
      "loss": 0.94374883,
      "num_input_tokens_seen": 17151945,
      "step": 804,
      "time_per_iteration": 2.7971975803375244
    },
    {
      "auxiliary_loss_clip": 0.01164046,
      "auxiliary_loss_mlp": 0.0100599,
      "balance_loss_clip": 1.04821467,
      "balance_loss_mlp": 1.00165081,
      "epoch": 0.09679552696446822,
      "flos": 70816455544320.0,
      "grad_norm": 0.8400571731083173,
      "language_loss": 0.61142415,
      "learning_rate": 3.953454520847964e-06,
      "loss": 0.63312453,
      "num_input_tokens_seen": 17216790,
      "step": 805,
      "time_per_iteration": 3.3890163898468018
    },
    {
      "auxiliary_loss_clip": 0.01243328,
      "auxiliary_loss_mlp": 0.01038656,
      "balance_loss_clip": 1.05995989,
      "balance_loss_mlp": 1.02558434,
      "epoch": 0.09691576985510732,
      "flos": 21945514020480.0,
      "grad_norm": 2.8813133042728136,
      "language_loss": 0.73572803,
      "learning_rate": 3.9532872948706395e-06,
      "loss": 0.7585479,
      "num_input_tokens_seen": 17236285,
      "step": 806,
      "time_per_iteration": 2.7489383220672607
    },
    {
      "auxiliary_loss_clip": 0.01247697,
      "auxiliary_loss_mlp": 0.01041157,
      "balance_loss_clip": 1.06127942,
      "balance_loss_mlp": 1.0294981,
      "epoch": 0.09703601274574641,
      "flos": 17965211103360.0,
      "grad_norm": 2.8841082272127703,
      "language_loss": 0.82868886,
      "learning_rate": 3.9531197725811845e-06,
      "loss": 0.8515774,
      "num_input_tokens_seen": 17251670,
      "step": 807,
      "time_per_iteration": 2.7216203212738037
    },
    {
      "auxiliary_loss_clip": 0.01245558,
      "auxiliary_loss_mlp": 0.01039037,
      "balance_loss_clip": 1.06817007,
      "balance_loss_mlp": 1.02814162,
      "epoch": 0.0971562556363855,
      "flos": 22162162901760.0,
      "grad_norm": 2.873600252546435,
      "language_loss": 0.88036788,
      "learning_rate": 3.952951954005013e-06,
      "loss": 0.90321386,
      "num_input_tokens_seen": 17271355,
      "step": 808,
      "time_per_iteration": 2.6474361419677734
    },
    {
      "auxiliary_loss_clip": 0.01244347,
      "auxiliary_loss_mlp": 0.01032855,
      "balance_loss_clip": 1.05756509,
      "balance_loss_mlp": 1.02263236,
      "epoch": 0.0972764985270246,
      "flos": 25848716394240.0,
      "grad_norm": 1.7977499500684682,
      "language_loss": 0.848194,
      "learning_rate": 3.952783839167584e-06,
      "loss": 0.87096608,
      "num_input_tokens_seen": 17291400,
      "step": 809,
      "time_per_iteration": 2.8147835731506348
    },
    {
      "auxiliary_loss_clip": 0.01243238,
      "auxiliary_loss_mlp": 0.01041212,
      "balance_loss_clip": 1.06106102,
      "balance_loss_mlp": 1.03001785,
      "epoch": 0.09739674141766368,
      "flos": 20339373375360.0,
      "grad_norm": 2.970629154899451,
      "language_loss": 0.74529469,
      "learning_rate": 3.952615428094398e-06,
      "loss": 0.76813912,
      "num_input_tokens_seen": 17310920,
      "step": 810,
      "time_per_iteration": 2.6720035076141357
    },
    {
      "auxiliary_loss_clip": 0.01240877,
      "auxiliary_loss_mlp": 0.01036353,
      "balance_loss_clip": 1.05361104,
      "balance_loss_mlp": 1.02570736,
      "epoch": 0.09751698430830277,
      "flos": 15743059188480.0,
      "grad_norm": 2.187025644476033,
      "language_loss": 0.73719263,
      "learning_rate": 3.952446720811004e-06,
      "loss": 0.75996494,
      "num_input_tokens_seen": 17329245,
      "step": 811,
      "time_per_iteration": 2.767745018005371
    },
    {
      "auxiliary_loss_clip": 0.01170407,
      "auxiliary_loss_mlp": 0.01011249,
      "balance_loss_clip": 1.04175198,
      "balance_loss_mlp": 1.00545585,
      "epoch": 0.09763722719894186,
      "flos": 63716806800000.0,
      "grad_norm": 0.8453187610369417,
      "language_loss": 0.63608384,
      "learning_rate": 3.952277717342995e-06,
      "loss": 0.65790045,
      "num_input_tokens_seen": 17395680,
      "step": 812,
      "time_per_iteration": 3.333613395690918
    },
    {
      "auxiliary_loss_clip": 0.01250556,
      "auxiliary_loss_mlp": 0.01039421,
      "balance_loss_clip": 1.06145048,
      "balance_loss_mlp": 1.02817333,
      "epoch": 0.09775747008958095,
      "flos": 22090916275200.0,
      "grad_norm": 3.437220036902223,
      "language_loss": 0.85864651,
      "learning_rate": 3.952108417716009e-06,
      "loss": 0.88154626,
      "num_input_tokens_seen": 17415135,
      "step": 813,
      "time_per_iteration": 2.674899101257324
    },
    {
      "auxiliary_loss_clip": 0.01250698,
      "auxiliary_loss_mlp": 0.0103366,
      "balance_loss_clip": 1.06802297,
      "balance_loss_mlp": 1.02236533,
      "epoch": 0.09787771298022005,
      "flos": 21286050272640.0,
      "grad_norm": 2.7514806700764605,
      "language_loss": 0.8493337,
      "learning_rate": 3.951938821955727e-06,
      "loss": 0.87217736,
      "num_input_tokens_seen": 17434535,
      "step": 814,
      "time_per_iteration": 2.6828436851501465
    },
    {
      "auxiliary_loss_clip": 0.01247008,
      "auxiliary_loss_mlp": 0.0104396,
      "balance_loss_clip": 1.06321764,
      "balance_loss_mlp": 1.03203893,
      "epoch": 0.09799795587085913,
      "flos": 22054574689920.0,
      "grad_norm": 3.4304501329058406,
      "language_loss": 0.76736152,
      "learning_rate": 3.9517689300878786e-06,
      "loss": 0.79027128,
      "num_input_tokens_seen": 17454270,
      "step": 815,
      "time_per_iteration": 2.6791090965270996
    },
    {
      "auxiliary_loss_clip": 0.01241663,
      "auxiliary_loss_mlp": 0.01038039,
      "balance_loss_clip": 1.06260586,
      "balance_loss_mlp": 1.0268209,
      "epoch": 0.09811819876149823,
      "flos": 22163743100160.0,
      "grad_norm": 2.2612517586885006,
      "language_loss": 0.78953046,
      "learning_rate": 3.951598742138236e-06,
      "loss": 0.81232744,
      "num_input_tokens_seen": 17472995,
      "step": 816,
      "time_per_iteration": 2.6254987716674805
    },
    {
      "auxiliary_loss_clip": 0.01251727,
      "auxiliary_loss_mlp": 0.01038625,
      "balance_loss_clip": 1.05839598,
      "balance_loss_mlp": 1.02793789,
      "epoch": 0.09823844165213731,
      "flos": 22231111057920.0,
      "grad_norm": 2.8960384946340763,
      "language_loss": 0.79975426,
      "learning_rate": 3.951428258132615e-06,
      "loss": 0.82265776,
      "num_input_tokens_seen": 17491115,
      "step": 817,
      "time_per_iteration": 2.7323713302612305
    },
    {
      "auxiliary_loss_clip": 0.01248699,
      "auxiliary_loss_mlp": 0.01032693,
      "balance_loss_clip": 1.06387234,
      "balance_loss_mlp": 1.02181494,
      "epoch": 0.09835868454277641,
      "flos": 22487728798080.0,
      "grad_norm": 2.8660031520149194,
      "language_loss": 0.84536862,
      "learning_rate": 3.951257478096879e-06,
      "loss": 0.86818254,
      "num_input_tokens_seen": 17509480,
      "step": 818,
      "time_per_iteration": 2.723167657852173
    },
    {
      "auxiliary_loss_clip": 0.01251199,
      "auxiliary_loss_mlp": 0.01209433,
      "balance_loss_clip": 1.0638659,
      "balance_loss_mlp": 1.00087285,
      "epoch": 0.0984789274334155,
      "flos": 16362554077440.0,
      "grad_norm": 6.035182411531741,
      "language_loss": 0.68807811,
      "learning_rate": 3.951086402056936e-06,
      "loss": 0.71268439,
      "num_input_tokens_seen": 17524080,
      "step": 819,
      "time_per_iteration": 2.6598899364471436
    },
    {
      "auxiliary_loss_clip": 0.01242196,
      "auxiliary_loss_mlp": 0.01209246,
      "balance_loss_clip": 1.05340266,
      "balance_loss_mlp": 1.00089288,
      "epoch": 0.09859917032405459,
      "flos": 24243545416320.0,
      "grad_norm": 2.161245089615326,
      "language_loss": 0.83800083,
      "learning_rate": 3.950915030038735e-06,
      "loss": 0.86251527,
      "num_input_tokens_seen": 17543875,
      "step": 820,
      "time_per_iteration": 2.8255205154418945
    },
    {
      "auxiliary_loss_clip": 0.01244637,
      "auxiliary_loss_mlp": 0.01039411,
      "balance_loss_clip": 1.06615901,
      "balance_loss_mlp": 1.02841353,
      "epoch": 0.09871941321469369,
      "flos": 17420195064960.0,
      "grad_norm": 4.285025797274739,
      "language_loss": 0.83457398,
      "learning_rate": 3.9507433620682765e-06,
      "loss": 0.85741448,
      "num_input_tokens_seen": 17560810,
      "step": 821,
      "time_per_iteration": 3.660824775695801
    },
    {
      "auxiliary_loss_clip": 0.01246835,
      "auxiliary_loss_mlp": 0.01035657,
      "balance_loss_clip": 1.05851007,
      "balance_loss_mlp": 1.02421331,
      "epoch": 0.09883965610533277,
      "flos": 28477341590400.0,
      "grad_norm": 1.8939294150934038,
      "language_loss": 0.88177621,
      "learning_rate": 3.9505713981716e-06,
      "loss": 0.9046011,
      "num_input_tokens_seen": 17583640,
      "step": 822,
      "time_per_iteration": 2.810194969177246
    },
    {
      "auxiliary_loss_clip": 0.01244359,
      "auxiliary_loss_mlp": 0.01038408,
      "balance_loss_clip": 1.06364131,
      "balance_loss_mlp": 1.02812052,
      "epoch": 0.09895989899597187,
      "flos": 23693932437120.0,
      "grad_norm": 2.2193672661106074,
      "language_loss": 0.8128562,
      "learning_rate": 3.950399138374795e-06,
      "loss": 0.83568388,
      "num_input_tokens_seen": 17602720,
      "step": 823,
      "time_per_iteration": 2.7560205459594727
    },
    {
      "auxiliary_loss_clip": 0.01245457,
      "auxiliary_loss_mlp": 0.01045988,
      "balance_loss_clip": 1.06315303,
      "balance_loss_mlp": 1.03408504,
      "epoch": 0.09908014188661095,
      "flos": 24679608526080.0,
      "grad_norm": 1.9344099141988493,
      "language_loss": 0.74350035,
      "learning_rate": 3.95022658270399e-06,
      "loss": 0.76641482,
      "num_input_tokens_seen": 17623085,
      "step": 824,
      "time_per_iteration": 3.5081846714019775
    },
    {
      "auxiliary_loss_clip": 0.01248674,
      "auxiliary_loss_mlp": 0.0104117,
      "balance_loss_clip": 1.06694913,
      "balance_loss_mlp": 1.03016067,
      "epoch": 0.09920038477725004,
      "flos": 14064307200000.0,
      "grad_norm": 2.136220655798877,
      "language_loss": 0.7799058,
      "learning_rate": 3.9500537311853635e-06,
      "loss": 0.80280423,
      "num_input_tokens_seen": 17641040,
      "step": 825,
      "time_per_iteration": 3.6025304794311523
    },
    {
      "auxiliary_loss_clip": 0.01243287,
      "auxiliary_loss_mlp": 0.01037809,
      "balance_loss_clip": 1.05871952,
      "balance_loss_mlp": 1.02593029,
      "epoch": 0.09932062766788914,
      "flos": 13407070095360.0,
      "grad_norm": 3.2755524359356345,
      "language_loss": 0.83181882,
      "learning_rate": 3.949880583845136e-06,
      "loss": 0.85462976,
      "num_input_tokens_seen": 17659115,
      "step": 826,
      "time_per_iteration": 3.6025211811065674
    },
    {
      "auxiliary_loss_clip": 0.01246196,
      "auxiliary_loss_mlp": 0.01032981,
      "balance_loss_clip": 1.06103873,
      "balance_loss_mlp": 1.02179933,
      "epoch": 0.09944087055852822,
      "flos": 19500751566720.0,
      "grad_norm": 2.800156320171497,
      "language_loss": 0.81102234,
      "learning_rate": 3.949707140709575e-06,
      "loss": 0.83381414,
      "num_input_tokens_seen": 17678845,
      "step": 827,
      "time_per_iteration": 2.681673526763916
    },
    {
      "auxiliary_loss_clip": 0.01246776,
      "auxiliary_loss_mlp": 0.01036836,
      "balance_loss_clip": 1.06103945,
      "balance_loss_mlp": 1.02484298,
      "epoch": 0.09956111344916732,
      "flos": 17749100926080.0,
      "grad_norm": 2.676751837897441,
      "language_loss": 0.83516669,
      "learning_rate": 3.949533401804991e-06,
      "loss": 0.85800278,
      "num_input_tokens_seen": 17695750,
      "step": 828,
      "time_per_iteration": 2.831484079360962
    },
    {
      "auxiliary_loss_clip": 0.01244313,
      "auxiliary_loss_mlp": 0.01209833,
      "balance_loss_clip": 1.06350458,
      "balance_loss_mlp": 1.00085688,
      "epoch": 0.0996813563398064,
      "flos": 17967581400960.0,
      "grad_norm": 2.579608801190047,
      "language_loss": 0.9091084,
      "learning_rate": 3.949359367157739e-06,
      "loss": 0.9336499,
      "num_input_tokens_seen": 17714445,
      "step": 829,
      "time_per_iteration": 2.672245502471924
    },
    {
      "auxiliary_loss_clip": 0.01251763,
      "auxiliary_loss_mlp": 0.01035983,
      "balance_loss_clip": 1.06640196,
      "balance_loss_mlp": 1.0240202,
      "epoch": 0.0998015992304455,
      "flos": 17457039440640.0,
      "grad_norm": 2.930116266501944,
      "language_loss": 0.75557131,
      "learning_rate": 3.949185036794222e-06,
      "loss": 0.77844882,
      "num_input_tokens_seen": 17732455,
      "step": 830,
      "time_per_iteration": 2.6820476055145264
    },
    {
      "auxiliary_loss_clip": 0.01244052,
      "auxiliary_loss_mlp": 0.01040831,
      "balance_loss_clip": 1.06552792,
      "balance_loss_mlp": 1.03021574,
      "epoch": 0.0999218421210846,
      "flos": 25888757080320.0,
      "grad_norm": 1.796983023658945,
      "language_loss": 0.79131001,
      "learning_rate": 3.949010410740884e-06,
      "loss": 0.81415892,
      "num_input_tokens_seen": 17755280,
      "step": 831,
      "time_per_iteration": 2.7196433544158936
    },
    {
      "auxiliary_loss_clip": 0.01238975,
      "auxiliary_loss_mlp": 0.01209635,
      "balance_loss_clip": 1.05968189,
      "balance_loss_mlp": 1.00087798,
      "epoch": 0.10004208501172368,
      "flos": 21215916967680.0,
      "grad_norm": 2.471287729033549,
      "language_loss": 0.86601472,
      "learning_rate": 3.948835489024216e-06,
      "loss": 0.89050084,
      "num_input_tokens_seen": 17775015,
      "step": 832,
      "time_per_iteration": 2.6995670795440674
    },
    {
      "auxiliary_loss_clip": 0.01249054,
      "auxiliary_loss_mlp": 0.01035374,
      "balance_loss_clip": 1.06340909,
      "balance_loss_mlp": 1.024526,
      "epoch": 0.10016232790236278,
      "flos": 17348409734400.0,
      "grad_norm": 4.740946240025885,
      "language_loss": 0.90335768,
      "learning_rate": 3.948660271670755e-06,
      "loss": 0.92620194,
      "num_input_tokens_seen": 17792165,
      "step": 833,
      "time_per_iteration": 2.6667604446411133
    },
    {
      "auxiliary_loss_clip": 0.01243551,
      "auxiliary_loss_mlp": 0.01037767,
      "balance_loss_clip": 1.06130433,
      "balance_loss_mlp": 1.02657294,
      "epoch": 0.10028257079300186,
      "flos": 25666541591040.0,
      "grad_norm": 2.7229120404202822,
      "language_loss": 0.84284663,
      "learning_rate": 3.948484758707079e-06,
      "loss": 0.86565983,
      "num_input_tokens_seen": 17811765,
      "step": 834,
      "time_per_iteration": 2.727243661880493
    },
    {
      "auxiliary_loss_clip": 0.01242564,
      "auxiliary_loss_mlp": 0.01038598,
      "balance_loss_clip": 1.05868125,
      "balance_loss_mlp": 1.02611709,
      "epoch": 0.10040281368364096,
      "flos": 25156035544320.0,
      "grad_norm": 2.6000849723402157,
      "language_loss": 0.8386941,
      "learning_rate": 3.948308950159815e-06,
      "loss": 0.86150569,
      "num_input_tokens_seen": 17830445,
      "step": 835,
      "time_per_iteration": 2.789336681365967
    },
    {
      "auxiliary_loss_clip": 0.01245537,
      "auxiliary_loss_mlp": 0.0104147,
      "balance_loss_clip": 1.0578537,
      "balance_loss_mlp": 1.02827978,
      "epoch": 0.10052305657428004,
      "flos": 17603303621760.0,
      "grad_norm": 2.8235186272506807,
      "language_loss": 0.7571131,
      "learning_rate": 3.9481328460556326e-06,
      "loss": 0.7799831,
      "num_input_tokens_seen": 17847665,
      "step": 836,
      "time_per_iteration": 2.7647945880889893
    },
    {
      "auxiliary_loss_clip": 0.01241454,
      "auxiliary_loss_mlp": 0.01031247,
      "balance_loss_clip": 1.06061995,
      "balance_loss_mlp": 1.01982117,
      "epoch": 0.10064329946491914,
      "flos": 18660154510080.0,
      "grad_norm": 2.201354454068575,
      "language_loss": 0.89793026,
      "learning_rate": 3.9479564464212455e-06,
      "loss": 0.92065728,
      "num_input_tokens_seen": 17866825,
      "step": 837,
      "time_per_iteration": 2.7320384979248047
    },
    {
      "auxiliary_loss_clip": 0.01252839,
      "auxiliary_loss_mlp": 0.01041517,
      "balance_loss_clip": 1.0667243,
      "balance_loss_mlp": 1.03000176,
      "epoch": 0.10076354235555823,
      "flos": 17199056983680.0,
      "grad_norm": 3.561529708945354,
      "language_loss": 0.7632966,
      "learning_rate": 3.947779751283414e-06,
      "loss": 0.7862401,
      "num_input_tokens_seen": 17883995,
      "step": 838,
      "time_per_iteration": 2.582597017288208
    },
    {
      "auxiliary_loss_clip": 0.01252923,
      "auxiliary_loss_mlp": 0.01210593,
      "balance_loss_clip": 1.07219124,
      "balance_loss_mlp": 1.00076294,
      "epoch": 0.10088378524619732,
      "flos": 22962252395520.0,
      "grad_norm": 3.750774537142868,
      "language_loss": 0.75949037,
      "learning_rate": 3.947602760668944e-06,
      "loss": 0.78412557,
      "num_input_tokens_seen": 17903785,
      "step": 839,
      "time_per_iteration": 2.690537214279175
    },
    {
      "auxiliary_loss_clip": 0.01249352,
      "auxiliary_loss_mlp": 0.01041694,
      "balance_loss_clip": 1.06703687,
      "balance_loss_mlp": 1.02945113,
      "epoch": 0.10100402813683641,
      "flos": 37885828746240.0,
      "grad_norm": 2.363695812724809,
      "language_loss": 0.71622562,
      "learning_rate": 3.947425474604684e-06,
      "loss": 0.7391361,
      "num_input_tokens_seen": 17927720,
      "step": 840,
      "time_per_iteration": 2.8664987087249756
    },
    {
      "auxiliary_loss_clip": 0.01246522,
      "auxiliary_loss_mlp": 0.01041717,
      "balance_loss_clip": 1.05947661,
      "balance_loss_mlp": 1.0302788,
      "epoch": 0.1011242710274755,
      "flos": 21543458112000.0,
      "grad_norm": 2.171113935934673,
      "language_loss": 0.92210031,
      "learning_rate": 3.947247893117528e-06,
      "loss": 0.94498265,
      "num_input_tokens_seen": 17946225,
      "step": 841,
      "time_per_iteration": 2.6892220973968506
    },
    {
      "auxiliary_loss_clip": 0.01245175,
      "auxiliary_loss_mlp": 0.01042574,
      "balance_loss_clip": 1.06268322,
      "balance_loss_mlp": 1.03084409,
      "epoch": 0.10124451391811459,
      "flos": 13621456419840.0,
      "grad_norm": 3.2010381506160925,
      "language_loss": 0.69587392,
      "learning_rate": 3.947070016234413e-06,
      "loss": 0.71875137,
      "num_input_tokens_seen": 17962015,
      "step": 842,
      "time_per_iteration": 2.6268422603607178
    },
    {
      "auxiliary_loss_clip": 0.01261636,
      "auxiliary_loss_mlp": 0.01035871,
      "balance_loss_clip": 1.06686831,
      "balance_loss_mlp": 1.02378941,
      "epoch": 0.10136475680875369,
      "flos": 16649228522880.0,
      "grad_norm": 2.58338289517023,
      "language_loss": 0.75077379,
      "learning_rate": 3.946891843982326e-06,
      "loss": 0.77374887,
      "num_input_tokens_seen": 17979680,
      "step": 843,
      "time_per_iteration": 2.602454900741577
    },
    {
      "auxiliary_loss_clip": 0.0124824,
      "auxiliary_loss_mlp": 0.01036821,
      "balance_loss_clip": 1.0649395,
      "balance_loss_mlp": 1.02470887,
      "epoch": 0.10148499969939277,
      "flos": 19461034103040.0,
      "grad_norm": 2.499449446174017,
      "language_loss": 0.74778116,
      "learning_rate": 3.9467133763882935e-06,
      "loss": 0.77063179,
      "num_input_tokens_seen": 17998145,
      "step": 844,
      "time_per_iteration": 2.6853151321411133
    },
    {
      "auxiliary_loss_clip": 0.01236985,
      "auxiliary_loss_mlp": 0.01042526,
      "balance_loss_clip": 1.06245494,
      "balance_loss_mlp": 1.03055108,
      "epoch": 0.10160524259003187,
      "flos": 21104988791040.0,
      "grad_norm": 3.9455222829505967,
      "language_loss": 0.86460268,
      "learning_rate": 3.9465346134793905e-06,
      "loss": 0.88739783,
      "num_input_tokens_seen": 18017955,
      "step": 845,
      "time_per_iteration": 2.6487910747528076
    },
    {
      "auxiliary_loss_clip": 0.01251965,
      "auxiliary_loss_mlp": 0.01035539,
      "balance_loss_clip": 1.06291986,
      "balance_loss_mlp": 1.02427971,
      "epoch": 0.10172548548067095,
      "flos": 17712687513600.0,
      "grad_norm": 2.1661211936597917,
      "language_loss": 0.79676199,
      "learning_rate": 3.9463555552827335e-06,
      "loss": 0.81963706,
      "num_input_tokens_seen": 18035125,
      "step": 846,
      "time_per_iteration": 2.6926681995391846
    },
    {
      "auxiliary_loss_clip": 0.01237186,
      "auxiliary_loss_mlp": 0.01041213,
      "balance_loss_clip": 1.06284785,
      "balance_loss_mlp": 1.02957249,
      "epoch": 0.10184572837131005,
      "flos": 21104845136640.0,
      "grad_norm": 3.2917071111654526,
      "language_loss": 0.86220527,
      "learning_rate": 3.946176201825487e-06,
      "loss": 0.8849892,
      "num_input_tokens_seen": 18053160,
      "step": 847,
      "time_per_iteration": 2.6727805137634277
    },
    {
      "auxiliary_loss_clip": 0.01252361,
      "auxiliary_loss_mlp": 0.01037904,
      "balance_loss_clip": 1.0666492,
      "balance_loss_mlp": 1.0265131,
      "epoch": 0.10196597126194913,
      "flos": 26067591918720.0,
      "grad_norm": 2.1326441966253347,
      "language_loss": 0.83591133,
      "learning_rate": 3.9459965531348575e-06,
      "loss": 0.858814,
      "num_input_tokens_seen": 18072815,
      "step": 848,
      "time_per_iteration": 3.7301788330078125
    },
    {
      "auxiliary_loss_clip": 0.01250815,
      "auxiliary_loss_mlp": 0.01210215,
      "balance_loss_clip": 1.06405902,
      "balance_loss_mlp": 1.00082827,
      "epoch": 0.10208621415258823,
      "flos": 29314634595840.0,
      "grad_norm": 2.1353435530369547,
      "language_loss": 0.85615158,
      "learning_rate": 3.945816609238098e-06,
      "loss": 0.88076186,
      "num_input_tokens_seen": 18092225,
      "step": 849,
      "time_per_iteration": 2.722771167755127
    },
    {
      "auxiliary_loss_clip": 0.01245012,
      "auxiliary_loss_mlp": 0.01042749,
      "balance_loss_clip": 1.06187189,
      "balance_loss_mlp": 1.0308578,
      "epoch": 0.10220645704322733,
      "flos": 23805794367360.0,
      "grad_norm": 2.732835992867475,
      "language_loss": 0.85010678,
      "learning_rate": 3.945636370162507e-06,
      "loss": 0.87298441,
      "num_input_tokens_seen": 18112335,
      "step": 850,
      "time_per_iteration": 3.6816413402557373
    },
    {
      "auxiliary_loss_clip": 0.01244941,
      "auxiliary_loss_mlp": 0.01037977,
      "balance_loss_clip": 1.06505394,
      "balance_loss_mlp": 1.02714682,
      "epoch": 0.10232669993386641,
      "flos": 23218546913280.0,
      "grad_norm": 2.2181926059830563,
      "language_loss": 0.79054767,
      "learning_rate": 3.945455835935425e-06,
      "loss": 0.81337684,
      "num_input_tokens_seen": 18131520,
      "step": 851,
      "time_per_iteration": 2.7151520252227783
    },
    {
      "auxiliary_loss_clip": 0.01252036,
      "auxiliary_loss_mlp": 0.01039355,
      "balance_loss_clip": 1.06345725,
      "balance_loss_mlp": 1.02813768,
      "epoch": 0.1024469428245055,
      "flos": 22922929981440.0,
      "grad_norm": 2.5651217059199642,
      "language_loss": 0.75137603,
      "learning_rate": 3.94527500658424e-06,
      "loss": 0.77428997,
      "num_input_tokens_seen": 18149185,
      "step": 852,
      "time_per_iteration": 3.577005624771118
    },
    {
      "auxiliary_loss_clip": 0.01253841,
      "auxiliary_loss_mlp": 0.01036999,
      "balance_loss_clip": 1.06176734,
      "balance_loss_mlp": 1.02625799,
      "epoch": 0.10256718571514459,
      "flos": 31359495957120.0,
      "grad_norm": 2.794128631363122,
      "language_loss": 0.81261802,
      "learning_rate": 3.945093882136382e-06,
      "loss": 0.83552647,
      "num_input_tokens_seen": 18172960,
      "step": 853,
      "time_per_iteration": 3.681680917739868
    },
    {
      "auxiliary_loss_clip": 0.01244503,
      "auxiliary_loss_mlp": 0.01208655,
      "balance_loss_clip": 1.06328511,
      "balance_loss_mlp": 1.00077641,
      "epoch": 0.10268742860578368,
      "flos": 23474877344640.0,
      "grad_norm": 2.1526856141419395,
      "language_loss": 0.84621489,
      "learning_rate": 3.944912462619329e-06,
      "loss": 0.87074649,
      "num_input_tokens_seen": 18191925,
      "step": 854,
      "time_per_iteration": 2.7048375606536865
    },
    {
      "auxiliary_loss_clip": 0.01251634,
      "auxiliary_loss_mlp": 0.01045596,
      "balance_loss_clip": 1.06374848,
      "balance_loss_mlp": 1.0334065,
      "epoch": 0.10280767149642277,
      "flos": 25520313323520.0,
      "grad_norm": 2.5410008244952937,
      "language_loss": 0.80889499,
      "learning_rate": 3.9447307480606025e-06,
      "loss": 0.83186734,
      "num_input_tokens_seen": 18212010,
      "step": 855,
      "time_per_iteration": 2.685891628265381
    },
    {
      "auxiliary_loss_clip": 0.01244353,
      "auxiliary_loss_mlp": 0.01038124,
      "balance_loss_clip": 1.06308389,
      "balance_loss_mlp": 1.02558947,
      "epoch": 0.10292791438706186,
      "flos": 17347691462400.0,
      "grad_norm": 2.5857550645279375,
      "language_loss": 0.90112662,
      "learning_rate": 3.944548738487767e-06,
      "loss": 0.92395139,
      "num_input_tokens_seen": 18229525,
      "step": 856,
      "time_per_iteration": 2.6716086864471436
    },
    {
      "auxiliary_loss_clip": 0.01254449,
      "auxiliary_loss_mlp": 0.01039174,
      "balance_loss_clip": 1.07138014,
      "balance_loss_mlp": 1.0283618,
      "epoch": 0.10304815727770096,
      "flos": 27052693390080.0,
      "grad_norm": 3.298314148859007,
      "language_loss": 0.90662026,
      "learning_rate": 3.944366433928434e-06,
      "loss": 0.92955655,
      "num_input_tokens_seen": 18249505,
      "step": 857,
      "time_per_iteration": 2.6502902507781982
    },
    {
      "auxiliary_loss_clip": 0.01244755,
      "auxiliary_loss_mlp": 0.01042424,
      "balance_loss_clip": 1.0612371,
      "balance_loss_mlp": 1.03109336,
      "epoch": 0.10316840016834004,
      "flos": 22782591544320.0,
      "grad_norm": 1.5721444718285182,
      "language_loss": 0.83658648,
      "learning_rate": 3.9441838344102594e-06,
      "loss": 0.85945821,
      "num_input_tokens_seen": 18269230,
      "step": 858,
      "time_per_iteration": 2.659022331237793
    },
    {
      "auxiliary_loss_clip": 0.01256921,
      "auxiliary_loss_mlp": 0.0103834,
      "balance_loss_clip": 1.06636333,
      "balance_loss_mlp": 1.02703917,
      "epoch": 0.10328864305897914,
      "flos": 20704584908160.0,
      "grad_norm": 2.760553849581445,
      "language_loss": 0.67193514,
      "learning_rate": 3.944000939960943e-06,
      "loss": 0.69488776,
      "num_input_tokens_seen": 18287955,
      "step": 859,
      "time_per_iteration": 2.642331838607788
    },
    {
      "auxiliary_loss_clip": 0.01250452,
      "auxiliary_loss_mlp": 0.01036801,
      "balance_loss_clip": 1.06237793,
      "balance_loss_mlp": 1.02625704,
      "epoch": 0.10340888594961822,
      "flos": 28478814048000.0,
      "grad_norm": 2.671337720701693,
      "language_loss": 0.80073237,
      "learning_rate": 3.943817750608229e-06,
      "loss": 0.82360482,
      "num_input_tokens_seen": 18310505,
      "step": 860,
      "time_per_iteration": 2.701404333114624
    },
    {
      "auxiliary_loss_clip": 0.01254166,
      "auxiliary_loss_mlp": 0.01034818,
      "balance_loss_clip": 1.06899977,
      "balance_loss_mlp": 1.02415466,
      "epoch": 0.10352912884025732,
      "flos": 13370333460480.0,
      "grad_norm": 2.8583309440509836,
      "language_loss": 0.82227838,
      "learning_rate": 3.943634266379908e-06,
      "loss": 0.84516823,
      "num_input_tokens_seen": 18327400,
      "step": 861,
      "time_per_iteration": 2.5969481468200684
    },
    {
      "auxiliary_loss_clip": 0.01250689,
      "auxiliary_loss_mlp": 0.01034293,
      "balance_loss_clip": 1.06380463,
      "balance_loss_mlp": 1.02313542,
      "epoch": 0.10364937173089642,
      "flos": 25558558329600.0,
      "grad_norm": 1.9174220531647939,
      "language_loss": 0.84874237,
      "learning_rate": 3.943450487303815e-06,
      "loss": 0.87159216,
      "num_input_tokens_seen": 18347895,
      "step": 862,
      "time_per_iteration": 2.710848808288574
    },
    {
      "auxiliary_loss_clip": 0.01247344,
      "auxiliary_loss_mlp": 0.01035615,
      "balance_loss_clip": 1.06597066,
      "balance_loss_mlp": 1.02426612,
      "epoch": 0.1037696146215355,
      "flos": 21215486004480.0,
      "grad_norm": 3.327128524421959,
      "language_loss": 0.85495776,
      "learning_rate": 3.943266413407827e-06,
      "loss": 0.87778735,
      "num_input_tokens_seen": 18367170,
      "step": 863,
      "time_per_iteration": 2.65531587600708
    },
    {
      "auxiliary_loss_clip": 0.01249937,
      "auxiliary_loss_mlp": 0.01040861,
      "balance_loss_clip": 1.0664798,
      "balance_loss_mlp": 1.02945828,
      "epoch": 0.1038898575121746,
      "flos": 25807382818560.0,
      "grad_norm": 3.181227685920338,
      "language_loss": 0.85070491,
      "learning_rate": 3.94308204471987e-06,
      "loss": 0.87361288,
      "num_input_tokens_seen": 18386185,
      "step": 864,
      "time_per_iteration": 2.674936532974243
    },
    {
      "auxiliary_loss_clip": 0.01252894,
      "auxiliary_loss_mlp": 0.0103093,
      "balance_loss_clip": 1.06221485,
      "balance_loss_mlp": 1.01944995,
      "epoch": 0.10401010040281368,
      "flos": 19062425900160.0,
      "grad_norm": 3.426620710230705,
      "language_loss": 0.74633867,
      "learning_rate": 3.942897381267912e-06,
      "loss": 0.7691769,
      "num_input_tokens_seen": 18402550,
      "step": 865,
      "time_per_iteration": 2.6565933227539062
    },
    {
      "auxiliary_loss_clip": 0.01254745,
      "auxiliary_loss_mlp": 0.01033111,
      "balance_loss_clip": 1.06807673,
      "balance_loss_mlp": 1.02192366,
      "epoch": 0.10413034329345278,
      "flos": 16355119962240.0,
      "grad_norm": 2.6056003905622753,
      "language_loss": 0.65884984,
      "learning_rate": 3.942712423079965e-06,
      "loss": 0.68172842,
      "num_input_tokens_seen": 18418940,
      "step": 866,
      "time_per_iteration": 2.604546070098877
    },
    {
      "auxiliary_loss_clip": 0.01232528,
      "auxiliary_loss_mlp": 0.01036081,
      "balance_loss_clip": 1.05302143,
      "balance_loss_mlp": 1.02591264,
      "epoch": 0.10425058618409186,
      "flos": 17236511890560.0,
      "grad_norm": 3.2911785508620133,
      "language_loss": 0.89699328,
      "learning_rate": 3.942527170184088e-06,
      "loss": 0.91967934,
      "num_input_tokens_seen": 18435560,
      "step": 867,
      "time_per_iteration": 2.688081979751587
    },
    {
      "auxiliary_loss_clip": 0.01251232,
      "auxiliary_loss_mlp": 0.01041248,
      "balance_loss_clip": 1.06938601,
      "balance_loss_mlp": 1.02948165,
      "epoch": 0.10437082907473096,
      "flos": 17967365919360.0,
      "grad_norm": 2.9858582786947125,
      "language_loss": 0.77689719,
      "learning_rate": 3.942341622608385e-06,
      "loss": 0.79982197,
      "num_input_tokens_seen": 18452590,
      "step": 868,
      "time_per_iteration": 2.555905342102051
    },
    {
      "auxiliary_loss_clip": 0.01254138,
      "auxiliary_loss_mlp": 0.01041043,
      "balance_loss_clip": 1.06912112,
      "balance_loss_mlp": 1.02973616,
      "epoch": 0.10449107196537005,
      "flos": 36283315374720.0,
      "grad_norm": 2.8829772863481646,
      "language_loss": 0.77748716,
      "learning_rate": 3.942155780381001e-06,
      "loss": 0.80043894,
      "num_input_tokens_seen": 18476325,
      "step": 869,
      "time_per_iteration": 2.8237595558166504
    },
    {
      "auxiliary_loss_clip": 0.01250807,
      "auxiliary_loss_mlp": 0.01039742,
      "balance_loss_clip": 1.0612489,
      "balance_loss_mlp": 1.02795804,
      "epoch": 0.10461131485600914,
      "flos": 23802095266560.0,
      "grad_norm": 2.8274237283617842,
      "language_loss": 0.76365876,
      "learning_rate": 3.94196964353013e-06,
      "loss": 0.78656429,
      "num_input_tokens_seen": 18495775,
      "step": 870,
      "time_per_iteration": 2.6638405323028564
    },
    {
      "auxiliary_loss_clip": 0.01246056,
      "auxiliary_loss_mlp": 0.0120927,
      "balance_loss_clip": 1.06050086,
      "balance_loss_mlp": 1.00085008,
      "epoch": 0.10473155774664823,
      "flos": 18405476104320.0,
      "grad_norm": 9.370298980696337,
      "language_loss": 0.80619371,
      "learning_rate": 3.941783212084008e-06,
      "loss": 0.83074695,
      "num_input_tokens_seen": 18513530,
      "step": 871,
      "time_per_iteration": 2.7136788368225098
    },
    {
      "auxiliary_loss_clip": 0.01235059,
      "auxiliary_loss_mlp": 0.01040477,
      "balance_loss_clip": 1.06240439,
      "balance_loss_mlp": 1.02871084,
      "epoch": 0.10485180063728732,
      "flos": 25592637358080.0,
      "grad_norm": 6.547284044075345,
      "language_loss": 0.79068327,
      "learning_rate": 3.941596486070916e-06,
      "loss": 0.81343865,
      "num_input_tokens_seen": 18531575,
      "step": 872,
      "time_per_iteration": 2.696571111679077
    },
    {
      "auxiliary_loss_clip": 0.01250377,
      "auxiliary_loss_mlp": 0.01033069,
      "balance_loss_clip": 1.06203973,
      "balance_loss_mlp": 1.02020049,
      "epoch": 0.10497204352792641,
      "flos": 27088747666560.0,
      "grad_norm": 2.837100787078263,
      "language_loss": 0.58671576,
      "learning_rate": 3.941409465519182e-06,
      "loss": 0.60955024,
      "num_input_tokens_seen": 18552100,
      "step": 873,
      "time_per_iteration": 2.7652292251586914
    },
    {
      "auxiliary_loss_clip": 0.01239679,
      "auxiliary_loss_mlp": 0.01040808,
      "balance_loss_clip": 1.06155157,
      "balance_loss_mlp": 1.02826679,
      "epoch": 0.10509228641856551,
      "flos": 32858479353600.0,
      "grad_norm": 1.9994073427634917,
      "language_loss": 0.85338092,
      "learning_rate": 3.941222150457176e-06,
      "loss": 0.87618577,
      "num_input_tokens_seen": 18575355,
      "step": 874,
      "time_per_iteration": 2.8200507164001465
    },
    {
      "auxiliary_loss_clip": 0.01254269,
      "auxiliary_loss_mlp": 0.01036527,
      "balance_loss_clip": 1.06563282,
      "balance_loss_mlp": 1.0247196,
      "epoch": 0.10521252930920459,
      "flos": 14319165173760.0,
      "grad_norm": 2.851307999391727,
      "language_loss": 0.71647167,
      "learning_rate": 3.941034540913311e-06,
      "loss": 0.73937964,
      "num_input_tokens_seen": 18592885,
      "step": 875,
      "time_per_iteration": 3.627767562866211
    },
    {
      "auxiliary_loss_clip": 0.01251334,
      "auxiliary_loss_mlp": 0.01209863,
      "balance_loss_clip": 1.06870818,
      "balance_loss_mlp": 1.00075698,
      "epoch": 0.10533277219984369,
      "flos": 21687028773120.0,
      "grad_norm": 2.005007694780548,
      "language_loss": 0.82604247,
      "learning_rate": 3.940846636916051e-06,
      "loss": 0.85065442,
      "num_input_tokens_seen": 18612920,
      "step": 876,
      "time_per_iteration": 3.6249685287475586
    },
    {
      "auxiliary_loss_clip": 0.0125051,
      "auxiliary_loss_mlp": 0.01042232,
      "balance_loss_clip": 1.06808519,
      "balance_loss_mlp": 1.02986407,
      "epoch": 0.10545301509048277,
      "flos": 22269787027200.0,
      "grad_norm": 2.591902724121219,
      "language_loss": 0.86540389,
      "learning_rate": 3.940658438493899e-06,
      "loss": 0.88833129,
      "num_input_tokens_seen": 18630765,
      "step": 877,
      "time_per_iteration": 2.6305010318756104
    },
    {
      "auxiliary_loss_clip": 0.01251153,
      "auxiliary_loss_mlp": 0.01040576,
      "balance_loss_clip": 1.06340981,
      "balance_loss_mlp": 1.02807689,
      "epoch": 0.10557325798112187,
      "flos": 22199725549440.0,
      "grad_norm": 2.3929242292776327,
      "language_loss": 0.75895059,
      "learning_rate": 3.940469945675405e-06,
      "loss": 0.78186786,
      "num_input_tokens_seen": 18649150,
      "step": 878,
      "time_per_iteration": 2.7077882289886475
    },
    {
      "auxiliary_loss_clip": 0.01229942,
      "auxiliary_loss_mlp": 0.01041153,
      "balance_loss_clip": 1.05533731,
      "balance_loss_mlp": 1.0299176,
      "epoch": 0.10569350087176095,
      "flos": 25775889569280.0,
      "grad_norm": 2.2125786505865532,
      "language_loss": 0.91649866,
      "learning_rate": 3.940281158489163e-06,
      "loss": 0.93920958,
      "num_input_tokens_seen": 18668380,
      "step": 879,
      "time_per_iteration": 3.685636520385742
    },
    {
      "auxiliary_loss_clip": 0.01250825,
      "auxiliary_loss_mlp": 0.0104081,
      "balance_loss_clip": 1.05502439,
      "balance_loss_mlp": 1.02944946,
      "epoch": 0.10581374376240005,
      "flos": 17311385790720.0,
      "grad_norm": 3.642024258626364,
      "language_loss": 0.83034974,
      "learning_rate": 3.940092076963812e-06,
      "loss": 0.85326612,
      "num_input_tokens_seen": 18685875,
      "step": 880,
      "time_per_iteration": 2.7187702655792236
    },
    {
      "auxiliary_loss_clip": 0.01247328,
      "auxiliary_loss_mlp": 0.01043518,
      "balance_loss_clip": 1.06120658,
      "balance_loss_mlp": 1.03135884,
      "epoch": 0.10593398665303914,
      "flos": 34349454017280.0,
      "grad_norm": 2.6554426513782765,
      "language_loss": 0.7899732,
      "learning_rate": 3.9399027011280355e-06,
      "loss": 0.81288165,
      "num_input_tokens_seen": 18707970,
      "step": 881,
      "time_per_iteration": 2.772758722305298
    },
    {
      "auxiliary_loss_clip": 0.01246607,
      "auxiliary_loss_mlp": 0.01038222,
      "balance_loss_clip": 1.06403542,
      "balance_loss_mlp": 1.02607453,
      "epoch": 0.10605422954367823,
      "flos": 23257977068160.0,
      "grad_norm": 2.534172107845413,
      "language_loss": 0.77449137,
      "learning_rate": 3.939713031010561e-06,
      "loss": 0.79733968,
      "num_input_tokens_seen": 18726335,
      "step": 882,
      "time_per_iteration": 2.643751621246338
    },
    {
      "auxiliary_loss_clip": 0.01251085,
      "auxiliary_loss_mlp": 0.01038108,
      "balance_loss_clip": 1.06550241,
      "balance_loss_mlp": 1.02503657,
      "epoch": 0.10617447243431732,
      "flos": 22820118278400.0,
      "grad_norm": 2.390277685825059,
      "language_loss": 0.77394772,
      "learning_rate": 3.939523066640163e-06,
      "loss": 0.79683965,
      "num_input_tokens_seen": 18745230,
      "step": 883,
      "time_per_iteration": 2.713380813598633
    },
    {
      "auxiliary_loss_clip": 0.01251303,
      "auxiliary_loss_mlp": 0.01036284,
      "balance_loss_clip": 1.06632185,
      "balance_loss_mlp": 1.02456009,
      "epoch": 0.10629471532495641,
      "flos": 24386577373440.0,
      "grad_norm": 2.6488271243600288,
      "language_loss": 0.81294286,
      "learning_rate": 3.939332808045657e-06,
      "loss": 0.83581877,
      "num_input_tokens_seen": 18764880,
      "step": 884,
      "time_per_iteration": 2.6516122817993164
    },
    {
      "auxiliary_loss_clip": 0.01250656,
      "auxiliary_loss_mlp": 0.01038672,
      "balance_loss_clip": 1.06232476,
      "balance_loss_mlp": 1.02758515,
      "epoch": 0.1064149582155955,
      "flos": 21105491581440.0,
      "grad_norm": 4.328518426406478,
      "language_loss": 0.84307551,
      "learning_rate": 3.939142255255906e-06,
      "loss": 0.86596882,
      "num_input_tokens_seen": 18785765,
      "step": 885,
      "time_per_iteration": 2.7460248470306396
    },
    {
      "auxiliary_loss_clip": 0.01251871,
      "auxiliary_loss_mlp": 0.01032922,
      "balance_loss_clip": 1.06833816,
      "balance_loss_mlp": 1.02073908,
      "epoch": 0.1065352011062346,
      "flos": 20702035042560.0,
      "grad_norm": 2.318317430669141,
      "language_loss": 0.86800438,
      "learning_rate": 3.938951408299817e-06,
      "loss": 0.89085233,
      "num_input_tokens_seen": 18804605,
      "step": 886,
      "time_per_iteration": 2.701629400253296
    },
    {
      "auxiliary_loss_clip": 0.01152794,
      "auxiliary_loss_mlp": 0.01003299,
      "balance_loss_clip": 1.03375721,
      "balance_loss_mlp": 0.99934149,
      "epoch": 0.10665544399687368,
      "flos": 62659632689280.0,
      "grad_norm": 0.8082970030150973,
      "language_loss": 0.54395735,
      "learning_rate": 3.938760267206342e-06,
      "loss": 0.56551832,
      "num_input_tokens_seen": 18866425,
      "step": 887,
      "time_per_iteration": 3.169332504272461
    },
    {
      "auxiliary_loss_clip": 0.01250404,
      "auxiliary_loss_mlp": 0.0103745,
      "balance_loss_clip": 1.06957364,
      "balance_loss_mlp": 1.02518356,
      "epoch": 0.10677568688751278,
      "flos": 26140382830080.0,
      "grad_norm": 3.4482426130936115,
      "language_loss": 0.78853166,
      "learning_rate": 3.938568832004475e-06,
      "loss": 0.81141019,
      "num_input_tokens_seen": 18885130,
      "step": 888,
      "time_per_iteration": 2.7107131481170654
    },
    {
      "auxiliary_loss_clip": 0.01241199,
      "auxiliary_loss_mlp": 0.01046801,
      "balance_loss_clip": 1.06065333,
      "balance_loss_mlp": 1.03424239,
      "epoch": 0.10689592977815186,
      "flos": 12786533712000.0,
      "grad_norm": 2.0435580279097683,
      "language_loss": 0.74907112,
      "learning_rate": 3.938377102723257e-06,
      "loss": 0.77195114,
      "num_input_tokens_seen": 18902265,
      "step": 889,
      "time_per_iteration": 2.6467597484588623
    },
    {
      "auxiliary_loss_clip": 0.01237254,
      "auxiliary_loss_mlp": 0.0104784,
      "balance_loss_clip": 1.05739164,
      "balance_loss_mlp": 1.03492379,
      "epoch": 0.10701617266879096,
      "flos": 22126683242880.0,
      "grad_norm": 2.244245316124162,
      "language_loss": 0.83357245,
      "learning_rate": 3.938185079391774e-06,
      "loss": 0.85642344,
      "num_input_tokens_seen": 18919310,
      "step": 890,
      "time_per_iteration": 2.765141487121582
    },
    {
      "auxiliary_loss_clip": 0.01248871,
      "auxiliary_loss_mlp": 0.01029709,
      "balance_loss_clip": 1.06775498,
      "balance_loss_mlp": 1.01813972,
      "epoch": 0.10713641555943004,
      "flos": 19745625559680.0,
      "grad_norm": 2.718949088366109,
      "language_loss": 1.05569744,
      "learning_rate": 3.937992762039157e-06,
      "loss": 1.07848334,
      "num_input_tokens_seen": 18932635,
      "step": 891,
      "time_per_iteration": 2.65024995803833
    },
    {
      "auxiliary_loss_clip": 0.01249923,
      "auxiliary_loss_mlp": 0.01046061,
      "balance_loss_clip": 1.06829238,
      "balance_loss_mlp": 1.03450334,
      "epoch": 0.10725665845006914,
      "flos": 23952992302080.0,
      "grad_norm": 4.337991794548775,
      "language_loss": 0.80358076,
      "learning_rate": 3.937800150694577e-06,
      "loss": 0.82654059,
      "num_input_tokens_seen": 18953810,
      "step": 892,
      "time_per_iteration": 2.6963698863983154
    },
    {
      "auxiliary_loss_clip": 0.01251564,
      "auxiliary_loss_mlp": 0.01036993,
      "balance_loss_clip": 1.0629673,
      "balance_loss_mlp": 1.02447593,
      "epoch": 0.10737690134070824,
      "flos": 18551704371840.0,
      "grad_norm": 2.541579103946323,
      "language_loss": 0.75802517,
      "learning_rate": 3.937607245387255e-06,
      "loss": 0.78091073,
      "num_input_tokens_seen": 18973175,
      "step": 893,
      "time_per_iteration": 2.6861088275909424
    },
    {
      "auxiliary_loss_clip": 0.01258438,
      "auxiliary_loss_mlp": 0.01041122,
      "balance_loss_clip": 1.06512773,
      "balance_loss_mlp": 1.03010094,
      "epoch": 0.10749714423134732,
      "flos": 22707609903360.0,
      "grad_norm": 1.928967712969956,
      "language_loss": 0.72350597,
      "learning_rate": 3.937414046146455e-06,
      "loss": 0.74650156,
      "num_input_tokens_seen": 18991130,
      "step": 894,
      "time_per_iteration": 2.874082088470459
    },
    {
      "auxiliary_loss_clip": 0.01251426,
      "auxiliary_loss_mlp": 0.01049057,
      "balance_loss_clip": 1.07085097,
      "balance_loss_mlp": 1.03674281,
      "epoch": 0.10761738712198642,
      "flos": 21106066199040.0,
      "grad_norm": 2.1214655535406544,
      "language_loss": 0.75419927,
      "learning_rate": 3.9372205530014845e-06,
      "loss": 0.77720416,
      "num_input_tokens_seen": 19009610,
      "step": 895,
      "time_per_iteration": 2.680744171142578
    },
    {
      "auxiliary_loss_clip": 0.01247897,
      "auxiliary_loss_mlp": 0.01049492,
      "balance_loss_clip": 1.0677247,
      "balance_loss_mlp": 1.0379765,
      "epoch": 0.1077376300126255,
      "flos": 23766723348480.0,
      "grad_norm": 2.4428495077899597,
      "language_loss": 0.71440482,
      "learning_rate": 3.937026765981696e-06,
      "loss": 0.73737872,
      "num_input_tokens_seen": 19029680,
      "step": 896,
      "time_per_iteration": 2.6927788257598877
    },
    {
      "auxiliary_loss_clip": 0.01255284,
      "auxiliary_loss_mlp": 0.01046966,
      "balance_loss_clip": 1.06415582,
      "balance_loss_mlp": 1.03476501,
      "epoch": 0.1078578729032646,
      "flos": 20919581763840.0,
      "grad_norm": 1.8700957359339074,
      "language_loss": 0.79089868,
      "learning_rate": 3.936832685116488e-06,
      "loss": 0.81392115,
      "num_input_tokens_seen": 19047775,
      "step": 897,
      "time_per_iteration": 2.713066339492798
    },
    {
      "auxiliary_loss_clip": 0.01248565,
      "auxiliary_loss_mlp": 0.01045349,
      "balance_loss_clip": 1.06813276,
      "balance_loss_mlp": 1.0337441,
      "epoch": 0.10797811579390369,
      "flos": 14829886702080.0,
      "grad_norm": 2.343952652028355,
      "language_loss": 0.9014743,
      "learning_rate": 3.936638310435301e-06,
      "loss": 0.92441344,
      "num_input_tokens_seen": 19065640,
      "step": 898,
      "time_per_iteration": 2.614562511444092
    },
    {
      "auxiliary_loss_clip": 0.01252613,
      "auxiliary_loss_mlp": 0.0103741,
      "balance_loss_clip": 1.06784773,
      "balance_loss_mlp": 1.02547705,
      "epoch": 0.10809835868454278,
      "flos": 19536985411200.0,
      "grad_norm": 1.847319751654452,
      "language_loss": 0.81505322,
      "learning_rate": 3.936443641967623e-06,
      "loss": 0.83795345,
      "num_input_tokens_seen": 19084470,
      "step": 899,
      "time_per_iteration": 2.691699743270874
    },
    {
      "auxiliary_loss_clip": 0.01251598,
      "auxiliary_loss_mlp": 0.01046708,
      "balance_loss_clip": 1.06710339,
      "balance_loss_mlp": 1.03395224,
      "epoch": 0.10821860157518187,
      "flos": 18442320480000.0,
      "grad_norm": 3.129639923222281,
      "language_loss": 0.83201241,
      "learning_rate": 3.936248679742983e-06,
      "loss": 0.85499549,
      "num_input_tokens_seen": 19102965,
      "step": 900,
      "time_per_iteration": 2.662074327468872
    },
    {
      "auxiliary_loss_clip": 0.01157768,
      "auxiliary_loss_mlp": 0.01005103,
      "balance_loss_clip": 1.04031658,
      "balance_loss_mlp": 1.00186026,
      "epoch": 0.10833884446582095,
      "flos": 49359468447360.0,
      "grad_norm": 1.0485468957692374,
      "language_loss": 0.70136535,
      "learning_rate": 3.936053423790959e-06,
      "loss": 0.72299409,
      "num_input_tokens_seen": 19151285,
      "step": 901,
      "time_per_iteration": 3.949571132659912
    },
    {
      "auxiliary_loss_clip": 0.01249185,
      "auxiliary_loss_mlp": 0.01051087,
      "balance_loss_clip": 1.07052898,
      "balance_loss_mlp": 1.03992355,
      "epoch": 0.10845908735646005,
      "flos": 20411912891520.0,
      "grad_norm": 1.9496729498778922,
      "language_loss": 0.77263916,
      "learning_rate": 3.935857874141168e-06,
      "loss": 0.7956419,
      "num_input_tokens_seen": 19170120,
      "step": 902,
      "time_per_iteration": 2.646045207977295
    },
    {
      "auxiliary_loss_clip": 0.01243954,
      "auxiliary_loss_mlp": 0.0103319,
      "balance_loss_clip": 1.06318498,
      "balance_loss_mlp": 1.02047086,
      "epoch": 0.10857933024709913,
      "flos": 14027750133120.0,
      "grad_norm": 2.4691747399000454,
      "language_loss": 0.83360708,
      "learning_rate": 3.935662030823279e-06,
      "loss": 0.8563785,
      "num_input_tokens_seen": 19186305,
      "step": 903,
      "time_per_iteration": 3.5472378730773926
    },
    {
      "auxiliary_loss_clip": 0.01249365,
      "auxiliary_loss_mlp": 0.01048676,
      "balance_loss_clip": 1.06627584,
      "balance_loss_mlp": 1.0370059,
      "epoch": 0.10869957313773823,
      "flos": 13369004657280.0,
      "grad_norm": 3.1871902969157264,
      "language_loss": 0.71999943,
      "learning_rate": 3.935465893866998e-06,
      "loss": 0.74297982,
      "num_input_tokens_seen": 19204530,
      "step": 904,
      "time_per_iteration": 2.649961471557617
    },
    {
      "auxiliary_loss_clip": 0.01254743,
      "auxiliary_loss_mlp": 0.01039469,
      "balance_loss_clip": 1.06846559,
      "balance_loss_mlp": 1.02757752,
      "epoch": 0.10881981602837733,
      "flos": 25807095509760.0,
      "grad_norm": 4.201828117290073,
      "language_loss": 0.80241013,
      "learning_rate": 3.935269463302079e-06,
      "loss": 0.82535219,
      "num_input_tokens_seen": 19222735,
      "step": 905,
      "time_per_iteration": 2.9186222553253174
    },
    {
      "auxiliary_loss_clip": 0.01252395,
      "auxiliary_loss_mlp": 0.01045946,
      "balance_loss_clip": 1.0673883,
      "balance_loss_mlp": 1.03334618,
      "epoch": 0.10894005891901641,
      "flos": 20777555387520.0,
      "grad_norm": 1.9957954958155628,
      "language_loss": 0.76615578,
      "learning_rate": 3.935072739158322e-06,
      "loss": 0.78913915,
      "num_input_tokens_seen": 19242445,
      "step": 906,
      "time_per_iteration": 3.543483257293701
    },
    {
      "auxiliary_loss_clip": 0.01256151,
      "auxiliary_loss_mlp": 0.01045341,
      "balance_loss_clip": 1.06653309,
      "balance_loss_mlp": 1.03318143,
      "epoch": 0.10906030180965551,
      "flos": 26649883296000.0,
      "grad_norm": 2.4402939352257476,
      "language_loss": 0.79678595,
      "learning_rate": 3.934875721465569e-06,
      "loss": 0.81980085,
      "num_input_tokens_seen": 19262865,
      "step": 907,
      "time_per_iteration": 2.739206075668335
    },
    {
      "auxiliary_loss_clip": 0.01245924,
      "auxiliary_loss_mlp": 0.01039513,
      "balance_loss_clip": 1.06079483,
      "balance_loss_mlp": 1.02679896,
      "epoch": 0.10918054470029459,
      "flos": 36534402420480.0,
      "grad_norm": 2.9780061073529196,
      "language_loss": 0.71618938,
      "learning_rate": 3.9346784102537076e-06,
      "loss": 0.73904383,
      "num_input_tokens_seen": 19285000,
      "step": 908,
      "time_per_iteration": 2.7780251502990723
    },
    {
      "auxiliary_loss_clip": 0.01248816,
      "auxiliary_loss_mlp": 0.01034607,
      "balance_loss_clip": 1.06867015,
      "balance_loss_mlp": 1.02278709,
      "epoch": 0.10930078759093369,
      "flos": 21762549118080.0,
      "grad_norm": 2.060835818112184,
      "language_loss": 0.78512591,
      "learning_rate": 3.934480805552669e-06,
      "loss": 0.80796015,
      "num_input_tokens_seen": 19306010,
      "step": 909,
      "time_per_iteration": 2.7440717220306396
    },
    {
      "auxiliary_loss_clip": 0.01250346,
      "auxiliary_loss_mlp": 0.01210259,
      "balance_loss_clip": 1.07070374,
      "balance_loss_mlp": 1.00067878,
      "epoch": 0.10942103048157277,
      "flos": 22601781457920.0,
      "grad_norm": 2.8653386969799866,
      "language_loss": 0.88273913,
      "learning_rate": 3.93428290739243e-06,
      "loss": 0.90734524,
      "num_input_tokens_seen": 19325380,
      "step": 910,
      "time_per_iteration": 2.6116561889648438
    },
    {
      "auxiliary_loss_clip": 0.012533,
      "auxiliary_loss_mlp": 0.01041397,
      "balance_loss_clip": 1.06646848,
      "balance_loss_mlp": 1.02896893,
      "epoch": 0.10954127337221187,
      "flos": 15045781397760.0,
      "grad_norm": 2.808663205824217,
      "language_loss": 0.80074573,
      "learning_rate": 3.9340847158030125e-06,
      "loss": 0.82369268,
      "num_input_tokens_seen": 19338960,
      "step": 911,
      "time_per_iteration": 2.6746959686279297
    },
    {
      "auxiliary_loss_clip": 0.0125108,
      "auxiliary_loss_mlp": 0.01047838,
      "balance_loss_clip": 1.06569302,
      "balance_loss_mlp": 1.03597045,
      "epoch": 0.10966151626285096,
      "flos": 21650974496640.0,
      "grad_norm": 1.890729462936453,
      "language_loss": 0.75617093,
      "learning_rate": 3.9338862308144814e-06,
      "loss": 0.77916014,
      "num_input_tokens_seen": 19357780,
      "step": 912,
      "time_per_iteration": 2.6441261768341064
    },
    {
      "auxiliary_loss_clip": 0.01249967,
      "auxiliary_loss_mlp": 0.01039835,
      "balance_loss_clip": 1.06896067,
      "balance_loss_mlp": 1.02751517,
      "epoch": 0.10978175915349005,
      "flos": 20121359777280.0,
      "grad_norm": 1.713048593824514,
      "language_loss": 0.84638858,
      "learning_rate": 3.933687452456946e-06,
      "loss": 0.86928654,
      "num_input_tokens_seen": 19377680,
      "step": 913,
      "time_per_iteration": 2.634003162384033
    },
    {
      "auxiliary_loss_clip": 0.01249938,
      "auxiliary_loss_mlp": 0.01038303,
      "balance_loss_clip": 1.05935836,
      "balance_loss_mlp": 1.02535701,
      "epoch": 0.10990200204412914,
      "flos": 20412667077120.0,
      "grad_norm": 2.1591088472930515,
      "language_loss": 0.86596811,
      "learning_rate": 3.933488380760562e-06,
      "loss": 0.88885051,
      "num_input_tokens_seen": 19397040,
      "step": 914,
      "time_per_iteration": 2.6939005851745605
    },
    {
      "auxiliary_loss_clip": 0.01248834,
      "auxiliary_loss_mlp": 0.01211132,
      "balance_loss_clip": 1.06826735,
      "balance_loss_mlp": 1.00058866,
      "epoch": 0.11002224493476823,
      "flos": 17530117660800.0,
      "grad_norm": 2.4026992384869343,
      "language_loss": 0.87389839,
      "learning_rate": 3.9332890157555286e-06,
      "loss": 0.89849806,
      "num_input_tokens_seen": 19413975,
      "step": 915,
      "time_per_iteration": 2.628365993499756
    },
    {
      "auxiliary_loss_clip": 0.01256751,
      "auxiliary_loss_mlp": 0.01043051,
      "balance_loss_clip": 1.06657398,
      "balance_loss_mlp": 1.03108263,
      "epoch": 0.11014248782540732,
      "flos": 12203093099520.0,
      "grad_norm": 3.118046185388723,
      "language_loss": 0.7652899,
      "learning_rate": 3.933089357472088e-06,
      "loss": 0.78828788,
      "num_input_tokens_seen": 19432005,
      "step": 916,
      "time_per_iteration": 2.6315627098083496
    },
    {
      "auxiliary_loss_clip": 0.01249421,
      "auxiliary_loss_mlp": 0.01038796,
      "balance_loss_clip": 1.07113385,
      "balance_loss_mlp": 1.02674437,
      "epoch": 0.11026273071604642,
      "flos": 22382977760640.0,
      "grad_norm": 2.9330817321509572,
      "language_loss": 0.85752946,
      "learning_rate": 3.932889405940529e-06,
      "loss": 0.88041162,
      "num_input_tokens_seen": 19450100,
      "step": 917,
      "time_per_iteration": 2.647952079772949
    },
    {
      "auxiliary_loss_clip": 0.01253318,
      "auxiliary_loss_mlp": 0.01040065,
      "balance_loss_clip": 1.07020569,
      "balance_loss_mlp": 1.02842975,
      "epoch": 0.1103829736066855,
      "flos": 19829046896640.0,
      "grad_norm": 2.181184954933588,
      "language_loss": 0.80255365,
      "learning_rate": 3.932689161191184e-06,
      "loss": 0.82548749,
      "num_input_tokens_seen": 19467805,
      "step": 918,
      "time_per_iteration": 2.679583787918091
    },
    {
      "auxiliary_loss_clip": 0.01249471,
      "auxiliary_loss_mlp": 0.01043109,
      "balance_loss_clip": 1.06703401,
      "balance_loss_mlp": 1.03024602,
      "epoch": 0.1105032164973246,
      "flos": 22669616292480.0,
      "grad_norm": 2.1723235245801478,
      "language_loss": 0.8838414,
      "learning_rate": 3.93248862325443e-06,
      "loss": 0.90676719,
      "num_input_tokens_seen": 19486710,
      "step": 919,
      "time_per_iteration": 2.83659291267395
    },
    {
      "auxiliary_loss_clip": 0.01156996,
      "auxiliary_loss_mlp": 0.01004209,
      "balance_loss_clip": 1.04136205,
      "balance_loss_mlp": 1.00044239,
      "epoch": 0.11062345938796368,
      "flos": 66483507876480.0,
      "grad_norm": 0.9296615370468743,
      "language_loss": 0.64398348,
      "learning_rate": 3.932287792160688e-06,
      "loss": 0.66559553,
      "num_input_tokens_seen": 19545170,
      "step": 920,
      "time_per_iteration": 3.150646209716797
    },
    {
      "auxiliary_loss_clip": 0.01252543,
      "auxiliary_loss_mlp": 0.01041448,
      "balance_loss_clip": 1.06552505,
      "balance_loss_mlp": 1.02819145,
      "epoch": 0.11074370227860278,
      "flos": 21907771804800.0,
      "grad_norm": 2.253582378039604,
      "language_loss": 0.81086236,
      "learning_rate": 3.932086667940424e-06,
      "loss": 0.83380228,
      "num_input_tokens_seen": 19561875,
      "step": 921,
      "time_per_iteration": 2.6856701374053955
    },
    {
      "auxiliary_loss_clip": 0.01249138,
      "auxiliary_loss_mlp": 0.01210436,
      "balance_loss_clip": 1.06877458,
      "balance_loss_mlp": 1.00061929,
      "epoch": 0.11086394516924186,
      "flos": 28658115763200.0,
      "grad_norm": 2.242992316234109,
      "language_loss": 0.81636333,
      "learning_rate": 3.93188525062415e-06,
      "loss": 0.84095907,
      "num_input_tokens_seen": 19582340,
      "step": 922,
      "time_per_iteration": 2.8087382316589355
    },
    {
      "auxiliary_loss_clip": 0.01251144,
      "auxiliary_loss_mlp": 0.01049289,
      "balance_loss_clip": 1.06709564,
      "balance_loss_mlp": 1.0366466,
      "epoch": 0.11098418805988096,
      "flos": 24535247765760.0,
      "grad_norm": 2.0298492808209283,
      "language_loss": 0.86532646,
      "learning_rate": 3.931683540242418e-06,
      "loss": 0.88833076,
      "num_input_tokens_seen": 19603405,
      "step": 923,
      "time_per_iteration": 2.638869285583496
    },
    {
      "auxiliary_loss_clip": 0.01243097,
      "auxiliary_loss_mlp": 0.01039983,
      "balance_loss_clip": 1.06483102,
      "balance_loss_mlp": 1.02723956,
      "epoch": 0.11110443095052006,
      "flos": 22960384888320.0,
      "grad_norm": 2.575620953500358,
      "language_loss": 0.91516697,
      "learning_rate": 3.9314815368258295e-06,
      "loss": 0.93799782,
      "num_input_tokens_seen": 19619885,
      "step": 924,
      "time_per_iteration": 2.6779942512512207
    },
    {
      "auxiliary_loss_clip": 0.01253628,
      "auxiliary_loss_mlp": 0.01039055,
      "balance_loss_clip": 1.07147145,
      "balance_loss_mlp": 1.02708089,
      "epoch": 0.11122467384115914,
      "flos": 18950025265920.0,
      "grad_norm": 1.8493736809712344,
      "language_loss": 0.78773642,
      "learning_rate": 3.9312792404050275e-06,
      "loss": 0.81066322,
      "num_input_tokens_seen": 19637940,
      "step": 925,
      "time_per_iteration": 2.6097652912139893
    },
    {
      "auxiliary_loss_clip": 0.01247123,
      "auxiliary_loss_mlp": 0.0104352,
      "balance_loss_clip": 1.07016349,
      "balance_loss_mlp": 1.03259993,
      "epoch": 0.11134491673179824,
      "flos": 25082957324160.0,
      "grad_norm": 1.9250570805187726,
      "language_loss": 0.77431554,
      "learning_rate": 3.9310766510107e-06,
      "loss": 0.79722202,
      "num_input_tokens_seen": 19657115,
      "step": 926,
      "time_per_iteration": 2.6680562496185303
    },
    {
      "auxiliary_loss_clip": 0.01251367,
      "auxiliary_loss_mlp": 0.01043389,
      "balance_loss_clip": 1.06048751,
      "balance_loss_mlp": 1.0305326,
      "epoch": 0.11146515962243732,
      "flos": 24499121662080.0,
      "grad_norm": 1.8771179496403565,
      "language_loss": 0.92180055,
      "learning_rate": 3.9308737686735806e-06,
      "loss": 0.9447481,
      "num_input_tokens_seen": 19677075,
      "step": 927,
      "time_per_iteration": 2.6891212463378906
    },
    {
      "auxiliary_loss_clip": 0.01252634,
      "auxiliary_loss_mlp": 0.01046361,
      "balance_loss_clip": 1.0720824,
      "balance_loss_mlp": 1.03451753,
      "epoch": 0.11158540251307641,
      "flos": 22343763087360.0,
      "grad_norm": 2.345875561396887,
      "language_loss": 0.82446569,
      "learning_rate": 3.9306705934244455e-06,
      "loss": 0.84745562,
      "num_input_tokens_seen": 19697155,
      "step": 928,
      "time_per_iteration": 3.5558481216430664
    },
    {
      "auxiliary_loss_clip": 0.01240034,
      "auxiliary_loss_mlp": 0.01033526,
      "balance_loss_clip": 1.06228709,
      "balance_loss_mlp": 1.02171206,
      "epoch": 0.11170564540371551,
      "flos": 19902304684800.0,
      "grad_norm": 1.7797386231933339,
      "language_loss": 0.88409871,
      "learning_rate": 3.930467125294116e-06,
      "loss": 0.90683436,
      "num_input_tokens_seen": 19716705,
      "step": 929,
      "time_per_iteration": 2.791775941848755
    },
    {
      "auxiliary_loss_clip": 0.01159938,
      "auxiliary_loss_mlp": 0.01007912,
      "balance_loss_clip": 1.03014278,
      "balance_loss_mlp": 1.00431144,
      "epoch": 0.1118258882943546,
      "flos": 64586239499520.0,
      "grad_norm": 1.0258206982966727,
      "language_loss": 0.60448337,
      "learning_rate": 3.930263364313458e-06,
      "loss": 0.62616187,
      "num_input_tokens_seen": 19767275,
      "step": 930,
      "time_per_iteration": 4.13698935508728
    },
    {
      "auxiliary_loss_clip": 0.01248587,
      "auxiliary_loss_mlp": 0.01048345,
      "balance_loss_clip": 1.06176758,
      "balance_loss_mlp": 1.03570914,
      "epoch": 0.11194613118499369,
      "flos": 17201965985280.0,
      "grad_norm": 3.0530944603265895,
      "language_loss": 0.83262587,
      "learning_rate": 3.930059310513384e-06,
      "loss": 0.85559517,
      "num_input_tokens_seen": 19786315,
      "step": 931,
      "time_per_iteration": 2.6794230937957764
    },
    {
      "auxiliary_loss_clip": 0.0123629,
      "auxiliary_loss_mlp": 0.01210259,
      "balance_loss_clip": 1.06046128,
      "balance_loss_mlp": 1.00056767,
      "epoch": 0.11206637407563277,
      "flos": 31863465728640.0,
      "grad_norm": 2.2229902480722283,
      "language_loss": 0.83876395,
      "learning_rate": 3.929854963924846e-06,
      "loss": 0.86322939,
      "num_input_tokens_seen": 19806580,
      "step": 932,
      "time_per_iteration": 4.697466850280762
    },
    {
      "auxiliary_loss_clip": 0.01253185,
      "auxiliary_loss_mlp": 0.01035742,
      "balance_loss_clip": 1.06185389,
      "balance_loss_mlp": 1.023911,
      "epoch": 0.11218661696627187,
      "flos": 21945621761280.0,
      "grad_norm": 1.9022503955738725,
      "language_loss": 0.77198094,
      "learning_rate": 3.929650324578845e-06,
      "loss": 0.79487026,
      "num_input_tokens_seen": 19826045,
      "step": 933,
      "time_per_iteration": 2.7269351482391357
    },
    {
      "auxiliary_loss_clip": 0.01256575,
      "auxiliary_loss_mlp": 0.01042248,
      "balance_loss_clip": 1.06681299,
      "balance_loss_mlp": 1.02930784,
      "epoch": 0.11230685985691095,
      "flos": 25878198481920.0,
      "grad_norm": 2.818579268861682,
      "language_loss": 0.81956983,
      "learning_rate": 3.929445392506423e-06,
      "loss": 0.84255809,
      "num_input_tokens_seen": 19843985,
      "step": 934,
      "time_per_iteration": 2.7226412296295166
    },
    {
      "auxiliary_loss_clip": 0.01250556,
      "auxiliary_loss_mlp": 0.01044829,
      "balance_loss_clip": 1.0711565,
      "balance_loss_mlp": 1.03315282,
      "epoch": 0.11242710274755005,
      "flos": 22231506107520.0,
      "grad_norm": 1.8469804458169827,
      "language_loss": 0.76406115,
      "learning_rate": 3.92924016773867e-06,
      "loss": 0.78701508,
      "num_input_tokens_seen": 19860480,
      "step": 935,
      "time_per_iteration": 2.695375442504883
    },
    {
      "auxiliary_loss_clip": 0.01251002,
      "auxiliary_loss_mlp": 0.01209844,
      "balance_loss_clip": 1.06307721,
      "balance_loss_mlp": 1.00057328,
      "epoch": 0.11254734563818915,
      "flos": 17712184723200.0,
      "grad_norm": 4.844426733673834,
      "language_loss": 0.73709786,
      "learning_rate": 3.9290346503067175e-06,
      "loss": 0.76170635,
      "num_input_tokens_seen": 19877145,
      "step": 936,
      "time_per_iteration": 2.651951551437378
    },
    {
      "auxiliary_loss_clip": 0.01253501,
      "auxiliary_loss_mlp": 0.01042379,
      "balance_loss_clip": 1.06734371,
      "balance_loss_mlp": 1.03029084,
      "epoch": 0.11266758852882823,
      "flos": 54930397334400.0,
      "grad_norm": 1.9853208788347305,
      "language_loss": 0.7894699,
      "learning_rate": 3.9288288402417415e-06,
      "loss": 0.81242871,
      "num_input_tokens_seen": 19903405,
      "step": 937,
      "time_per_iteration": 3.003933906555176
    },
    {
      "auxiliary_loss_clip": 0.0125566,
      "auxiliary_loss_mlp": 0.01037674,
      "balance_loss_clip": 1.07089472,
      "balance_loss_mlp": 1.0249666,
      "epoch": 0.11278783141946733,
      "flos": 18878132194560.0,
      "grad_norm": 2.257062619762604,
      "language_loss": 0.70645022,
      "learning_rate": 3.928622737574964e-06,
      "loss": 0.72938359,
      "num_input_tokens_seen": 19918740,
      "step": 938,
      "time_per_iteration": 2.6314072608947754
    },
    {
      "auxiliary_loss_clip": 0.01252095,
      "auxiliary_loss_mlp": 0.01043125,
      "balance_loss_clip": 1.06526589,
      "balance_loss_mlp": 1.03087044,
      "epoch": 0.11290807431010641,
      "flos": 26469252777600.0,
      "grad_norm": 2.0452369946440254,
      "language_loss": 0.91211796,
      "learning_rate": 3.928416342337652e-06,
      "loss": 0.93507016,
      "num_input_tokens_seen": 19938475,
      "step": 939,
      "time_per_iteration": 2.749629020690918
    },
    {
      "auxiliary_loss_clip": 0.01254664,
      "auxiliary_loss_mlp": 0.01046328,
      "balance_loss_clip": 1.06744647,
      "balance_loss_mlp": 1.03369164,
      "epoch": 0.1130283172007455,
      "flos": 22710590732160.0,
      "grad_norm": 2.528039523056551,
      "language_loss": 0.8303076,
      "learning_rate": 3.928209654561113e-06,
      "loss": 0.85331756,
      "num_input_tokens_seen": 19959310,
      "step": 940,
      "time_per_iteration": 2.6575357913970947
    },
    {
      "auxiliary_loss_clip": 0.01244273,
      "auxiliary_loss_mlp": 0.01039967,
      "balance_loss_clip": 1.06514359,
      "balance_loss_mlp": 1.02811718,
      "epoch": 0.1131485600913846,
      "flos": 23219911630080.0,
      "grad_norm": 2.9874548306202624,
      "language_loss": 0.81735778,
      "learning_rate": 3.928002674276703e-06,
      "loss": 0.84020019,
      "num_input_tokens_seen": 19978700,
      "step": 941,
      "time_per_iteration": 2.663496971130371
    },
    {
      "auxiliary_loss_clip": 0.01234695,
      "auxiliary_loss_mlp": 0.01042157,
      "balance_loss_clip": 1.0557332,
      "balance_loss_mlp": 1.02968156,
      "epoch": 0.11326880298202369,
      "flos": 14064271286400.0,
      "grad_norm": 1.971710503525863,
      "language_loss": 0.75735795,
      "learning_rate": 3.92779540151582e-06,
      "loss": 0.78012645,
      "num_input_tokens_seen": 19995785,
      "step": 942,
      "time_per_iteration": 2.733708381652832
    },
    {
      "auxiliary_loss_clip": 0.01249821,
      "auxiliary_loss_mlp": 0.01037389,
      "balance_loss_clip": 1.06515396,
      "balance_loss_mlp": 1.02569485,
      "epoch": 0.11338904587266278,
      "flos": 16325386479360.0,
      "grad_norm": 1.8421897362840443,
      "language_loss": 0.85505253,
      "learning_rate": 3.927587836309907e-06,
      "loss": 0.87792462,
      "num_input_tokens_seen": 20013615,
      "step": 943,
      "time_per_iteration": 2.654275894165039
    },
    {
      "auxiliary_loss_clip": 0.01246948,
      "auxiliary_loss_mlp": 0.01044717,
      "balance_loss_clip": 1.06378937,
      "balance_loss_mlp": 1.03290927,
      "epoch": 0.11350928876330187,
      "flos": 24426258923520.0,
      "grad_norm": 1.9108985794210807,
      "language_loss": 0.78343785,
      "learning_rate": 3.927379978690452e-06,
      "loss": 0.80635446,
      "num_input_tokens_seen": 20032880,
      "step": 944,
      "time_per_iteration": 2.7180826663970947
    },
    {
      "auxiliary_loss_clip": 0.01238821,
      "auxiliary_loss_mlp": 0.01043782,
      "balance_loss_clip": 1.0557878,
      "balance_loss_mlp": 1.03167069,
      "epoch": 0.11362953165394096,
      "flos": 24497074586880.0,
      "grad_norm": 2.1021783384799426,
      "language_loss": 0.87261438,
      "learning_rate": 3.927171828688987e-06,
      "loss": 0.89544034,
      "num_input_tokens_seen": 20052405,
      "step": 945,
      "time_per_iteration": 2.7090659141540527
    },
    {
      "auxiliary_loss_clip": 0.01251353,
      "auxiliary_loss_mlp": 0.01037337,
      "balance_loss_clip": 1.0724442,
      "balance_loss_mlp": 1.02528501,
      "epoch": 0.11374977454458005,
      "flos": 24060831909120.0,
      "grad_norm": 2.5067130319307696,
      "language_loss": 0.82168221,
      "learning_rate": 3.926963386337088e-06,
      "loss": 0.84456915,
      "num_input_tokens_seen": 20070635,
      "step": 946,
      "time_per_iteration": 2.69012188911438
    },
    {
      "auxiliary_loss_clip": 0.01251786,
      "auxiliary_loss_mlp": 0.01042617,
      "balance_loss_clip": 1.06849861,
      "balance_loss_mlp": 1.02905107,
      "epoch": 0.11387001743521914,
      "flos": 39457638967680.0,
      "grad_norm": 2.836269774695643,
      "language_loss": 0.70011401,
      "learning_rate": 3.926754651666375e-06,
      "loss": 0.72305799,
      "num_input_tokens_seen": 20091195,
      "step": 947,
      "time_per_iteration": 2.7753982543945312
    },
    {
      "auxiliary_loss_clip": 0.01253643,
      "auxiliary_loss_mlp": 0.01042748,
      "balance_loss_clip": 1.06639016,
      "balance_loss_mlp": 1.03085113,
      "epoch": 0.11399026032585824,
      "flos": 25082454533760.0,
      "grad_norm": 2.5470083254021403,
      "language_loss": 0.77991545,
      "learning_rate": 3.926545624708513e-06,
      "loss": 0.80287939,
      "num_input_tokens_seen": 20110435,
      "step": 948,
      "time_per_iteration": 2.7526915073394775
    },
    {
      "auxiliary_loss_clip": 0.01247716,
      "auxiliary_loss_mlp": 0.01043899,
      "balance_loss_clip": 1.06202126,
      "balance_loss_mlp": 1.03200781,
      "epoch": 0.11411050321649732,
      "flos": 17961835224960.0,
      "grad_norm": 2.0143750625543198,
      "language_loss": 0.85624564,
      "learning_rate": 3.926336305495213e-06,
      "loss": 0.87916178,
      "num_input_tokens_seen": 20128995,
      "step": 949,
      "time_per_iteration": 2.683868646621704
    },
    {
      "auxiliary_loss_clip": 0.01239491,
      "auxiliary_loss_mlp": 0.01044533,
      "balance_loss_clip": 1.06292868,
      "balance_loss_mlp": 1.03127074,
      "epoch": 0.11423074610713642,
      "flos": 22455409536000.0,
      "grad_norm": 2.0239029481325512,
      "language_loss": 0.88731754,
      "learning_rate": 3.926126694058226e-06,
      "loss": 0.91015768,
      "num_input_tokens_seen": 20148145,
      "step": 950,
      "time_per_iteration": 2.756082057952881
    },
    {
      "auxiliary_loss_clip": 0.01248948,
      "auxiliary_loss_mlp": 0.01041592,
      "balance_loss_clip": 1.06406784,
      "balance_loss_mlp": 1.03082788,
      "epoch": 0.1143509889977755,
      "flos": 19717687756800.0,
      "grad_norm": 1.8168865399903797,
      "language_loss": 0.8214283,
      "learning_rate": 3.92591679042935e-06,
      "loss": 0.84433377,
      "num_input_tokens_seen": 20168035,
      "step": 951,
      "time_per_iteration": 2.6693103313446045
    },
    {
      "auxiliary_loss_clip": 0.01249392,
      "auxiliary_loss_mlp": 0.01040921,
      "balance_loss_clip": 1.06842566,
      "balance_loss_mlp": 1.02813566,
      "epoch": 0.1144712318884146,
      "flos": 19822869757440.0,
      "grad_norm": 1.6690711243521605,
      "language_loss": 0.82418513,
      "learning_rate": 3.92570659464043e-06,
      "loss": 0.84708828,
      "num_input_tokens_seen": 20186095,
      "step": 952,
      "time_per_iteration": 2.7290167808532715
    },
    {
      "auxiliary_loss_clip": 0.0124619,
      "auxiliary_loss_mlp": 0.01210205,
      "balance_loss_clip": 1.06847692,
      "balance_loss_mlp": 1.0005641,
      "epoch": 0.1145914747790537,
      "flos": 14939198766720.0,
      "grad_norm": 2.0292907929225357,
      "language_loss": 0.80048025,
      "learning_rate": 3.925496106723349e-06,
      "loss": 0.82504421,
      "num_input_tokens_seen": 20203535,
      "step": 953,
      "time_per_iteration": 2.6037991046905518
    },
    {
      "auxiliary_loss_clip": 0.01251383,
      "auxiliary_loss_mlp": 0.01040543,
      "balance_loss_clip": 1.06763005,
      "balance_loss_mlp": 1.02943301,
      "epoch": 0.11471171766969278,
      "flos": 19865029345920.0,
      "grad_norm": 2.0602272465119382,
      "language_loss": 0.83888394,
      "learning_rate": 3.9252853267100405e-06,
      "loss": 0.86180317,
      "num_input_tokens_seen": 20222780,
      "step": 954,
      "time_per_iteration": 3.651399850845337
    },
    {
      "auxiliary_loss_clip": 0.01241231,
      "auxiliary_loss_mlp": 0.01042695,
      "balance_loss_clip": 1.06105709,
      "balance_loss_mlp": 1.03070831,
      "epoch": 0.11483196056033187,
      "flos": 22526476594560.0,
      "grad_norm": 1.8215636428587698,
      "language_loss": 0.83965695,
      "learning_rate": 3.9250742546324786e-06,
      "loss": 0.86249626,
      "num_input_tokens_seen": 20243015,
      "step": 955,
      "time_per_iteration": 2.732365131378174
    },
    {
      "auxiliary_loss_clip": 0.01247019,
      "auxiliary_loss_mlp": 0.01039663,
      "balance_loss_clip": 1.06335115,
      "balance_loss_mlp": 1.02890456,
      "epoch": 0.11495220345097096,
      "flos": 28220292887040.0,
      "grad_norm": 1.8881281102154923,
      "language_loss": 0.86698067,
      "learning_rate": 3.924862890522683e-06,
      "loss": 0.88984752,
      "num_input_tokens_seen": 20263025,
      "step": 956,
      "time_per_iteration": 2.729874610900879
    },
    {
      "auxiliary_loss_clip": 0.01246167,
      "auxiliary_loss_mlp": 0.01035607,
      "balance_loss_clip": 1.0636003,
      "balance_loss_mlp": 1.02393055,
      "epoch": 0.11507244634161005,
      "flos": 17492267704320.0,
      "grad_norm": 2.315160791997177,
      "language_loss": 0.86306632,
      "learning_rate": 3.9246512344127174e-06,
      "loss": 0.88588405,
      "num_input_tokens_seen": 20280685,
      "step": 957,
      "time_per_iteration": 3.5540668964385986
    },
    {
      "auxiliary_loss_clip": 0.01232644,
      "auxiliary_loss_mlp": 0.01038202,
      "balance_loss_clip": 1.05520797,
      "balance_loss_mlp": 1.02665114,
      "epoch": 0.11519268923224914,
      "flos": 22564937082240.0,
      "grad_norm": 1.848704053352634,
      "language_loss": 0.82211173,
      "learning_rate": 3.9244392863346895e-06,
      "loss": 0.8448202,
      "num_input_tokens_seen": 20300090,
      "step": 958,
      "time_per_iteration": 2.729116201400757
    },
    {
      "auxiliary_loss_clip": 0.01253852,
      "auxiliary_loss_mlp": 0.01043148,
      "balance_loss_clip": 1.06901932,
      "balance_loss_mlp": 1.030339,
      "epoch": 0.11531293212288823,
      "flos": 16982839065600.0,
      "grad_norm": 1.8876877327136017,
      "language_loss": 0.9252581,
      "learning_rate": 3.9242270463207524e-06,
      "loss": 0.94822806,
      "num_input_tokens_seen": 20318480,
      "step": 959,
      "time_per_iteration": 3.594456672668457
    },
    {
      "auxiliary_loss_clip": 0.01236582,
      "auxiliary_loss_mlp": 0.01038229,
      "balance_loss_clip": 1.05872989,
      "balance_loss_mlp": 1.02617669,
      "epoch": 0.11543317501352733,
      "flos": 12422004537600.0,
      "grad_norm": 2.966335902426363,
      "language_loss": 0.85298157,
      "learning_rate": 3.924014514403102e-06,
      "loss": 0.87572968,
      "num_input_tokens_seen": 20334635,
      "step": 960,
      "time_per_iteration": 2.757211446762085
    },
    {
      "auxiliary_loss_clip": 0.01237805,
      "auxiliary_loss_mlp": 0.0104508,
      "balance_loss_clip": 1.05766606,
      "balance_loss_mlp": 1.03202701,
      "epoch": 0.11555341790416641,
      "flos": 19821648695040.0,
      "grad_norm": 2.276666506787491,
      "language_loss": 0.9132632,
      "learning_rate": 3.92380169061398e-06,
      "loss": 0.93609208,
      "num_input_tokens_seen": 20352415,
      "step": 961,
      "time_per_iteration": 2.7350287437438965
    },
    {
      "auxiliary_loss_clip": 0.01240766,
      "auxiliary_loss_mlp": 0.0120977,
      "balance_loss_clip": 1.05801857,
      "balance_loss_mlp": 1.00056434,
      "epoch": 0.11567366079480551,
      "flos": 25738865625600.0,
      "grad_norm": 2.564631505626613,
      "language_loss": 0.84200782,
      "learning_rate": 3.9235885749856705e-06,
      "loss": 0.86651313,
      "num_input_tokens_seen": 20371095,
      "step": 962,
      "time_per_iteration": 2.7739336490631104
    },
    {
      "auxiliary_loss_clip": 0.01251745,
      "auxiliary_loss_mlp": 0.01041748,
      "balance_loss_clip": 1.07025385,
      "balance_loss_mlp": 1.02946937,
      "epoch": 0.1157939036854446,
      "flos": 18223301301120.0,
      "grad_norm": 19.443639493518553,
      "language_loss": 0.82797551,
      "learning_rate": 3.9233751675505035e-06,
      "loss": 0.85091054,
      "num_input_tokens_seen": 20389805,
      "step": 963,
      "time_per_iteration": 2.700101375579834
    },
    {
      "auxiliary_loss_clip": 0.01243962,
      "auxiliary_loss_mlp": 0.01039548,
      "balance_loss_clip": 1.06640005,
      "balance_loss_mlp": 1.02676892,
      "epoch": 0.11591414657608369,
      "flos": 23073755189760.0,
      "grad_norm": 2.871154361871057,
      "language_loss": 0.84924161,
      "learning_rate": 3.923161468340853e-06,
      "loss": 0.87207675,
      "num_input_tokens_seen": 20409640,
      "step": 964,
      "time_per_iteration": 2.6961114406585693
    },
    {
      "auxiliary_loss_clip": 0.01237331,
      "auxiliary_loss_mlp": 0.01038977,
      "balance_loss_clip": 1.05639362,
      "balance_loss_mlp": 1.02738428,
      "epoch": 0.11603438946672277,
      "flos": 19461716461440.0,
      "grad_norm": 2.8802500976155767,
      "language_loss": 0.81778544,
      "learning_rate": 3.9229474773891374e-06,
      "loss": 0.84054857,
      "num_input_tokens_seen": 20428180,
      "step": 965,
      "time_per_iteration": 2.696503162384033
    },
    {
      "auxiliary_loss_clip": 0.01253735,
      "auxiliary_loss_mlp": 0.0104625,
      "balance_loss_clip": 1.05809271,
      "balance_loss_mlp": 1.03371501,
      "epoch": 0.11615463235736187,
      "flos": 26831986272000.0,
      "grad_norm": 2.5357576417151924,
      "language_loss": 0.83563268,
      "learning_rate": 3.922733194727818e-06,
      "loss": 0.85863256,
      "num_input_tokens_seen": 20447975,
      "step": 966,
      "time_per_iteration": 2.7304978370666504
    },
    {
      "auxiliary_loss_clip": 0.01252516,
      "auxiliary_loss_mlp": 0.0104021,
      "balance_loss_clip": 1.06790531,
      "balance_loss_mlp": 1.02778244,
      "epoch": 0.11627487524800097,
      "flos": 18580324533120.0,
      "grad_norm": 2.318576837551501,
      "language_loss": 0.8785181,
      "learning_rate": 3.922518620389402e-06,
      "loss": 0.90144533,
      "num_input_tokens_seen": 20464840,
      "step": 967,
      "time_per_iteration": 2.661937713623047
    },
    {
      "auxiliary_loss_clip": 0.01223374,
      "auxiliary_loss_mlp": 0.01041557,
      "balance_loss_clip": 1.05295074,
      "balance_loss_mlp": 1.02923083,
      "epoch": 0.11639511813864005,
      "flos": 18150474476160.0,
      "grad_norm": 1.8969510659670494,
      "language_loss": 0.89614916,
      "learning_rate": 3.922303754406439e-06,
      "loss": 0.91879845,
      "num_input_tokens_seen": 20482680,
      "step": 968,
      "time_per_iteration": 2.769179344177246
    },
    {
      "auxiliary_loss_clip": 0.01231909,
      "auxiliary_loss_mlp": 0.01046911,
      "balance_loss_clip": 1.05819023,
      "balance_loss_mlp": 1.03410184,
      "epoch": 0.11651536102927915,
      "flos": 20922023888640.0,
      "grad_norm": 1.7348813957775364,
      "language_loss": 0.78970194,
      "learning_rate": 3.922088596811526e-06,
      "loss": 0.81249017,
      "num_input_tokens_seen": 20501810,
      "step": 969,
      "time_per_iteration": 2.7430012226104736
    },
    {
      "auxiliary_loss_clip": 0.01234926,
      "auxiliary_loss_mlp": 0.01040768,
      "balance_loss_clip": 1.06387806,
      "balance_loss_mlp": 1.02976513,
      "epoch": 0.11663560391991823,
      "flos": 16508602776960.0,
      "grad_norm": 3.618921707702469,
      "language_loss": 0.86980963,
      "learning_rate": 3.9218731476373e-06,
      "loss": 0.89256662,
      "num_input_tokens_seen": 20517995,
      "step": 970,
      "time_per_iteration": 2.5986790657043457
    },
    {
      "auxiliary_loss_clip": 0.01252201,
      "auxiliary_loss_mlp": 0.01045446,
      "balance_loss_clip": 1.06839085,
      "balance_loss_mlp": 1.03270245,
      "epoch": 0.11675584681055733,
      "flos": 19865029345920.0,
      "grad_norm": 2.872333538928712,
      "language_loss": 0.84787649,
      "learning_rate": 3.9216574069164455e-06,
      "loss": 0.87085301,
      "num_input_tokens_seen": 20536970,
      "step": 971,
      "time_per_iteration": 2.7115817070007324
    },
    {
      "auxiliary_loss_clip": 0.01244531,
      "auxiliary_loss_mlp": 0.01040771,
      "balance_loss_clip": 1.06810927,
      "balance_loss_mlp": 1.02933884,
      "epoch": 0.11687608970119642,
      "flos": 21944364785280.0,
      "grad_norm": 1.6716424478610294,
      "language_loss": 0.80339825,
      "learning_rate": 3.921441374681691e-06,
      "loss": 0.82625127,
      "num_input_tokens_seen": 20557030,
      "step": 972,
      "time_per_iteration": 2.627502202987671
    },
    {
      "auxiliary_loss_clip": 0.01243548,
      "auxiliary_loss_mlp": 0.01038428,
      "balance_loss_clip": 1.06359529,
      "balance_loss_mlp": 1.02688241,
      "epoch": 0.1169963325918355,
      "flos": 24061155131520.0,
      "grad_norm": 1.7874320497892329,
      "language_loss": 0.65171689,
      "learning_rate": 3.921225050965808e-06,
      "loss": 0.67453671,
      "num_input_tokens_seen": 20576915,
      "step": 973,
      "time_per_iteration": 2.7379274368286133
    },
    {
      "auxiliary_loss_clip": 0.01244716,
      "auxiliary_loss_mlp": 0.01040011,
      "balance_loss_clip": 1.06072092,
      "balance_loss_mlp": 1.02785718,
      "epoch": 0.1171165754824746,
      "flos": 23368151059200.0,
      "grad_norm": 2.2490641244991525,
      "language_loss": 0.74892658,
      "learning_rate": 3.921008435801612e-06,
      "loss": 0.77177382,
      "num_input_tokens_seen": 20596000,
      "step": 974,
      "time_per_iteration": 2.702404499053955
    },
    {
      "auxiliary_loss_clip": 0.01235542,
      "auxiliary_loss_mlp": 0.01039679,
      "balance_loss_clip": 1.06522751,
      "balance_loss_mlp": 1.02745414,
      "epoch": 0.11723681837311369,
      "flos": 18552243075840.0,
      "grad_norm": 2.2694228628990736,
      "language_loss": 0.75601447,
      "learning_rate": 3.920791529221963e-06,
      "loss": 0.77876669,
      "num_input_tokens_seen": 20614675,
      "step": 975,
      "time_per_iteration": 2.6542985439300537
    },
    {
      "auxiliary_loss_clip": 0.01247334,
      "auxiliary_loss_mlp": 0.01209921,
      "balance_loss_clip": 1.06242871,
      "balance_loss_mlp": 1.00064158,
      "epoch": 0.11735706126375278,
      "flos": 23550541344000.0,
      "grad_norm": 2.0481331943141754,
      "language_loss": 0.762564,
      "learning_rate": 3.920574331259768e-06,
      "loss": 0.78713655,
      "num_input_tokens_seen": 20635875,
      "step": 976,
      "time_per_iteration": 2.677151679992676
    },
    {
      "auxiliary_loss_clip": 0.01236284,
      "auxiliary_loss_mlp": 0.01034193,
      "balance_loss_clip": 1.06102943,
      "balance_loss_mlp": 1.02354741,
      "epoch": 0.11747730415439187,
      "flos": 22381541216640.0,
      "grad_norm": 2.2011399162955025,
      "language_loss": 0.79352719,
      "learning_rate": 3.9203568419479716e-06,
      "loss": 0.81623197,
      "num_input_tokens_seen": 20656430,
      "step": 977,
      "time_per_iteration": 2.699854850769043
    },
    {
      "auxiliary_loss_clip": 0.01245297,
      "auxiliary_loss_mlp": 0.01029999,
      "balance_loss_clip": 1.06369758,
      "balance_loss_mlp": 1.01922262,
      "epoch": 0.11759754704503096,
      "flos": 22200731130240.0,
      "grad_norm": 2.2762613574775545,
      "language_loss": 0.74923265,
      "learning_rate": 3.92013906131957e-06,
      "loss": 0.77198553,
      "num_input_tokens_seen": 20675360,
      "step": 978,
      "time_per_iteration": 2.666102170944214
    },
    {
      "auxiliary_loss_clip": 0.01247878,
      "auxiliary_loss_mlp": 0.0105261,
      "balance_loss_clip": 1.0642848,
      "balance_loss_mlp": 1.04217315,
      "epoch": 0.11771778993567006,
      "flos": 22309755886080.0,
      "grad_norm": 1.972173016837716,
      "language_loss": 0.82802701,
      "learning_rate": 3.9199209894076e-06,
      "loss": 0.85103184,
      "num_input_tokens_seen": 20695675,
      "step": 979,
      "time_per_iteration": 2.787997007369995
    },
    {
      "auxiliary_loss_clip": 0.01245717,
      "auxiliary_loss_mlp": 0.01037604,
      "balance_loss_clip": 1.06643486,
      "balance_loss_mlp": 1.0245266,
      "epoch": 0.11783803282630914,
      "flos": 21288169175040.0,
      "grad_norm": 2.135104138169243,
      "language_loss": 0.89840877,
      "learning_rate": 3.919702626245142e-06,
      "loss": 0.92124194,
      "num_input_tokens_seen": 20715330,
      "step": 980,
      "time_per_iteration": 2.741669178009033
    },
    {
      "auxiliary_loss_clip": 0.01232881,
      "auxiliary_loss_mlp": 0.01035882,
      "balance_loss_clip": 1.06089902,
      "balance_loss_mlp": 1.02416992,
      "epoch": 0.11795827571694824,
      "flos": 25371535190400.0,
      "grad_norm": 3.2519229821893454,
      "language_loss": 0.66894549,
      "learning_rate": 3.919483971865322e-06,
      "loss": 0.69163311,
      "num_input_tokens_seen": 20735325,
      "step": 981,
      "time_per_iteration": 3.571516275405884
    },
    {
      "auxiliary_loss_clip": 0.01243497,
      "auxiliary_loss_mlp": 0.01039035,
      "balance_loss_clip": 1.06504095,
      "balance_loss_mlp": 1.02791858,
      "epoch": 0.11807851860758732,
      "flos": 23622218933760.0,
      "grad_norm": 2.322091805919089,
      "language_loss": 0.87928498,
      "learning_rate": 3.91926502630131e-06,
      "loss": 0.90211022,
      "num_input_tokens_seen": 20755940,
      "step": 982,
      "time_per_iteration": 2.6944220066070557
    },
    {
      "auxiliary_loss_clip": 0.01250773,
      "auxiliary_loss_mlp": 0.01041307,
      "balance_loss_clip": 1.07118702,
      "balance_loss_mlp": 1.03005958,
      "epoch": 0.11819876149822642,
      "flos": 24972496024320.0,
      "grad_norm": 3.246570643809827,
      "language_loss": 0.7229048,
      "learning_rate": 3.91904578958632e-06,
      "loss": 0.74582553,
      "num_input_tokens_seen": 20775355,
      "step": 983,
      "time_per_iteration": 2.6784870624542236
    },
    {
      "auxiliary_loss_clip": 0.0124737,
      "auxiliary_loss_mlp": 0.01045367,
      "balance_loss_clip": 1.06894159,
      "balance_loss_mlp": 1.03351736,
      "epoch": 0.11831900438886551,
      "flos": 23003226835200.0,
      "grad_norm": 3.7417997632797335,
      "language_loss": 0.84205377,
      "learning_rate": 3.918826261753608e-06,
      "loss": 0.86498117,
      "num_input_tokens_seen": 20794935,
      "step": 984,
      "time_per_iteration": 3.5323898792266846
    },
    {
      "auxiliary_loss_clip": 0.01244287,
      "auxiliary_loss_mlp": 0.01032144,
      "balance_loss_clip": 1.06284308,
      "balance_loss_mlp": 1.022017,
      "epoch": 0.1184392472795046,
      "flos": 27965147604480.0,
      "grad_norm": 2.6313111415915724,
      "language_loss": 0.71216458,
      "learning_rate": 3.918606442836478e-06,
      "loss": 0.73492885,
      "num_input_tokens_seen": 20817155,
      "step": 985,
      "time_per_iteration": 2.7864856719970703
    },
    {
      "auxiliary_loss_clip": 0.01243661,
      "auxiliary_loss_mlp": 0.01037986,
      "balance_loss_clip": 1.06885338,
      "balance_loss_mlp": 1.02727461,
      "epoch": 0.1185594901701437,
      "flos": 19898497843200.0,
      "grad_norm": 2.148243641752004,
      "language_loss": 0.77725017,
      "learning_rate": 3.918386332868277e-06,
      "loss": 0.80006665,
      "num_input_tokens_seen": 20835125,
      "step": 986,
      "time_per_iteration": 4.52485203742981
    },
    {
      "auxiliary_loss_clip": 0.01235675,
      "auxiliary_loss_mlp": 0.01042419,
      "balance_loss_clip": 1.06432104,
      "balance_loss_mlp": 1.03084993,
      "epoch": 0.11867973306078278,
      "flos": 18912354877440.0,
      "grad_norm": 3.0797518344736687,
      "language_loss": 0.94552463,
      "learning_rate": 3.918165931882394e-06,
      "loss": 0.96830553,
      "num_input_tokens_seen": 20853525,
      "step": 987,
      "time_per_iteration": 2.6054766178131104
    },
    {
      "auxiliary_loss_clip": 0.01235925,
      "auxiliary_loss_mlp": 0.01037362,
      "balance_loss_clip": 1.05230117,
      "balance_loss_mlp": 1.02545261,
      "epoch": 0.11879997595142187,
      "flos": 16982803152000.0,
      "grad_norm": 2.4986507205779662,
      "language_loss": 0.75884897,
      "learning_rate": 3.917945239912264e-06,
      "loss": 0.78158182,
      "num_input_tokens_seen": 20871000,
      "step": 988,
      "time_per_iteration": 2.7343733310699463
    },
    {
      "auxiliary_loss_clip": 0.01243484,
      "auxiliary_loss_mlp": 0.01038324,
      "balance_loss_clip": 1.05811954,
      "balance_loss_mlp": 1.02800012,
      "epoch": 0.11892021884206096,
      "flos": 17530369056000.0,
      "grad_norm": 3.58504154875349,
      "language_loss": 0.75689304,
      "learning_rate": 3.917724256991367e-06,
      "loss": 0.77971113,
      "num_input_tokens_seen": 20889745,
      "step": 989,
      "time_per_iteration": 2.849810838699341
    },
    {
      "auxiliary_loss_clip": 0.01234852,
      "auxiliary_loss_mlp": 0.01047993,
      "balance_loss_clip": 1.06206453,
      "balance_loss_mlp": 1.03665638,
      "epoch": 0.11904046173270005,
      "flos": 30955895763840.0,
      "grad_norm": 2.2412671323073954,
      "language_loss": 0.81953108,
      "learning_rate": 3.9175029831532245e-06,
      "loss": 0.84235954,
      "num_input_tokens_seen": 20909260,
      "step": 990,
      "time_per_iteration": 2.7362546920776367
    },
    {
      "auxiliary_loss_clip": 0.01250948,
      "auxiliary_loss_mlp": 0.01031808,
      "balance_loss_clip": 1.06698871,
      "balance_loss_mlp": 1.02168107,
      "epoch": 0.11916070462333915,
      "flos": 20157234485760.0,
      "grad_norm": 3.354857935587684,
      "language_loss": 0.88837719,
      "learning_rate": 3.917281418431404e-06,
      "loss": 0.91120476,
      "num_input_tokens_seen": 20928305,
      "step": 991,
      "time_per_iteration": 2.6597273349761963
    },
    {
      "auxiliary_loss_clip": 0.01244127,
      "auxiliary_loss_mlp": 0.01042918,
      "balance_loss_clip": 1.06647038,
      "balance_loss_mlp": 1.03073478,
      "epoch": 0.11928094751397823,
      "flos": 23551115961600.0,
      "grad_norm": 1.9044991873501171,
      "language_loss": 0.77404773,
      "learning_rate": 3.917059562859516e-06,
      "loss": 0.79691821,
      "num_input_tokens_seen": 20947630,
      "step": 992,
      "time_per_iteration": 2.7316441535949707
    },
    {
      "auxiliary_loss_clip": 0.01239151,
      "auxiliary_loss_mlp": 0.0105076,
      "balance_loss_clip": 1.06720483,
      "balance_loss_mlp": 1.03804016,
      "epoch": 0.11940119040461733,
      "flos": 23908426502400.0,
      "grad_norm": 2.014700183192491,
      "language_loss": 0.88700801,
      "learning_rate": 3.916837416471218e-06,
      "loss": 0.9099071,
      "num_input_tokens_seen": 20964250,
      "step": 993,
      "time_per_iteration": 2.6128737926483154
    },
    {
      "auxiliary_loss_clip": 0.01238461,
      "auxiliary_loss_mlp": 0.01041087,
      "balance_loss_clip": 1.06321657,
      "balance_loss_mlp": 1.02993512,
      "epoch": 0.11952143329525641,
      "flos": 13844533835520.0,
      "grad_norm": 2.631598412806415,
      "language_loss": 0.72312456,
      "learning_rate": 3.916614979300207e-06,
      "loss": 0.74592006,
      "num_input_tokens_seen": 20979095,
      "step": 994,
      "time_per_iteration": 2.6879265308380127
    },
    {
      "auxiliary_loss_clip": 0.01237779,
      "auxiliary_loss_mlp": 0.010396,
      "balance_loss_clip": 1.06017828,
      "balance_loss_mlp": 1.02910972,
      "epoch": 0.11964167618589551,
      "flos": 27015525792000.0,
      "grad_norm": 1.8619851168956714,
      "language_loss": 0.78876275,
      "learning_rate": 3.9163922513802274e-06,
      "loss": 0.81153655,
      "num_input_tokens_seen": 21001430,
      "step": 995,
      "time_per_iteration": 2.7834954261779785
    },
    {
      "auxiliary_loss_clip": 0.0124675,
      "auxiliary_loss_mlp": 0.01036228,
      "balance_loss_clip": 1.0684371,
      "balance_loss_mlp": 1.02480769,
      "epoch": 0.1197619190765346,
      "flos": 12567622273920.0,
      "grad_norm": 2.8144454315196112,
      "language_loss": 0.8239429,
      "learning_rate": 3.916169232745067e-06,
      "loss": 0.84677267,
      "num_input_tokens_seen": 21019105,
      "step": 996,
      "time_per_iteration": 2.7013843059539795
    },
    {
      "auxiliary_loss_clip": 0.01233833,
      "auxiliary_loss_mlp": 0.01043853,
      "balance_loss_clip": 1.06197,
      "balance_loss_mlp": 1.03224826,
      "epoch": 0.11988216196717369,
      "flos": 16909437623040.0,
      "grad_norm": 2.8725616377195577,
      "language_loss": 0.91867214,
      "learning_rate": 3.915945923428559e-06,
      "loss": 0.94144905,
      "num_input_tokens_seen": 21035630,
      "step": 997,
      "time_per_iteration": 2.6573352813720703
    },
    {
      "auxiliary_loss_clip": 0.01240231,
      "auxiliary_loss_mlp": 0.01037608,
      "balance_loss_clip": 1.06383836,
      "balance_loss_mlp": 1.02584171,
      "epoch": 0.12000240485781279,
      "flos": 16216577205120.0,
      "grad_norm": 2.794758880327502,
      "language_loss": 0.83227551,
      "learning_rate": 3.915722323464577e-06,
      "loss": 0.8550539,
      "num_input_tokens_seen": 21054235,
      "step": 998,
      "time_per_iteration": 2.6055469512939453
    },
    {
      "auxiliary_loss_clip": 0.01244735,
      "auxiliary_loss_mlp": 0.01040613,
      "balance_loss_clip": 1.06652939,
      "balance_loss_mlp": 1.02897859,
      "epoch": 0.12012264774845187,
      "flos": 49344887525760.0,
      "grad_norm": 2.258363437020132,
      "language_loss": 0.70817041,
      "learning_rate": 3.91549843288704e-06,
      "loss": 0.73102391,
      "num_input_tokens_seen": 21077915,
      "step": 999,
      "time_per_iteration": 2.911414623260498
    },
    {
      "auxiliary_loss_clip": 0.01243325,
      "auxiliary_loss_mlp": 0.01209663,
      "balance_loss_clip": 1.0579989,
      "balance_loss_mlp": 1.00058734,
      "epoch": 0.12024289063909097,
      "flos": 26979435601920.0,
      "grad_norm": 2.092210396412502,
      "language_loss": 0.79504532,
      "learning_rate": 3.915274251729916e-06,
      "loss": 0.81957519,
      "num_input_tokens_seen": 21099205,
      "step": 1000,
      "time_per_iteration": 2.7195065021514893
    },
    {
      "auxiliary_loss_clip": 0.01246973,
      "auxiliary_loss_mlp": 0.0103452,
      "balance_loss_clip": 1.06388366,
      "balance_loss_mlp": 1.02318954,
      "epoch": 0.12036313352973005,
      "flos": 19537308633600.0,
      "grad_norm": 3.6876867957172625,
      "language_loss": 0.90393138,
      "learning_rate": 3.91504978002721e-06,
      "loss": 0.92674637,
      "num_input_tokens_seen": 21118260,
      "step": 1001,
      "time_per_iteration": 2.733802318572998
    },
    {
      "auxiliary_loss_clip": 0.01248605,
      "auxiliary_loss_mlp": 0.01209556,
      "balance_loss_clip": 1.06321704,
      "balance_loss_mlp": 1.00055754,
      "epoch": 0.12048337642036915,
      "flos": 17268256535040.0,
      "grad_norm": 2.1161658123597635,
      "language_loss": 0.7631076,
      "learning_rate": 3.914825017812974e-06,
      "loss": 0.78768921,
      "num_input_tokens_seen": 21134910,
      "step": 1002,
      "time_per_iteration": 2.6309218406677246
    },
    {
      "auxiliary_loss_clip": 0.01246442,
      "auxiliary_loss_mlp": 0.01037648,
      "balance_loss_clip": 1.06610584,
      "balance_loss_mlp": 1.02601898,
      "epoch": 0.12060361931100824,
      "flos": 22856962654080.0,
      "grad_norm": 2.5646639298165916,
      "language_loss": 0.72868526,
      "learning_rate": 3.9145999651213065e-06,
      "loss": 0.75152612,
      "num_input_tokens_seen": 21154150,
      "step": 1003,
      "time_per_iteration": 2.711174726486206
    },
    {
      "auxiliary_loss_clip": 0.01245329,
      "auxiliary_loss_mlp": 0.01042263,
      "balance_loss_clip": 1.06603527,
      "balance_loss_mlp": 1.03006184,
      "epoch": 0.12072386220164733,
      "flos": 16726795943040.0,
      "grad_norm": 3.7334468077405525,
      "language_loss": 0.88510609,
      "learning_rate": 3.9143746219863465e-06,
      "loss": 0.90798199,
      "num_input_tokens_seen": 21171255,
      "step": 1004,
      "time_per_iteration": 2.594132900238037
    },
    {
      "auxiliary_loss_clip": 0.01163577,
      "auxiliary_loss_mlp": 0.01006451,
      "balance_loss_clip": 1.03809845,
      "balance_loss_mlp": 1.00301743,
      "epoch": 0.12084410509228642,
      "flos": 55144176105600.0,
      "grad_norm": 0.9648021106793967,
      "language_loss": 0.647883,
      "learning_rate": 3.914148988442278e-06,
      "loss": 0.66958326,
      "num_input_tokens_seen": 21227045,
      "step": 1005,
      "time_per_iteration": 3.1365630626678467
    },
    {
      "auxiliary_loss_clip": 0.01232422,
      "auxiliary_loss_mlp": 0.01036256,
      "balance_loss_clip": 1.0620687,
      "balance_loss_mlp": 1.02457905,
      "epoch": 0.1209643479829255,
      "flos": 26760236855040.0,
      "grad_norm": 3.5330631914959443,
      "language_loss": 0.95630825,
      "learning_rate": 3.91392306452333e-06,
      "loss": 0.97899503,
      "num_input_tokens_seen": 21244120,
      "step": 1006,
      "time_per_iteration": 2.6930930614471436
    },
    {
      "auxiliary_loss_clip": 0.01248539,
      "auxiliary_loss_mlp": 0.01035651,
      "balance_loss_clip": 1.06952333,
      "balance_loss_mlp": 1.02449322,
      "epoch": 0.1210845908735646,
      "flos": 11035026725760.0,
      "grad_norm": 2.806972835746599,
      "language_loss": 0.66572618,
      "learning_rate": 3.913696850263774e-06,
      "loss": 0.68856812,
      "num_input_tokens_seen": 21258485,
      "step": 1007,
      "time_per_iteration": 2.681570053100586
    },
    {
      "auxiliary_loss_clip": 0.01243737,
      "auxiliary_loss_mlp": 0.010358,
      "balance_loss_clip": 1.06591582,
      "balance_loss_mlp": 1.02450454,
      "epoch": 0.1212048337642037,
      "flos": 20484631975680.0,
      "grad_norm": 2.022092103503406,
      "language_loss": 0.79114687,
      "learning_rate": 3.913470345697929e-06,
      "loss": 0.81394225,
      "num_input_tokens_seen": 21277115,
      "step": 1008,
      "time_per_iteration": 3.6168770790100098
    },
    {
      "auxiliary_loss_clip": 0.01250219,
      "auxiliary_loss_mlp": 0.01042809,
      "balance_loss_clip": 1.06103182,
      "balance_loss_mlp": 1.03195548,
      "epoch": 0.12132507665484278,
      "flos": 22346061557760.0,
      "grad_norm": 2.0916383568813828,
      "language_loss": 0.85618931,
      "learning_rate": 3.913243550860153e-06,
      "loss": 0.87911958,
      "num_input_tokens_seen": 21294880,
      "step": 1009,
      "time_per_iteration": 2.6710257530212402
    },
    {
      "auxiliary_loss_clip": 0.01251516,
      "auxiliary_loss_mlp": 0.01043238,
      "balance_loss_clip": 1.07257962,
      "balance_loss_mlp": 1.03131747,
      "epoch": 0.12144531954548188,
      "flos": 29314957818240.0,
      "grad_norm": 2.065180717899735,
      "language_loss": 0.76159549,
      "learning_rate": 3.913016465784852e-06,
      "loss": 0.7845431,
      "num_input_tokens_seen": 21315555,
      "step": 1010,
      "time_per_iteration": 3.6980063915252686
    },
    {
      "auxiliary_loss_clip": 0.01246816,
      "auxiliary_loss_mlp": 0.01036918,
      "balance_loss_clip": 1.05811632,
      "balance_loss_mlp": 1.02490115,
      "epoch": 0.12156556243612096,
      "flos": 20485242506880.0,
      "grad_norm": 3.0461845921970148,
      "language_loss": 0.71724856,
      "learning_rate": 3.912789090506474e-06,
      "loss": 0.74008584,
      "num_input_tokens_seen": 21334815,
      "step": 1011,
      "time_per_iteration": 2.7149815559387207
    },
    {
      "auxiliary_loss_clip": 0.01254425,
      "auxiliary_loss_mlp": 0.01045558,
      "balance_loss_clip": 1.06150031,
      "balance_loss_mlp": 1.03317797,
      "epoch": 0.12168580532676006,
      "flos": 16472009796480.0,
      "grad_norm": 2.7054485056935134,
      "language_loss": 0.7205978,
      "learning_rate": 3.9125614250595114e-06,
      "loss": 0.74359763,
      "num_input_tokens_seen": 21351025,
      "step": 1012,
      "time_per_iteration": 3.6152279376983643
    },
    {
      "auxiliary_loss_clip": 0.01245102,
      "auxiliary_loss_mlp": 0.01036652,
      "balance_loss_clip": 1.0650034,
      "balance_loss_mlp": 1.02464175,
      "epoch": 0.12180604821739914,
      "flos": 15341290588800.0,
      "grad_norm": 3.0440585572746213,
      "language_loss": 0.88997269,
      "learning_rate": 3.912333469478502e-06,
      "loss": 0.91279018,
      "num_input_tokens_seen": 21368990,
      "step": 1013,
      "time_per_iteration": 3.533604383468628
    },
    {
      "auxiliary_loss_clip": 0.01243441,
      "auxiliary_loss_mlp": 0.01032647,
      "balance_loss_clip": 1.06184995,
      "balance_loss_mlp": 1.02173305,
      "epoch": 0.12192629110803824,
      "flos": 19318038059520.0,
      "grad_norm": 2.7870906473496904,
      "language_loss": 0.78339368,
      "learning_rate": 3.912105223798025e-06,
      "loss": 0.80615461,
      "num_input_tokens_seen": 21388410,
      "step": 1014,
      "time_per_iteration": 2.6517655849456787
    },
    {
      "auxiliary_loss_clip": 0.01160189,
      "auxiliary_loss_mlp": 0.01007032,
      "balance_loss_clip": 1.02968025,
      "balance_loss_mlp": 1.00336063,
      "epoch": 0.12204653399867733,
      "flos": 47725354085760.0,
      "grad_norm": 0.9971720910549715,
      "language_loss": 0.67680073,
      "learning_rate": 3.9118766880527065e-06,
      "loss": 0.69847298,
      "num_input_tokens_seen": 21442845,
      "step": 1015,
      "time_per_iteration": 3.0868778228759766
    },
    {
      "auxiliary_loss_clip": 0.01237554,
      "auxiliary_loss_mlp": 0.01033034,
      "balance_loss_clip": 1.05740714,
      "balance_loss_mlp": 1.02235913,
      "epoch": 0.12216677688931642,
      "flos": 18221936584320.0,
      "grad_norm": 2.2154659024682517,
      "language_loss": 0.73906851,
      "learning_rate": 3.9116478622772145e-06,
      "loss": 0.76177442,
      "num_input_tokens_seen": 21461420,
      "step": 1016,
      "time_per_iteration": 2.7062487602233887
    },
    {
      "auxiliary_loss_clip": 0.01242483,
      "auxiliary_loss_mlp": 0.0104661,
      "balance_loss_clip": 1.06614733,
      "balance_loss_mlp": 1.03479075,
      "epoch": 0.12228701977995551,
      "flos": 27525636789120.0,
      "grad_norm": 1.7161849780077119,
      "language_loss": 0.88132101,
      "learning_rate": 3.911418746506261e-06,
      "loss": 0.904212,
      "num_input_tokens_seen": 21481550,
      "step": 1017,
      "time_per_iteration": 2.7500791549682617
    },
    {
      "auxiliary_loss_clip": 0.01249783,
      "auxiliary_loss_mlp": 0.01048095,
      "balance_loss_clip": 1.07044625,
      "balance_loss_mlp": 1.03631735,
      "epoch": 0.1224072626705946,
      "flos": 21798136517760.0,
      "grad_norm": 1.7806425615249883,
      "language_loss": 0.78549588,
      "learning_rate": 3.911189340774604e-06,
      "loss": 0.80847472,
      "num_input_tokens_seen": 21501680,
      "step": 1018,
      "time_per_iteration": 2.6735448837280273
    },
    {
      "auxiliary_loss_clip": 0.01255489,
      "auxiliary_loss_mlp": 0.01041584,
      "balance_loss_clip": 1.06545854,
      "balance_loss_mlp": 1.02980626,
      "epoch": 0.1225275055612337,
      "flos": 20703758895360.0,
      "grad_norm": 2.5901732547106344,
      "language_loss": 0.7946105,
      "learning_rate": 3.910959645117043e-06,
      "loss": 0.81758118,
      "num_input_tokens_seen": 21521015,
      "step": 1019,
      "time_per_iteration": 2.6821625232696533
    },
    {
      "auxiliary_loss_clip": 0.01149108,
      "auxiliary_loss_mlp": 0.01201065,
      "balance_loss_clip": 1.03243327,
      "balance_loss_mlp": 1.00014734,
      "epoch": 0.12264774845187278,
      "flos": 57745294462080.0,
      "grad_norm": 0.8305373618240293,
      "language_loss": 0.56701285,
      "learning_rate": 3.910729659568423e-06,
      "loss": 0.59051454,
      "num_input_tokens_seen": 21578200,
      "step": 1020,
      "time_per_iteration": 3.208744764328003
    },
    {
      "auxiliary_loss_clip": 0.01247816,
      "auxiliary_loss_mlp": 0.01044172,
      "balance_loss_clip": 1.06601191,
      "balance_loss_mlp": 1.033324,
      "epoch": 0.12276799134251187,
      "flos": 26396282298240.0,
      "grad_norm": 2.1788363354472327,
      "language_loss": 0.82268435,
      "learning_rate": 3.9104993841636344e-06,
      "loss": 0.84560424,
      "num_input_tokens_seen": 21598770,
      "step": 1021,
      "time_per_iteration": 2.7502827644348145
    },
    {
      "auxiliary_loss_clip": 0.01247829,
      "auxiliary_loss_mlp": 0.01209304,
      "balance_loss_clip": 1.06877148,
      "balance_loss_mlp": 1.00070477,
      "epoch": 0.12288823423315097,
      "flos": 21064193919360.0,
      "grad_norm": 1.717383935467254,
      "language_loss": 0.81052327,
      "learning_rate": 3.910268818937608e-06,
      "loss": 0.83509457,
      "num_input_tokens_seen": 21616925,
      "step": 1022,
      "time_per_iteration": 2.750255823135376
    },
    {
      "auxiliary_loss_clip": 0.01247577,
      "auxiliary_loss_mlp": 0.01038351,
      "balance_loss_clip": 1.06257105,
      "balance_loss_mlp": 1.02697837,
      "epoch": 0.12300847712379005,
      "flos": 12312441077760.0,
      "grad_norm": 3.4585452561137644,
      "language_loss": 0.87512541,
      "learning_rate": 3.9100379639253196e-06,
      "loss": 0.89798468,
      "num_input_tokens_seen": 21633645,
      "step": 1023,
      "time_per_iteration": 2.680448293685913
    },
    {
      "auxiliary_loss_clip": 0.01244209,
      "auxiliary_loss_mlp": 0.01041159,
      "balance_loss_clip": 1.06066942,
      "balance_loss_mlp": 1.02935696,
      "epoch": 0.12312872001442915,
      "flos": 16762239688320.0,
      "grad_norm": 2.9229796106674675,
      "language_loss": 0.8643859,
      "learning_rate": 3.909806819161791e-06,
      "loss": 0.88723958,
      "num_input_tokens_seen": 21649120,
      "step": 1024,
      "time_per_iteration": 2.709076166152954
    },
    {
      "auxiliary_loss_clip": 0.0125023,
      "auxiliary_loss_mlp": 0.01032152,
      "balance_loss_clip": 1.06193042,
      "balance_loss_mlp": 1.02042222,
      "epoch": 0.12324896290506823,
      "flos": 18404937400320.0,
      "grad_norm": 2.0310582140210736,
      "language_loss": 0.86301655,
      "learning_rate": 3.909575384682086e-06,
      "loss": 0.8858403,
      "num_input_tokens_seen": 21668000,
      "step": 1025,
      "time_per_iteration": 2.7299606800079346
    },
    {
      "auxiliary_loss_clip": 0.01247836,
      "auxiliary_loss_mlp": 0.01053901,
      "balance_loss_clip": 1.06666613,
      "balance_loss_mlp": 1.04164624,
      "epoch": 0.12336920579570733,
      "flos": 18915407533440.0,
      "grad_norm": 2.3586704526095614,
      "language_loss": 0.69258916,
      "learning_rate": 3.9093436605213144e-06,
      "loss": 0.71560657,
      "num_input_tokens_seen": 21688500,
      "step": 1026,
      "time_per_iteration": 2.6266136169433594
    },
    {
      "auxiliary_loss_clip": 0.01248553,
      "auxiliary_loss_mlp": 0.01046003,
      "balance_loss_clip": 1.06480503,
      "balance_loss_mlp": 1.03495204,
      "epoch": 0.12348944868634643,
      "flos": 23878369797120.0,
      "grad_norm": 2.4836240570145427,
      "language_loss": 0.79593742,
      "learning_rate": 3.909111646714627e-06,
      "loss": 0.81888294,
      "num_input_tokens_seen": 21709345,
      "step": 1027,
      "time_per_iteration": 2.741152048110962
    },
    {
      "auxiliary_loss_clip": 0.01240747,
      "auxiliary_loss_mlp": 0.01034459,
      "balance_loss_clip": 1.06697416,
      "balance_loss_mlp": 1.02369487,
      "epoch": 0.12360969157698551,
      "flos": 19026084314880.0,
      "grad_norm": 2.065322196461481,
      "language_loss": 0.72507071,
      "learning_rate": 3.9088793432972206e-06,
      "loss": 0.7478227,
      "num_input_tokens_seen": 21728165,
      "step": 1028,
      "time_per_iteration": 2.6338231563568115
    },
    {
      "auxiliary_loss_clip": 0.01248192,
      "auxiliary_loss_mlp": 0.01040946,
      "balance_loss_clip": 1.06201911,
      "balance_loss_mlp": 1.02945995,
      "epoch": 0.1237299344676246,
      "flos": 13224607983360.0,
      "grad_norm": 2.766248027726742,
      "language_loss": 0.82269323,
      "learning_rate": 3.908646750304336e-06,
      "loss": 0.84558463,
      "num_input_tokens_seen": 21745850,
      "step": 1029,
      "time_per_iteration": 2.7245287895202637
    },
    {
      "auxiliary_loss_clip": 0.01248921,
      "auxiliary_loss_mlp": 0.01039361,
      "balance_loss_clip": 1.06574619,
      "balance_loss_mlp": 1.0281198,
      "epoch": 0.12385017735826369,
      "flos": 20485673470080.0,
      "grad_norm": 2.3863026130149922,
      "language_loss": 0.87519062,
      "learning_rate": 3.908413867771257e-06,
      "loss": 0.89807343,
      "num_input_tokens_seen": 21764760,
      "step": 1030,
      "time_per_iteration": 2.6377248764038086
    },
    {
      "auxiliary_loss_clip": 0.01245452,
      "auxiliary_loss_mlp": 0.01043365,
      "balance_loss_clip": 1.06850863,
      "balance_loss_mlp": 1.03122973,
      "epoch": 0.12397042024890279,
      "flos": 17347835116800.0,
      "grad_norm": 1.826294318629917,
      "language_loss": 0.808321,
      "learning_rate": 3.908180695733311e-06,
      "loss": 0.83120912,
      "num_input_tokens_seen": 21784250,
      "step": 1031,
      "time_per_iteration": 2.643932819366455
    },
    {
      "auxiliary_loss_clip": 0.01225909,
      "auxiliary_loss_mlp": 0.01047805,
      "balance_loss_clip": 1.05397511,
      "balance_loss_mlp": 1.03628373,
      "epoch": 0.12409066313954187,
      "flos": 20412343854720.0,
      "grad_norm": 8.134708573396336,
      "language_loss": 0.82747358,
      "learning_rate": 3.907947234225871e-06,
      "loss": 0.85021067,
      "num_input_tokens_seen": 21803260,
      "step": 1032,
      "time_per_iteration": 2.6893198490142822
    },
    {
      "auxiliary_loss_clip": 0.01244304,
      "auxiliary_loss_mlp": 0.01038851,
      "balance_loss_clip": 1.05849242,
      "balance_loss_mlp": 1.027776,
      "epoch": 0.12421090603018096,
      "flos": 20736688688640.0,
      "grad_norm": 1.9232516138259979,
      "language_loss": 0.87401736,
      "learning_rate": 3.907713483284352e-06,
      "loss": 0.89684886,
      "num_input_tokens_seen": 21822735,
      "step": 1033,
      "time_per_iteration": 2.7310149669647217
    },
    {
      "auxiliary_loss_clip": 0.01237794,
      "auxiliary_loss_mlp": 0.01043904,
      "balance_loss_clip": 1.05102253,
      "balance_loss_mlp": 1.0303973,
      "epoch": 0.12433114892082006,
      "flos": 24498834353280.0,
      "grad_norm": 2.386476360421127,
      "language_loss": 0.9754144,
      "learning_rate": 3.907479442944216e-06,
      "loss": 0.99823135,
      "num_input_tokens_seen": 21841140,
      "step": 1034,
      "time_per_iteration": 2.7859933376312256
    },
    {
      "auxiliary_loss_clip": 0.01244932,
      "auxiliary_loss_mlp": 0.01037143,
      "balance_loss_clip": 1.06880331,
      "balance_loss_mlp": 1.02700973,
      "epoch": 0.12445139181145914,
      "flos": 19682315838720.0,
      "grad_norm": 4.051073608890809,
      "language_loss": 0.92635125,
      "learning_rate": 3.907245113240963e-06,
      "loss": 0.94917202,
      "num_input_tokens_seen": 21859260,
      "step": 1035,
      "time_per_iteration": 3.554722785949707
    },
    {
      "auxiliary_loss_clip": 0.01245053,
      "auxiliary_loss_mlp": 0.01037779,
      "balance_loss_clip": 1.05935204,
      "balance_loss_mlp": 1.02619219,
      "epoch": 0.12457163470209824,
      "flos": 46423087522560.0,
      "grad_norm": 1.7418514470881425,
      "language_loss": 0.73827398,
      "learning_rate": 3.907010494210144e-06,
      "loss": 0.76110232,
      "num_input_tokens_seen": 21881920,
      "step": 1036,
      "time_per_iteration": 3.8579976558685303
    },
    {
      "auxiliary_loss_clip": 0.01247206,
      "auxiliary_loss_mlp": 0.01046418,
      "balance_loss_clip": 1.06836939,
      "balance_loss_mlp": 1.03397262,
      "epoch": 0.12469187759273732,
      "flos": 20376289578240.0,
      "grad_norm": 2.19570643243272,
      "language_loss": 0.92078352,
      "learning_rate": 3.9067755858873495e-06,
      "loss": 0.94371974,
      "num_input_tokens_seen": 21898720,
      "step": 1037,
      "time_per_iteration": 2.6754071712493896
    },
    {
      "auxiliary_loss_clip": 0.01148948,
      "auxiliary_loss_mlp": 0.01004265,
      "balance_loss_clip": 1.03014708,
      "balance_loss_mlp": 1.00135636,
      "epoch": 0.12481212048337642,
      "flos": 69224641447680.0,
      "grad_norm": 0.8576018445032713,
      "language_loss": 0.62773317,
      "learning_rate": 3.906540388308214e-06,
      "loss": 0.64926535,
      "num_input_tokens_seen": 21958305,
      "step": 1038,
      "time_per_iteration": 3.210991144180298
    },
    {
      "auxiliary_loss_clip": 0.01234283,
      "auxiliary_loss_mlp": 0.01045502,
      "balance_loss_clip": 1.06098866,
      "balance_loss_mlp": 1.03373623,
      "epoch": 0.12493236337401552,
      "flos": 18223696350720.0,
      "grad_norm": 1.74982578601882,
      "language_loss": 0.81684744,
      "learning_rate": 3.906304901508417e-06,
      "loss": 0.83964533,
      "num_input_tokens_seen": 21977205,
      "step": 1039,
      "time_per_iteration": 4.551063299179077
    },
    {
      "auxiliary_loss_clip": 0.01248842,
      "auxiliary_loss_mlp": 0.01043077,
      "balance_loss_clip": 1.07286751,
      "balance_loss_mlp": 1.03269958,
      "epoch": 0.12505260626465461,
      "flos": 30044375303040.0,
      "grad_norm": 3.855802296595132,
      "language_loss": 0.75645006,
      "learning_rate": 3.9060691255236835e-06,
      "loss": 0.77936918,
      "num_input_tokens_seen": 21997770,
      "step": 1040,
      "time_per_iteration": 2.732199192047119
    },
    {
      "auxiliary_loss_clip": 0.0123666,
      "auxiliary_loss_mlp": 0.01040036,
      "balance_loss_clip": 1.06120729,
      "balance_loss_mlp": 1.02804959,
      "epoch": 0.1251728491552937,
      "flos": 24433980347520.0,
      "grad_norm": 1.697262681658903,
      "language_loss": 0.80735993,
      "learning_rate": 3.905833060389778e-06,
      "loss": 0.83012688,
      "num_input_tokens_seen": 22021890,
      "step": 1041,
      "time_per_iteration": 2.7897212505340576
    },
    {
      "auxiliary_loss_clip": 0.01245565,
      "auxiliary_loss_mlp": 0.01209796,
      "balance_loss_clip": 1.07095838,
      "balance_loss_mlp": 1.0007726,
      "epoch": 0.12529309204593278,
      "flos": 27119809952640.0,
      "grad_norm": 2.19116967546814,
      "language_loss": 0.78452271,
      "learning_rate": 3.905596706142513e-06,
      "loss": 0.80907631,
      "num_input_tokens_seen": 22043300,
      "step": 1042,
      "time_per_iteration": 2.682654619216919
    },
    {
      "auxiliary_loss_clip": 0.01241405,
      "auxiliary_loss_mlp": 0.01046257,
      "balance_loss_clip": 1.06023049,
      "balance_loss_mlp": 1.0343833,
      "epoch": 0.12541333493657186,
      "flos": 30774151923840.0,
      "grad_norm": 2.2777874607522715,
      "language_loss": 0.86011779,
      "learning_rate": 3.9053600628177435e-06,
      "loss": 0.88299441,
      "num_input_tokens_seen": 22062910,
      "step": 1043,
      "time_per_iteration": 2.7456629276275635
    },
    {
      "auxiliary_loss_clip": 0.01241708,
      "auxiliary_loss_mlp": 0.01038003,
      "balance_loss_clip": 1.06787348,
      "balance_loss_mlp": 1.02689254,
      "epoch": 0.12553357782721097,
      "flos": 23659566099840.0,
      "grad_norm": 2.2548010352889243,
      "language_loss": 0.85330898,
      "learning_rate": 3.905123130451367e-06,
      "loss": 0.87610608,
      "num_input_tokens_seen": 22084010,
      "step": 1044,
      "time_per_iteration": 2.7017178535461426
    },
    {
      "auxiliary_loss_clip": 0.01244301,
      "auxiliary_loss_mlp": 0.01038204,
      "balance_loss_clip": 1.07003689,
      "balance_loss_mlp": 1.02636039,
      "epoch": 0.12565382071785006,
      "flos": 24863758577280.0,
      "grad_norm": 2.272960764158344,
      "language_loss": 0.79709947,
      "learning_rate": 3.904885909079326e-06,
      "loss": 0.81992453,
      "num_input_tokens_seen": 22102795,
      "step": 1045,
      "time_per_iteration": 2.6295101642608643
    },
    {
      "auxiliary_loss_clip": 0.01245686,
      "auxiliary_loss_mlp": 0.01036497,
      "balance_loss_clip": 1.0661974,
      "balance_loss_mlp": 1.02507114,
      "epoch": 0.12577406360848914,
      "flos": 21360780518400.0,
      "grad_norm": 7.175245295693797,
      "language_loss": 0.78423345,
      "learning_rate": 3.904648398737607e-06,
      "loss": 0.80705535,
      "num_input_tokens_seen": 22121360,
      "step": 1046,
      "time_per_iteration": 2.6731860637664795
    },
    {
      "auxiliary_loss_clip": 0.0124116,
      "auxiliary_loss_mlp": 0.01042522,
      "balance_loss_clip": 1.06705654,
      "balance_loss_mlp": 1.03147149,
      "epoch": 0.12589430649912825,
      "flos": 36138056774400.0,
      "grad_norm": 1.9166587944323037,
      "language_loss": 0.78169763,
      "learning_rate": 3.9044105994622406e-06,
      "loss": 0.80453444,
      "num_input_tokens_seen": 22142505,
      "step": 1047,
      "time_per_iteration": 2.734363317489624
    },
    {
      "auxiliary_loss_clip": 0.01246498,
      "auxiliary_loss_mlp": 0.01210125,
      "balance_loss_clip": 1.06344008,
      "balance_loss_mlp": 1.00088525,
      "epoch": 0.12601454938976733,
      "flos": 25337671643520.0,
      "grad_norm": 1.9525387169457789,
      "language_loss": 0.81921494,
      "learning_rate": 3.9041725112893005e-06,
      "loss": 0.84378111,
      "num_input_tokens_seen": 22163730,
      "step": 1048,
      "time_per_iteration": 2.7312982082366943
    },
    {
      "auxiliary_loss_clip": 0.01240574,
      "auxiliary_loss_mlp": 0.01038424,
      "balance_loss_clip": 1.06222141,
      "balance_loss_mlp": 1.02723646,
      "epoch": 0.12613479228040642,
      "flos": 15560094286080.0,
      "grad_norm": 1.8639227727643861,
      "language_loss": 0.75101042,
      "learning_rate": 3.903934134254904e-06,
      "loss": 0.77380037,
      "num_input_tokens_seen": 22181520,
      "step": 1049,
      "time_per_iteration": 2.67281174659729
    },
    {
      "auxiliary_loss_clip": 0.01247468,
      "auxiliary_loss_mlp": 0.01043856,
      "balance_loss_clip": 1.06525731,
      "balance_loss_mlp": 1.03206015,
      "epoch": 0.1262550351710455,
      "flos": 21470595373440.0,
      "grad_norm": 3.529864120197015,
      "language_loss": 0.85098267,
      "learning_rate": 3.903695468395213e-06,
      "loss": 0.87389588,
      "num_input_tokens_seen": 22199390,
      "step": 1050,
      "time_per_iteration": 2.7132339477539062
    },
    {
      "auxiliary_loss_clip": 0.01246233,
      "auxiliary_loss_mlp": 0.01043175,
      "balance_loss_clip": 1.06026649,
      "balance_loss_mlp": 1.03244042,
      "epoch": 0.1263752780616846,
      "flos": 31576719456000.0,
      "grad_norm": 3.756145154343612,
      "language_loss": 0.56000793,
      "learning_rate": 3.903456513746434e-06,
      "loss": 0.58290201,
      "num_input_tokens_seen": 22220365,
      "step": 1051,
      "time_per_iteration": 2.767103672027588
    },
    {
      "auxiliary_loss_clip": 0.01237441,
      "auxiliary_loss_mlp": 0.01037686,
      "balance_loss_clip": 1.06567085,
      "balance_loss_mlp": 1.02722502,
      "epoch": 0.1264955209523237,
      "flos": 28768217927040.0,
      "grad_norm": 1.9052589654848113,
      "language_loss": 0.87874728,
      "learning_rate": 3.903217270344815e-06,
      "loss": 0.90149856,
      "num_input_tokens_seen": 22240615,
      "step": 1052,
      "time_per_iteration": 2.6925463676452637
    },
    {
      "auxiliary_loss_clip": 0.01236837,
      "auxiliary_loss_mlp": 0.01039309,
      "balance_loss_clip": 1.05914223,
      "balance_loss_mlp": 1.028157,
      "epoch": 0.12661576384296278,
      "flos": 29241125412480.0,
      "grad_norm": 1.9314945893802589,
      "language_loss": 0.82391244,
      "learning_rate": 3.902977738226648e-06,
      "loss": 0.84667397,
      "num_input_tokens_seen": 22261350,
      "step": 1053,
      "time_per_iteration": 2.7987000942230225
    },
    {
      "auxiliary_loss_clip": 0.01245823,
      "auxiliary_loss_mlp": 0.01042198,
      "balance_loss_clip": 1.0666908,
      "balance_loss_mlp": 1.03032458,
      "epoch": 0.12673600673360189,
      "flos": 20850346298880.0,
      "grad_norm": 2.168172957881554,
      "language_loss": 0.90889537,
      "learning_rate": 3.902737917428273e-06,
      "loss": 0.93177563,
      "num_input_tokens_seen": 22279515,
      "step": 1054,
      "time_per_iteration": 2.6364850997924805
    },
    {
      "auxiliary_loss_clip": 0.01240219,
      "auxiliary_loss_mlp": 0.01034047,
      "balance_loss_clip": 1.06643629,
      "balance_loss_mlp": 1.02295482,
      "epoch": 0.12685624962424097,
      "flos": 25263695583360.0,
      "grad_norm": 1.9382281475215086,
      "language_loss": 0.84032071,
      "learning_rate": 3.902497807986068e-06,
      "loss": 0.8630634,
      "num_input_tokens_seen": 22299535,
      "step": 1055,
      "time_per_iteration": 2.64939546585083
    },
    {
      "auxiliary_loss_clip": 0.01245555,
      "auxiliary_loss_mlp": 0.01039172,
      "balance_loss_clip": 1.05934012,
      "balance_loss_mlp": 1.02724481,
      "epoch": 0.12697649251488005,
      "flos": 27527109246720.0,
      "grad_norm": 3.4510957253621934,
      "language_loss": 0.83919966,
      "learning_rate": 3.902257409936458e-06,
      "loss": 0.86204696,
      "num_input_tokens_seen": 22320300,
      "step": 1056,
      "time_per_iteration": 2.718945264816284
    },
    {
      "auxiliary_loss_clip": 0.01239723,
      "auxiliary_loss_mlp": 0.01041663,
      "balance_loss_clip": 1.06446004,
      "balance_loss_mlp": 1.03085113,
      "epoch": 0.12709673540551916,
      "flos": 21251863503360.0,
      "grad_norm": 2.1514501794515897,
      "language_loss": 0.84293175,
      "learning_rate": 3.902016723315912e-06,
      "loss": 0.86574554,
      "num_input_tokens_seen": 22338240,
      "step": 1057,
      "time_per_iteration": 2.677978992462158
    },
    {
      "auxiliary_loss_clip": 0.01238458,
      "auxiliary_loss_mlp": 0.01041512,
      "balance_loss_clip": 1.06240153,
      "balance_loss_mlp": 1.03047335,
      "epoch": 0.12721697829615825,
      "flos": 25337707557120.0,
      "grad_norm": 2.837982700437704,
      "language_loss": 0.69997889,
      "learning_rate": 3.901775748160941e-06,
      "loss": 0.72277856,
      "num_input_tokens_seen": 22357420,
      "step": 1058,
      "time_per_iteration": 2.658637762069702
    },
    {
      "auxiliary_loss_clip": 0.01148993,
      "auxiliary_loss_mlp": 0.01002942,
      "balance_loss_clip": 1.03963506,
      "balance_loss_mlp": 1.00016439,
      "epoch": 0.12733722118679733,
      "flos": 61943287754880.0,
      "grad_norm": 0.7951293559129718,
      "language_loss": 0.60817671,
      "learning_rate": 3.901534484508101e-06,
      "loss": 0.62969601,
      "num_input_tokens_seen": 22420095,
      "step": 1059,
      "time_per_iteration": 3.2138593196868896
    },
    {
      "auxiliary_loss_clip": 0.0123326,
      "auxiliary_loss_mlp": 0.01034215,
      "balance_loss_clip": 1.05932081,
      "balance_loss_mlp": 1.02294421,
      "epoch": 0.1274574640774364,
      "flos": 26976742081920.0,
      "grad_norm": 3.5031974267419272,
      "language_loss": 0.74826342,
      "learning_rate": 3.901292932393991e-06,
      "loss": 0.77093816,
      "num_input_tokens_seen": 22438975,
      "step": 1060,
      "time_per_iteration": 2.6961851119995117
    },
    {
      "auxiliary_loss_clip": 0.01240142,
      "auxiliary_loss_mlp": 0.01040856,
      "balance_loss_clip": 1.06803381,
      "balance_loss_mlp": 1.02928066,
      "epoch": 0.12757770696807552,
      "flos": 22236318529920.0,
      "grad_norm": 3.474755433103945,
      "language_loss": 0.85367304,
      "learning_rate": 3.9010510918552555e-06,
      "loss": 0.87648302,
      "num_input_tokens_seen": 22458050,
      "step": 1061,
      "time_per_iteration": 2.5847859382629395
    },
    {
      "auxiliary_loss_clip": 0.01239331,
      "auxiliary_loss_mlp": 0.01046207,
      "balance_loss_clip": 1.06107092,
      "balance_loss_mlp": 1.03353453,
      "epoch": 0.1276979498587146,
      "flos": 28547905858560.0,
      "grad_norm": 4.8773182330573945,
      "language_loss": 0.7481764,
      "learning_rate": 3.900808962928581e-06,
      "loss": 0.77103174,
      "num_input_tokens_seen": 22475665,
      "step": 1062,
      "time_per_iteration": 3.662301540374756
    },
    {
      "auxiliary_loss_clip": 0.01241854,
      "auxiliary_loss_mlp": 0.01037464,
      "balance_loss_clip": 1.07026064,
      "balance_loss_mlp": 1.02591848,
      "epoch": 0.1278181927493537,
      "flos": 17420338719360.0,
      "grad_norm": 2.214405600038654,
      "language_loss": 0.89542794,
      "learning_rate": 3.900566545650698e-06,
      "loss": 0.91822112,
      "num_input_tokens_seen": 22493335,
      "step": 1063,
      "time_per_iteration": 3.554638624191284
    },
    {
      "auxiliary_loss_clip": 0.0124117,
      "auxiliary_loss_mlp": 0.0103481,
      "balance_loss_clip": 1.06641746,
      "balance_loss_mlp": 1.02270985,
      "epoch": 0.1279384356399928,
      "flos": 21138636856320.0,
      "grad_norm": 2.336627305158109,
      "language_loss": 0.81885082,
      "learning_rate": 3.900323840058381e-06,
      "loss": 0.84161055,
      "num_input_tokens_seen": 22511045,
      "step": 1064,
      "time_per_iteration": 2.655080795288086
    },
    {
      "auxiliary_loss_clip": 0.01239959,
      "auxiliary_loss_mlp": 0.01035754,
      "balance_loss_clip": 1.06248391,
      "balance_loss_mlp": 1.02567482,
      "epoch": 0.12805867853063188,
      "flos": 26576733248640.0,
      "grad_norm": 1.9130117762963423,
      "language_loss": 0.81558174,
      "learning_rate": 3.900080846188449e-06,
      "loss": 0.83833885,
      "num_input_tokens_seen": 22529635,
      "step": 1065,
      "time_per_iteration": 3.5163943767547607
    },
    {
      "auxiliary_loss_clip": 0.01238408,
      "auxiliary_loss_mlp": 0.01032885,
      "balance_loss_clip": 1.06612802,
      "balance_loss_mlp": 1.02145898,
      "epoch": 0.12817892142127096,
      "flos": 16436206915200.0,
      "grad_norm": 1.9121397756936107,
      "language_loss": 0.81472528,
      "learning_rate": 3.8998375640777625e-06,
      "loss": 0.83743823,
      "num_input_tokens_seen": 22547505,
      "step": 1066,
      "time_per_iteration": 3.491523265838623
    },
    {
      "auxiliary_loss_clip": 0.01142685,
      "auxiliary_loss_mlp": 0.01018865,
      "balance_loss_clip": 1.03073239,
      "balance_loss_mlp": 1.01562262,
      "epoch": 0.12829916431191005,
      "flos": 60757049099520.0,
      "grad_norm": 0.7066611832958803,
      "language_loss": 0.52614737,
      "learning_rate": 3.899593993763229e-06,
      "loss": 0.54776287,
      "num_input_tokens_seen": 22608465,
      "step": 1067,
      "time_per_iteration": 3.139594316482544
    },
    {
      "auxiliary_loss_clip": 0.01236904,
      "auxiliary_loss_mlp": 0.01043474,
      "balance_loss_clip": 1.06055629,
      "balance_loss_mlp": 1.03075409,
      "epoch": 0.12841940720254916,
      "flos": 29786895636480.0,
      "grad_norm": 3.8513256461272043,
      "language_loss": 0.81223297,
      "learning_rate": 3.899350135281796e-06,
      "loss": 0.83503675,
      "num_input_tokens_seen": 22629465,
      "step": 1068,
      "time_per_iteration": 2.7608797550201416
    },
    {
      "auxiliary_loss_clip": 0.01243257,
      "auxiliary_loss_mlp": 0.01036498,
      "balance_loss_clip": 1.06146443,
      "balance_loss_mlp": 1.02599013,
      "epoch": 0.12853965009318824,
      "flos": 25951851319680.0,
      "grad_norm": 2.5444405611380367,
      "language_loss": 0.79341078,
      "learning_rate": 3.8991059886704585e-06,
      "loss": 0.81620836,
      "num_input_tokens_seen": 22648970,
      "step": 1069,
      "time_per_iteration": 2.722672462463379
    },
    {
      "auxiliary_loss_clip": 0.01237281,
      "auxiliary_loss_mlp": 0.01041511,
      "balance_loss_clip": 1.0596832,
      "balance_loss_mlp": 1.03059769,
      "epoch": 0.12865989298382732,
      "flos": 30846871008000.0,
      "grad_norm": 2.090175831579378,
      "language_loss": 0.83058,
      "learning_rate": 3.898861553966252e-06,
      "loss": 0.85336792,
      "num_input_tokens_seen": 22668620,
      "step": 1070,
      "time_per_iteration": 2.848907470703125
    },
    {
      "auxiliary_loss_clip": 0.01248008,
      "auxiliary_loss_mlp": 0.01039672,
      "balance_loss_clip": 1.05250299,
      "balance_loss_mlp": 1.0286932,
      "epoch": 0.12878013587446643,
      "flos": 25885776251520.0,
      "grad_norm": 1.7918388974866144,
      "language_loss": 0.88112277,
      "learning_rate": 3.898616831206257e-06,
      "loss": 0.90399963,
      "num_input_tokens_seen": 22689045,
      "step": 1071,
      "time_per_iteration": 2.7908616065979004
    },
    {
      "auxiliary_loss_clip": 0.01237702,
      "auxiliary_loss_mlp": 0.01038411,
      "balance_loss_clip": 1.05646563,
      "balance_loss_mlp": 1.02559638,
      "epoch": 0.12890037876510552,
      "flos": 23333138277120.0,
      "grad_norm": 1.9748019380664121,
      "language_loss": 0.77154106,
      "learning_rate": 3.8983718204276e-06,
      "loss": 0.79430217,
      "num_input_tokens_seen": 22711265,
      "step": 1072,
      "time_per_iteration": 2.7961483001708984
    },
    {
      "auxiliary_loss_clip": 0.0123621,
      "auxiliary_loss_mlp": 0.0103926,
      "balance_loss_clip": 1.05957389,
      "balance_loss_mlp": 1.02869236,
      "epoch": 0.1290206216557446,
      "flos": 23587242065280.0,
      "grad_norm": 1.7453889544665693,
      "language_loss": 0.82422465,
      "learning_rate": 3.898126521667446e-06,
      "loss": 0.84697938,
      "num_input_tokens_seen": 22731420,
      "step": 1073,
      "time_per_iteration": 2.7051949501037598
    },
    {
      "auxiliary_loss_clip": 0.01236434,
      "auxiliary_loss_mlp": 0.01045998,
      "balance_loss_clip": 1.06120694,
      "balance_loss_mlp": 1.03441668,
      "epoch": 0.12914086454638368,
      "flos": 24170610850560.0,
      "grad_norm": 1.7279698166285675,
      "language_loss": 0.83494574,
      "learning_rate": 3.897880934963007e-06,
      "loss": 0.85777003,
      "num_input_tokens_seen": 22750970,
      "step": 1074,
      "time_per_iteration": 2.7139389514923096
    },
    {
      "auxiliary_loss_clip": 0.01233837,
      "auxiliary_loss_mlp": 0.01034142,
      "balance_loss_clip": 1.0569706,
      "balance_loss_mlp": 1.02260911,
      "epoch": 0.1292611074370228,
      "flos": 20267157081600.0,
      "grad_norm": 3.864138113261689,
      "language_loss": 0.78099209,
      "learning_rate": 3.89763506035154e-06,
      "loss": 0.80367184,
      "num_input_tokens_seen": 22768820,
      "step": 1075,
      "time_per_iteration": 2.666330575942993
    },
    {
      "auxiliary_loss_clip": 0.01225516,
      "auxiliary_loss_mlp": 0.01034523,
      "balance_loss_clip": 1.06017303,
      "balance_loss_mlp": 1.02379394,
      "epoch": 0.12938135032766188,
      "flos": 27377684668800.0,
      "grad_norm": 2.025047878615437,
      "language_loss": 0.81179774,
      "learning_rate": 3.897388897870343e-06,
      "loss": 0.83439815,
      "num_input_tokens_seen": 22789460,
      "step": 1076,
      "time_per_iteration": 2.7078113555908203
    },
    {
      "auxiliary_loss_clip": 0.01252749,
      "auxiliary_loss_mlp": 0.01035541,
      "balance_loss_clip": 1.06182218,
      "balance_loss_mlp": 1.02356601,
      "epoch": 0.12950159321830096,
      "flos": 29277107861760.0,
      "grad_norm": 2.8560386397251913,
      "language_loss": 0.74891901,
      "learning_rate": 3.89714244755676e-06,
      "loss": 0.77180183,
      "num_input_tokens_seen": 22810820,
      "step": 1077,
      "time_per_iteration": 2.7360336780548096
    },
    {
      "auxiliary_loss_clip": 0.01229053,
      "auxiliary_loss_mlp": 0.0104008,
      "balance_loss_clip": 1.05337119,
      "balance_loss_mlp": 1.02907085,
      "epoch": 0.12962183610894007,
      "flos": 24534888629760.0,
      "grad_norm": 2.3694669863886313,
      "language_loss": 0.86326605,
      "learning_rate": 3.896895709448175e-06,
      "loss": 0.88595736,
      "num_input_tokens_seen": 22830570,
      "step": 1078,
      "time_per_iteration": 2.7434301376342773
    },
    {
      "auxiliary_loss_clip": 0.01238475,
      "auxiliary_loss_mlp": 0.01041295,
      "balance_loss_clip": 1.05212855,
      "balance_loss_mlp": 1.03045309,
      "epoch": 0.12974207899957915,
      "flos": 11215944552960.0,
      "grad_norm": 3.5373347314907377,
      "language_loss": 0.77888751,
      "learning_rate": 3.896648683582019e-06,
      "loss": 0.80168521,
      "num_input_tokens_seen": 22845905,
      "step": 1079,
      "time_per_iteration": 2.7139034271240234
    },
    {
      "auxiliary_loss_clip": 0.01239477,
      "auxiliary_loss_mlp": 0.01037958,
      "balance_loss_clip": 1.0599463,
      "balance_loss_mlp": 1.02744365,
      "epoch": 0.12986232189021824,
      "flos": 24717889445760.0,
      "grad_norm": 14.31202838898249,
      "language_loss": 0.81083071,
      "learning_rate": 3.896401369995766e-06,
      "loss": 0.83360505,
      "num_input_tokens_seen": 22865710,
      "step": 1080,
      "time_per_iteration": 2.723466157913208
    },
    {
      "auxiliary_loss_clip": 0.01239886,
      "auxiliary_loss_mlp": 0.01044937,
      "balance_loss_clip": 1.06814623,
      "balance_loss_mlp": 1.03424394,
      "epoch": 0.12998256478085732,
      "flos": 23915357827200.0,
      "grad_norm": 1.9010835704322808,
      "language_loss": 0.79509175,
      "learning_rate": 3.896153768726932e-06,
      "loss": 0.81794,
      "num_input_tokens_seen": 22886020,
      "step": 1081,
      "time_per_iteration": 2.640894651412964
    },
    {
      "auxiliary_loss_clip": 0.01239529,
      "auxiliary_loss_mlp": 0.01038716,
      "balance_loss_clip": 1.0650481,
      "balance_loss_mlp": 1.02756989,
      "epoch": 0.13010280767149643,
      "flos": 18624207974400.0,
      "grad_norm": 2.6429867579437123,
      "language_loss": 0.87988561,
      "learning_rate": 3.8959058798130806e-06,
      "loss": 0.90266806,
      "num_input_tokens_seen": 22903995,
      "step": 1082,
      "time_per_iteration": 2.617307662963867
    },
    {
      "auxiliary_loss_clip": 0.0124313,
      "auxiliary_loss_mlp": 0.01210006,
      "balance_loss_clip": 1.0627377,
      "balance_loss_mlp": 1.00102425,
      "epoch": 0.1302230505621355,
      "flos": 22783992174720.0,
      "grad_norm": 7.017184032408312,
      "language_loss": 0.75084525,
      "learning_rate": 3.895657703291814e-06,
      "loss": 0.77537656,
      "num_input_tokens_seen": 22924100,
      "step": 1083,
      "time_per_iteration": 2.7093758583068848
    },
    {
      "auxiliary_loss_clip": 0.01247593,
      "auxiliary_loss_mlp": 0.01033771,
      "balance_loss_clip": 1.06015289,
      "balance_loss_mlp": 1.02259552,
      "epoch": 0.1303432934527746,
      "flos": 21323612920320.0,
      "grad_norm": 19.302739428883477,
      "language_loss": 0.79910821,
      "learning_rate": 3.895409239200781e-06,
      "loss": 0.82192183,
      "num_input_tokens_seen": 22939985,
      "step": 1084,
      "time_per_iteration": 2.6600492000579834
    },
    {
      "auxiliary_loss_clip": 0.01232342,
      "auxiliary_loss_mlp": 0.0103781,
      "balance_loss_clip": 1.06061494,
      "balance_loss_mlp": 1.02562046,
      "epoch": 0.1304635363434137,
      "flos": 20922490765440.0,
      "grad_norm": 4.356620508138738,
      "language_loss": 0.91200262,
      "learning_rate": 3.895160487577673e-06,
      "loss": 0.93470407,
      "num_input_tokens_seen": 22957555,
      "step": 1085,
      "time_per_iteration": 2.6820228099823
    },
    {
      "auxiliary_loss_clip": 0.01137122,
      "auxiliary_loss_mlp": 0.01004759,
      "balance_loss_clip": 1.029598,
      "balance_loss_mlp": 1.001755,
      "epoch": 0.1305837792340528,
      "flos": 63245659080960.0,
      "grad_norm": 0.7895553309030635,
      "language_loss": 0.60904491,
      "learning_rate": 3.894911448460226e-06,
      "loss": 0.63046372,
      "num_input_tokens_seen": 23016870,
      "step": 1086,
      "time_per_iteration": 3.0426855087280273
    },
    {
      "auxiliary_loss_clip": 0.01224873,
      "auxiliary_loss_mlp": 0.01043252,
      "balance_loss_clip": 1.04985118,
      "balance_loss_mlp": 1.03157568,
      "epoch": 0.13070402212469187,
      "flos": 26428852955520.0,
      "grad_norm": 2.3916151102793775,
      "language_loss": 0.73446536,
      "learning_rate": 3.8946621218862195e-06,
      "loss": 0.7571466,
      "num_input_tokens_seen": 23037870,
      "step": 1087,
      "time_per_iteration": 2.8209068775177
    },
    {
      "auxiliary_loss_clip": 0.01238765,
      "auxiliary_loss_mlp": 0.01045333,
      "balance_loss_clip": 1.05979729,
      "balance_loss_mlp": 1.03505111,
      "epoch": 0.13082426501533098,
      "flos": 27673409341440.0,
      "grad_norm": 2.464705198459352,
      "language_loss": 0.887371,
      "learning_rate": 3.894412507893475e-06,
      "loss": 0.91021198,
      "num_input_tokens_seen": 23058150,
      "step": 1088,
      "time_per_iteration": 2.735931873321533
    },
    {
      "auxiliary_loss_clip": 0.01247467,
      "auxiliary_loss_mlp": 0.01040574,
      "balance_loss_clip": 1.05704737,
      "balance_loss_mlp": 1.02908218,
      "epoch": 0.13094450790597006,
      "flos": 24826770547200.0,
      "grad_norm": 2.1834143511850583,
      "language_loss": 0.72007525,
      "learning_rate": 3.894162606519859e-06,
      "loss": 0.74295568,
      "num_input_tokens_seen": 23077100,
      "step": 1089,
      "time_per_iteration": 3.6547605991363525
    },
    {
      "auxiliary_loss_clip": 0.01239098,
      "auxiliary_loss_mlp": 0.01039191,
      "balance_loss_clip": 1.05742574,
      "balance_loss_mlp": 1.02885556,
      "epoch": 0.13106475079660915,
      "flos": 19062605468160.0,
      "grad_norm": 2.0280851213243527,
      "language_loss": 0.77129626,
      "learning_rate": 3.893912417803282e-06,
      "loss": 0.79407918,
      "num_input_tokens_seen": 23096815,
      "step": 1090,
      "time_per_iteration": 2.7010321617126465
    },
    {
      "auxiliary_loss_clip": 0.01242694,
      "auxiliary_loss_mlp": 0.01036803,
      "balance_loss_clip": 1.05354762,
      "balance_loss_mlp": 1.02497125,
      "epoch": 0.13118499368724823,
      "flos": 28913189218560.0,
      "grad_norm": 1.8420203540791948,
      "language_loss": 0.77035081,
      "learning_rate": 3.8936619417816975e-06,
      "loss": 0.79314584,
      "num_input_tokens_seen": 23117145,
      "step": 1091,
      "time_per_iteration": 3.6641807556152344
    },
    {
      "auxiliary_loss_clip": 0.01240211,
      "auxiliary_loss_mlp": 0.01031932,
      "balance_loss_clip": 1.06183839,
      "balance_loss_mlp": 1.02143013,
      "epoch": 0.13130523657788734,
      "flos": 14283398206080.0,
      "grad_norm": 2.872688019641648,
      "language_loss": 0.71411431,
      "learning_rate": 3.8934111784931015e-06,
      "loss": 0.73683572,
      "num_input_tokens_seen": 23134595,
      "step": 1092,
      "time_per_iteration": 3.579416036605835
    },
    {
      "auxiliary_loss_clip": 0.01147567,
      "auxiliary_loss_mlp": 0.01003795,
      "balance_loss_clip": 1.03138816,
      "balance_loss_mlp": 1.00077951,
      "epoch": 0.13142547946852642,
      "flos": 70174155519360.0,
      "grad_norm": 0.9137577638077861,
      "language_loss": 0.59094375,
      "learning_rate": 3.893160127975535e-06,
      "loss": 0.61245739,
      "num_input_tokens_seen": 23195285,
      "step": 1093,
      "time_per_iteration": 3.2925894260406494
    },
    {
      "auxiliary_loss_clip": 0.01244519,
      "auxiliary_loss_mlp": 0.01033062,
      "balance_loss_clip": 1.0548296,
      "balance_loss_mlp": 1.02242851,
      "epoch": 0.1315457223591655,
      "flos": 45805998844800.0,
      "grad_norm": 2.327089808731587,
      "language_loss": 0.81509966,
      "learning_rate": 3.8929087902670826e-06,
      "loss": 0.83787549,
      "num_input_tokens_seen": 23216915,
      "step": 1094,
      "time_per_iteration": 2.911893606185913
    },
    {
      "auxiliary_loss_clip": 0.01140775,
      "auxiliary_loss_mlp": 0.01005619,
      "balance_loss_clip": 1.02897072,
      "balance_loss_mlp": 1.00261486,
      "epoch": 0.13166596524980462,
      "flos": 62881165820160.0,
      "grad_norm": 0.9378813717527529,
      "language_loss": 0.60697985,
      "learning_rate": 3.8926571654058715e-06,
      "loss": 0.62844384,
      "num_input_tokens_seen": 23273560,
      "step": 1095,
      "time_per_iteration": 3.0796358585357666
    },
    {
      "auxiliary_loss_clip": 0.01235385,
      "auxiliary_loss_mlp": 0.01034734,
      "balance_loss_clip": 1.05851698,
      "balance_loss_mlp": 1.02410603,
      "epoch": 0.1317862081404437,
      "flos": 23586523793280.0,
      "grad_norm": 3.6009245629939493,
      "language_loss": 0.77161729,
      "learning_rate": 3.892405253430074e-06,
      "loss": 0.79431844,
      "num_input_tokens_seen": 23291080,
      "step": 1096,
      "time_per_iteration": 2.6974287033081055
    },
    {
      "auxiliary_loss_clip": 0.01241188,
      "auxiliary_loss_mlp": 0.01210058,
      "balance_loss_clip": 1.06211686,
      "balance_loss_mlp": 1.0011785,
      "epoch": 0.13190645103108278,
      "flos": 20260764460800.0,
      "grad_norm": 2.5034860411897015,
      "language_loss": 0.82613319,
      "learning_rate": 3.892153054377904e-06,
      "loss": 0.85064566,
      "num_input_tokens_seen": 23308485,
      "step": 1097,
      "time_per_iteration": 2.630796194076538
    },
    {
      "auxiliary_loss_clip": 0.01152695,
      "auxiliary_loss_mlp": 0.01001511,
      "balance_loss_clip": 1.03177226,
      "balance_loss_mlp": 0.99891204,
      "epoch": 0.13202669392172187,
      "flos": 53455440136320.0,
      "grad_norm": 0.9370266439309044,
      "language_loss": 0.59460956,
      "learning_rate": 3.891900568287619e-06,
      "loss": 0.61615163,
      "num_input_tokens_seen": 23360870,
      "step": 1098,
      "time_per_iteration": 3.083665370941162
    },
    {
      "auxiliary_loss_clip": 0.01242044,
      "auxiliary_loss_mlp": 0.01034675,
      "balance_loss_clip": 1.05887091,
      "balance_loss_mlp": 1.02306974,
      "epoch": 0.13214693681236098,
      "flos": 15851293845120.0,
      "grad_norm": 2.785953929103142,
      "language_loss": 0.72661012,
      "learning_rate": 3.891647795197523e-06,
      "loss": 0.74937737,
      "num_input_tokens_seen": 23376910,
      "step": 1099,
      "time_per_iteration": 2.6590769290924072
    },
    {
      "auxiliary_loss_clip": 0.01245408,
      "auxiliary_loss_mlp": 0.01041165,
      "balance_loss_clip": 1.05678427,
      "balance_loss_mlp": 1.02911878,
      "epoch": 0.13226717970300006,
      "flos": 19353840940800.0,
      "grad_norm": 2.7460440551448957,
      "language_loss": 0.68805528,
      "learning_rate": 3.8913947351459605e-06,
      "loss": 0.71092105,
      "num_input_tokens_seen": 23394450,
      "step": 1100,
      "time_per_iteration": 2.674081802368164
    },
    {
      "auxiliary_loss_clip": 0.01237651,
      "auxiliary_loss_mlp": 0.0103444,
      "balance_loss_clip": 1.06578851,
      "balance_loss_mlp": 1.0239377,
      "epoch": 0.13238742259363914,
      "flos": 20698084546560.0,
      "grad_norm": 2.582845116093576,
      "language_loss": 0.67444074,
      "learning_rate": 3.89114138817132e-06,
      "loss": 0.69716167,
      "num_input_tokens_seen": 23411115,
      "step": 1101,
      "time_per_iteration": 2.580817699432373
    },
    {
      "auxiliary_loss_clip": 0.01237894,
      "auxiliary_loss_mlp": 0.01033641,
      "balance_loss_clip": 1.06461918,
      "balance_loss_mlp": 1.02294779,
      "epoch": 0.13250766548427825,
      "flos": 21032449274880.0,
      "grad_norm": 2.079262308160518,
      "language_loss": 0.84138066,
      "learning_rate": 3.890887754312035e-06,
      "loss": 0.86409599,
      "num_input_tokens_seen": 23429360,
      "step": 1102,
      "time_per_iteration": 2.6210081577301025
    },
    {
      "auxiliary_loss_clip": 0.01229385,
      "auxiliary_loss_mlp": 0.0104293,
      "balance_loss_clip": 1.0553081,
      "balance_loss_mlp": 1.03186107,
      "epoch": 0.13262790837491734,
      "flos": 22637871648000.0,
      "grad_norm": 1.9334609052403813,
      "language_loss": 0.87460417,
      "learning_rate": 3.890633833606581e-06,
      "loss": 0.8973273,
      "num_input_tokens_seen": 23449050,
      "step": 1103,
      "time_per_iteration": 2.6344077587127686
    },
    {
      "auxiliary_loss_clip": 0.01236103,
      "auxiliary_loss_mlp": 0.01035132,
      "balance_loss_clip": 1.06531405,
      "balance_loss_mlp": 1.0249697,
      "epoch": 0.13274815126555642,
      "flos": 19683141851520.0,
      "grad_norm": 2.145087768504033,
      "language_loss": 0.69406509,
      "learning_rate": 3.890379626093477e-06,
      "loss": 0.71677744,
      "num_input_tokens_seen": 23468800,
      "step": 1104,
      "time_per_iteration": 2.5877158641815186
    },
    {
      "auxiliary_loss_clip": 0.01226435,
      "auxiliary_loss_mlp": 0.0103482,
      "balance_loss_clip": 1.05363369,
      "balance_loss_mlp": 1.02376366,
      "epoch": 0.1328683941561955,
      "flos": 21317687176320.0,
      "grad_norm": 2.013641189189029,
      "language_loss": 0.92576361,
      "learning_rate": 3.890125131811287e-06,
      "loss": 0.94837624,
      "num_input_tokens_seen": 23486850,
      "step": 1105,
      "time_per_iteration": 2.7955689430236816
    },
    {
      "auxiliary_loss_clip": 0.01224801,
      "auxiliary_loss_mlp": 0.01034838,
      "balance_loss_clip": 1.05404902,
      "balance_loss_mlp": 1.02495599,
      "epoch": 0.1329886370468346,
      "flos": 13699131580800.0,
      "grad_norm": 2.0451703731646593,
      "language_loss": 0.75556815,
      "learning_rate": 3.889870350798618e-06,
      "loss": 0.77816457,
      "num_input_tokens_seen": 23504195,
      "step": 1106,
      "time_per_iteration": 2.6173951625823975
    },
    {
      "auxiliary_loss_clip": 0.01235925,
      "auxiliary_loss_mlp": 0.01033179,
      "balance_loss_clip": 1.06334066,
      "balance_loss_mlp": 1.02294517,
      "epoch": 0.1331088799374737,
      "flos": 21032413361280.0,
      "grad_norm": 1.8569089584422906,
      "language_loss": 0.78564471,
      "learning_rate": 3.889615283094119e-06,
      "loss": 0.80833578,
      "num_input_tokens_seen": 23523385,
      "step": 1107,
      "time_per_iteration": 2.6543967723846436
    },
    {
      "auxiliary_loss_clip": 0.01241085,
      "auxiliary_loss_mlp": 0.01037978,
      "balance_loss_clip": 1.06471562,
      "balance_loss_mlp": 1.0267067,
      "epoch": 0.13322912282811278,
      "flos": 18260432985600.0,
      "grad_norm": 2.602937316295164,
      "language_loss": 0.84784275,
      "learning_rate": 3.889359928736485e-06,
      "loss": 0.87063336,
      "num_input_tokens_seen": 23541330,
      "step": 1108,
      "time_per_iteration": 2.5827887058258057
    },
    {
      "auxiliary_loss_clip": 0.01231491,
      "auxiliary_loss_mlp": 0.01209888,
      "balance_loss_clip": 1.06052864,
      "balance_loss_mlp": 1.00112557,
      "epoch": 0.1333493657187519,
      "flos": 24460876656000.0,
      "grad_norm": 2.084726015430629,
      "language_loss": 0.91059995,
      "learning_rate": 3.889104287764451e-06,
      "loss": 0.93501377,
      "num_input_tokens_seen": 23561705,
      "step": 1109,
      "time_per_iteration": 2.6741549968719482
    },
    {
      "auxiliary_loss_clip": 0.01236853,
      "auxiliary_loss_mlp": 0.01037025,
      "balance_loss_clip": 1.06133652,
      "balance_loss_mlp": 1.02643883,
      "epoch": 0.13346960860939097,
      "flos": 22158930677760.0,
      "grad_norm": 2.070591686043817,
      "language_loss": 0.90622008,
      "learning_rate": 3.888848360216798e-06,
      "loss": 0.92895889,
      "num_input_tokens_seen": 23579350,
      "step": 1110,
      "time_per_iteration": 2.6142594814300537
    },
    {
      "auxiliary_loss_clip": 0.01145849,
      "auxiliary_loss_mlp": 0.01001774,
      "balance_loss_clip": 1.02825129,
      "balance_loss_mlp": 0.99886495,
      "epoch": 0.13358985150003005,
      "flos": 67931212608000.0,
      "grad_norm": 0.801547617154548,
      "language_loss": 0.5653187,
      "learning_rate": 3.888592146132351e-06,
      "loss": 0.58679497,
      "num_input_tokens_seen": 23640620,
      "step": 1111,
      "time_per_iteration": 3.3018360137939453
    },
    {
      "auxiliary_loss_clip": 0.01236659,
      "auxiliary_loss_mlp": 0.010433,
      "balance_loss_clip": 1.06389427,
      "balance_loss_mlp": 1.03279805,
      "epoch": 0.13371009439066917,
      "flos": 26834284742400.0,
      "grad_norm": 1.91422304618114,
      "language_loss": 0.78156823,
      "learning_rate": 3.888335645549978e-06,
      "loss": 0.80436778,
      "num_input_tokens_seen": 23661040,
      "step": 1112,
      "time_per_iteration": 2.6532554626464844
    },
    {
      "auxiliary_loss_clip": 0.01237758,
      "auxiliary_loss_mlp": 0.01042665,
      "balance_loss_clip": 1.06666613,
      "balance_loss_mlp": 1.03198957,
      "epoch": 0.13383033728130825,
      "flos": 26322844942080.0,
      "grad_norm": 2.8545309624790582,
      "language_loss": 0.81474692,
      "learning_rate": 3.888078858508588e-06,
      "loss": 0.83755112,
      "num_input_tokens_seen": 23680900,
      "step": 1113,
      "time_per_iteration": 2.617018938064575
    },
    {
      "auxiliary_loss_clip": 0.01235043,
      "auxiliary_loss_mlp": 0.01033533,
      "balance_loss_clip": 1.06214476,
      "balance_loss_mlp": 1.02262509,
      "epoch": 0.13395058017194733,
      "flos": 22563931501440.0,
      "grad_norm": 2.2487964312546596,
      "language_loss": 0.84345949,
      "learning_rate": 3.8878217850471365e-06,
      "loss": 0.86614525,
      "num_input_tokens_seen": 23700815,
      "step": 1114,
      "time_per_iteration": 2.6377267837524414
    },
    {
      "auxiliary_loss_clip": 0.01238961,
      "auxiliary_loss_mlp": 0.01043178,
      "balance_loss_clip": 1.06568527,
      "balance_loss_mlp": 1.03138196,
      "epoch": 0.13407082306258641,
      "flos": 25810938264960.0,
      "grad_norm": 2.283529874199181,
      "language_loss": 0.73990703,
      "learning_rate": 3.887564425204621e-06,
      "loss": 0.76272845,
      "num_input_tokens_seen": 23722500,
      "step": 1115,
      "time_per_iteration": 3.5799484252929688
    },
    {
      "auxiliary_loss_clip": 0.01139409,
      "auxiliary_loss_mlp": 0.01004486,
      "balance_loss_clip": 1.02786779,
      "balance_loss_mlp": 1.00175595,
      "epoch": 0.13419106595322552,
      "flos": 68338365269760.0,
      "grad_norm": 0.8449482032514781,
      "language_loss": 0.54642415,
      "learning_rate": 3.887306779020083e-06,
      "loss": 0.56786311,
      "num_input_tokens_seen": 23777155,
      "step": 1116,
      "time_per_iteration": 4.018868923187256
    },
    {
      "auxiliary_loss_clip": 0.01241145,
      "auxiliary_loss_mlp": 0.01037405,
      "balance_loss_clip": 1.06472993,
      "balance_loss_mlp": 1.02606809,
      "epoch": 0.1343113088438646,
      "flos": 20449080489600.0,
      "grad_norm": 13.747155399712954,
      "language_loss": 0.70266306,
      "learning_rate": 3.887048846532608e-06,
      "loss": 0.72544861,
      "num_input_tokens_seen": 23794130,
      "step": 1117,
      "time_per_iteration": 2.6104393005371094
    },
    {
      "auxiliary_loss_clip": 0.01146472,
      "auxiliary_loss_mlp": 0.01006419,
      "balance_loss_clip": 1.02799857,
      "balance_loss_mlp": 1.00376081,
      "epoch": 0.1344315517345037,
      "flos": 67389784951680.0,
      "grad_norm": 0.7629902409717455,
      "language_loss": 0.5812937,
      "learning_rate": 3.8867906277813224e-06,
      "loss": 0.60282254,
      "num_input_tokens_seen": 23852285,
      "step": 1118,
      "time_per_iteration": 4.037387371063232
    },
    {
      "auxiliary_loss_clip": 0.01236823,
      "auxiliary_loss_mlp": 0.01209296,
      "balance_loss_clip": 1.06003237,
      "balance_loss_mlp": 1.0010066,
      "epoch": 0.1345517946251428,
      "flos": 40734442788480.0,
      "grad_norm": 2.497311474127778,
      "language_loss": 0.73907769,
      "learning_rate": 3.886532122805399e-06,
      "loss": 0.76353896,
      "num_input_tokens_seen": 23874765,
      "step": 1119,
      "time_per_iteration": 3.716259717941284
    },
    {
      "auxiliary_loss_clip": 0.01228238,
      "auxiliary_loss_mlp": 0.01036648,
      "balance_loss_clip": 1.05020857,
      "balance_loss_mlp": 1.02560973,
      "epoch": 0.13467203751578188,
      "flos": 22816850140800.0,
      "grad_norm": 2.0521267412088076,
      "language_loss": 0.89704716,
      "learning_rate": 3.886273331644053e-06,
      "loss": 0.91969603,
      "num_input_tokens_seen": 23893635,
      "step": 1120,
      "time_per_iteration": 2.730241060256958
    },
    {
      "auxiliary_loss_clip": 0.0123579,
      "auxiliary_loss_mlp": 0.01030698,
      "balance_loss_clip": 1.05635166,
      "balance_loss_mlp": 1.02015448,
      "epoch": 0.13479228040642097,
      "flos": 17091576512640.0,
      "grad_norm": 3.3290161219834724,
      "language_loss": 0.82637453,
      "learning_rate": 3.886014254336542e-06,
      "loss": 0.84903944,
      "num_input_tokens_seen": 23910110,
      "step": 1121,
      "time_per_iteration": 2.62040376663208
    },
    {
      "auxiliary_loss_clip": 0.01231713,
      "auxiliary_loss_mlp": 0.01032924,
      "balance_loss_clip": 1.05837977,
      "balance_loss_mlp": 1.02279735,
      "epoch": 0.13491252329706005,
      "flos": 23730525417600.0,
      "grad_norm": 1.8624896973007155,
      "language_loss": 0.92581761,
      "learning_rate": 3.885754890922168e-06,
      "loss": 0.94846404,
      "num_input_tokens_seen": 23930440,
      "step": 1122,
      "time_per_iteration": 2.690134286880493
    },
    {
      "auxiliary_loss_clip": 0.01225611,
      "auxiliary_loss_mlp": 0.01039713,
      "balance_loss_clip": 1.0492382,
      "balance_loss_mlp": 1.02876365,
      "epoch": 0.13503276618769916,
      "flos": 34127058960000.0,
      "grad_norm": 3.413164728424845,
      "language_loss": 0.78622133,
      "learning_rate": 3.885495241440277e-06,
      "loss": 0.80887461,
      "num_input_tokens_seen": 23954535,
      "step": 1123,
      "time_per_iteration": 2.7774901390075684
    },
    {
      "auxiliary_loss_clip": 0.01236395,
      "auxiliary_loss_mlp": 0.01038051,
      "balance_loss_clip": 1.06297588,
      "balance_loss_mlp": 1.02758396,
      "epoch": 0.13515300907833824,
      "flos": 17712328377600.0,
      "grad_norm": 1.8959063618477123,
      "language_loss": 0.74230278,
      "learning_rate": 3.885235305930257e-06,
      "loss": 0.76504719,
      "num_input_tokens_seen": 23972735,
      "step": 1124,
      "time_per_iteration": 2.591749668121338
    },
    {
      "auxiliary_loss_clip": 0.01238641,
      "auxiliary_loss_mlp": 0.01043379,
      "balance_loss_clip": 1.06238616,
      "balance_loss_mlp": 1.03161275,
      "epoch": 0.13527325196897733,
      "flos": 20260872201600.0,
      "grad_norm": 2.154985017726641,
      "language_loss": 0.85692096,
      "learning_rate": 3.884975084431539e-06,
      "loss": 0.87974107,
      "num_input_tokens_seen": 23987685,
      "step": 1125,
      "time_per_iteration": 2.6585984230041504
    },
    {
      "auxiliary_loss_clip": 0.01227992,
      "auxiliary_loss_mlp": 0.01209834,
      "balance_loss_clip": 1.05950427,
      "balance_loss_mlp": 1.00114977,
      "epoch": 0.13539349485961644,
      "flos": 18186492839040.0,
      "grad_norm": 3.9302117691263327,
      "language_loss": 0.91637552,
      "learning_rate": 3.8847145769836e-06,
      "loss": 0.94075382,
      "num_input_tokens_seen": 24004105,
      "step": 1126,
      "time_per_iteration": 2.6438117027282715
    },
    {
      "auxiliary_loss_clip": 0.01237623,
      "auxiliary_loss_mlp": 0.01036704,
      "balance_loss_clip": 1.06237912,
      "balance_loss_mlp": 1.02556431,
      "epoch": 0.13551373775025552,
      "flos": 19317463441920.0,
      "grad_norm": 2.5166299479930423,
      "language_loss": 0.66571122,
      "learning_rate": 3.884453783625959e-06,
      "loss": 0.68845445,
      "num_input_tokens_seen": 24021715,
      "step": 1127,
      "time_per_iteration": 2.6135358810424805
    },
    {
      "auxiliary_loss_clip": 0.01231783,
      "auxiliary_loss_mlp": 0.0103118,
      "balance_loss_clip": 1.05961704,
      "balance_loss_mlp": 1.02120876,
      "epoch": 0.1356339806408946,
      "flos": 20850813175680.0,
      "grad_norm": 3.0430726077845076,
      "language_loss": 0.84923446,
      "learning_rate": 3.884192704398176e-06,
      "loss": 0.87186408,
      "num_input_tokens_seen": 24038915,
      "step": 1128,
      "time_per_iteration": 2.700430154800415
    },
    {
      "auxiliary_loss_clip": 0.01236402,
      "auxiliary_loss_mlp": 0.01047455,
      "balance_loss_clip": 1.05915368,
      "balance_loss_mlp": 1.03691065,
      "epoch": 0.13575422353153369,
      "flos": 50476037696640.0,
      "grad_norm": 2.543170129548867,
      "language_loss": 0.74585617,
      "learning_rate": 3.883931339339858e-06,
      "loss": 0.76869476,
      "num_input_tokens_seen": 24063300,
      "step": 1129,
      "time_per_iteration": 2.8869714736938477
    },
    {
      "auxiliary_loss_clip": 0.01239044,
      "auxiliary_loss_mlp": 0.01034776,
      "balance_loss_clip": 1.05982447,
      "balance_loss_mlp": 1.02345753,
      "epoch": 0.1358744664221728,
      "flos": 18150797698560.0,
      "grad_norm": 3.0494718308720583,
      "language_loss": 0.79036903,
      "learning_rate": 3.883669688490654e-06,
      "loss": 0.81310725,
      "num_input_tokens_seen": 24081070,
      "step": 1130,
      "time_per_iteration": 2.6421279907226562
    },
    {
      "auxiliary_loss_clip": 0.01225163,
      "auxiliary_loss_mlp": 0.01209653,
      "balance_loss_clip": 1.05502367,
      "balance_loss_mlp": 1.00113392,
      "epoch": 0.13599470931281188,
      "flos": 18442966924800.0,
      "grad_norm": 2.6824477709780146,
      "language_loss": 0.85261989,
      "learning_rate": 3.883407751890256e-06,
      "loss": 0.87696803,
      "num_input_tokens_seen": 24099675,
      "step": 1131,
      "time_per_iteration": 2.6240081787109375
    },
    {
      "auxiliary_loss_clip": 0.01236954,
      "auxiliary_loss_mlp": 0.01042484,
      "balance_loss_clip": 1.05529571,
      "balance_loss_mlp": 1.03034294,
      "epoch": 0.13611495220345096,
      "flos": 26680766014080.0,
      "grad_norm": 5.066281737700317,
      "language_loss": 0.85696495,
      "learning_rate": 3.8831455295783994e-06,
      "loss": 0.87975931,
      "num_input_tokens_seen": 24118925,
      "step": 1132,
      "time_per_iteration": 2.687469720840454
    },
    {
      "auxiliary_loss_clip": 0.01232673,
      "auxiliary_loss_mlp": 0.01037897,
      "balance_loss_clip": 1.05937004,
      "balance_loss_mlp": 1.02687025,
      "epoch": 0.13623519509409007,
      "flos": 21686238673920.0,
      "grad_norm": 2.16418090501351,
      "language_loss": 0.74072647,
      "learning_rate": 3.882883021594864e-06,
      "loss": 0.76343215,
      "num_input_tokens_seen": 24137065,
      "step": 1133,
      "time_per_iteration": 2.6555027961730957
    },
    {
      "auxiliary_loss_clip": 0.01228893,
      "auxiliary_loss_mlp": 0.01036204,
      "balance_loss_clip": 1.05783141,
      "balance_loss_mlp": 1.02544558,
      "epoch": 0.13635543798472916,
      "flos": 14830389492480.0,
      "grad_norm": 2.104086976391345,
      "language_loss": 0.86628187,
      "learning_rate": 3.8826202279794705e-06,
      "loss": 0.88893282,
      "num_input_tokens_seen": 24154125,
      "step": 1134,
      "time_per_iteration": 2.6296911239624023
    },
    {
      "auxiliary_loss_clip": 0.01237292,
      "auxiliary_loss_mlp": 0.01036175,
      "balance_loss_clip": 1.06445277,
      "balance_loss_mlp": 1.02567232,
      "epoch": 0.13647568087536824,
      "flos": 22890323410560.0,
      "grad_norm": 2.676765285228512,
      "language_loss": 0.70054781,
      "learning_rate": 3.882357148772085e-06,
      "loss": 0.72328246,
      "num_input_tokens_seen": 24171550,
      "step": 1135,
      "time_per_iteration": 2.6664609909057617
    },
    {
      "auxiliary_loss_clip": 0.01222076,
      "auxiliary_loss_mlp": 0.01040251,
      "balance_loss_clip": 1.05490363,
      "balance_loss_mlp": 1.02897382,
      "epoch": 0.13659592376600732,
      "flos": 19937927998080.0,
      "grad_norm": 3.373450306647739,
      "language_loss": 0.8457669,
      "learning_rate": 3.882093784012617e-06,
      "loss": 0.86839014,
      "num_input_tokens_seen": 24190190,
      "step": 1136,
      "time_per_iteration": 2.6566452980041504
    },
    {
      "auxiliary_loss_clip": 0.01237238,
      "auxiliary_loss_mlp": 0.01034276,
      "balance_loss_clip": 1.06071329,
      "balance_loss_mlp": 1.02302909,
      "epoch": 0.13671616665664643,
      "flos": 21428579439360.0,
      "grad_norm": 2.0862699193783762,
      "language_loss": 0.84252644,
      "learning_rate": 3.881830133741019e-06,
      "loss": 0.86524153,
      "num_input_tokens_seen": 24209055,
      "step": 1137,
      "time_per_iteration": 2.6414566040039062
    },
    {
      "auxiliary_loss_clip": 0.01237806,
      "auxiliary_loss_mlp": 0.01041853,
      "balance_loss_clip": 1.0616864,
      "balance_loss_mlp": 1.03053999,
      "epoch": 0.13683640954728551,
      "flos": 22778138257920.0,
      "grad_norm": 2.4512232857447103,
      "language_loss": 0.76337802,
      "learning_rate": 3.881566197997285e-06,
      "loss": 0.7861746,
      "num_input_tokens_seen": 24225490,
      "step": 1138,
      "time_per_iteration": 2.6582887172698975
    },
    {
      "auxiliary_loss_clip": 0.01234832,
      "auxiliary_loss_mlp": 0.01038131,
      "balance_loss_clip": 1.06256902,
      "balance_loss_mlp": 1.02738428,
      "epoch": 0.1369566524379246,
      "flos": 21725884310400.0,
      "grad_norm": 1.6447166860461435,
      "language_loss": 0.74903351,
      "learning_rate": 3.881301976821456e-06,
      "loss": 0.77176309,
      "num_input_tokens_seen": 24245520,
      "step": 1139,
      "time_per_iteration": 2.717454195022583
    },
    {
      "auxiliary_loss_clip": 0.01233962,
      "auxiliary_loss_mlp": 0.01038498,
      "balance_loss_clip": 1.06169248,
      "balance_loss_mlp": 1.02820385,
      "epoch": 0.1370768953285637,
      "flos": 18624459369600.0,
      "grad_norm": 1.952924298306541,
      "language_loss": 0.90381241,
      "learning_rate": 3.881037470253612e-06,
      "loss": 0.92653704,
      "num_input_tokens_seen": 24265035,
      "step": 1140,
      "time_per_iteration": 2.573523759841919
    },
    {
      "auxiliary_loss_clip": 0.01241222,
      "auxiliary_loss_mlp": 0.01039334,
      "balance_loss_clip": 1.05838418,
      "balance_loss_mlp": 1.02903461,
      "epoch": 0.1371971382192028,
      "flos": 14939521989120.0,
      "grad_norm": 3.6441204416505193,
      "language_loss": 0.79724985,
      "learning_rate": 3.88077267833388e-06,
      "loss": 0.82005537,
      "num_input_tokens_seen": 24281550,
      "step": 1141,
      "time_per_iteration": 2.6604273319244385
    },
    {
      "auxiliary_loss_clip": 0.01236697,
      "auxiliary_loss_mlp": 0.01043938,
      "balance_loss_clip": 1.05624616,
      "balance_loss_mlp": 1.03273869,
      "epoch": 0.13731738110984187,
      "flos": 19023785844480.0,
      "grad_norm": 2.1049141696635445,
      "language_loss": 0.84164548,
      "learning_rate": 3.880507601102427e-06,
      "loss": 0.86445189,
      "num_input_tokens_seen": 24299485,
      "step": 1142,
      "time_per_iteration": 3.6442737579345703
    },
    {
      "auxiliary_loss_clip": 0.01237399,
      "auxiliary_loss_mlp": 0.01042873,
      "balance_loss_clip": 1.06572747,
      "balance_loss_mlp": 1.03213787,
      "epoch": 0.13743762400048098,
      "flos": 18187462506240.0,
      "grad_norm": 2.898636806699506,
      "language_loss": 0.82148021,
      "learning_rate": 3.880242238599467e-06,
      "loss": 0.84428298,
      "num_input_tokens_seen": 24316010,
      "step": 1143,
      "time_per_iteration": 2.5963385105133057
    },
    {
      "auxiliary_loss_clip": 0.01234609,
      "auxiliary_loss_mlp": 0.01038621,
      "balance_loss_clip": 1.06393421,
      "balance_loss_mlp": 1.02763605,
      "epoch": 0.13755786689112007,
      "flos": 21031982398080.0,
      "grad_norm": 2.3706586393410785,
      "language_loss": 0.83322096,
      "learning_rate": 3.879976590865254e-06,
      "loss": 0.85595334,
      "num_input_tokens_seen": 24335465,
      "step": 1144,
      "time_per_iteration": 2.6508121490478516
    },
    {
      "auxiliary_loss_clip": 0.01238192,
      "auxiliary_loss_mlp": 0.01039962,
      "balance_loss_clip": 1.0631597,
      "balance_loss_mlp": 1.0287627,
      "epoch": 0.13767810978175915,
      "flos": 21360636864000.0,
      "grad_norm": 2.6482673284203346,
      "language_loss": 0.87355387,
      "learning_rate": 3.879710657940087e-06,
      "loss": 0.89633548,
      "num_input_tokens_seen": 24354415,
      "step": 1145,
      "time_per_iteration": 3.5846824645996094
    },
    {
      "auxiliary_loss_clip": 0.01240993,
      "auxiliary_loss_mlp": 0.01047861,
      "balance_loss_clip": 1.06350863,
      "balance_loss_mlp": 1.03589284,
      "epoch": 0.13779835267239823,
      "flos": 30592084861440.0,
      "grad_norm": 2.39860860620535,
      "language_loss": 0.70367992,
      "learning_rate": 3.879444439864308e-06,
      "loss": 0.72656846,
      "num_input_tokens_seen": 24373990,
      "step": 1146,
      "time_per_iteration": 3.581679105758667
    },
    {
      "auxiliary_loss_clip": 0.01236909,
      "auxiliary_loss_mlp": 0.01209978,
      "balance_loss_clip": 1.06132185,
      "balance_loss_mlp": 1.00128126,
      "epoch": 0.13791859556303734,
      "flos": 22669867687680.0,
      "grad_norm": 1.8352391297369621,
      "language_loss": 0.85963756,
      "learning_rate": 3.879177936678301e-06,
      "loss": 0.88410652,
      "num_input_tokens_seen": 24392995,
      "step": 1147,
      "time_per_iteration": 2.6900312900543213
    },
    {
      "auxiliary_loss_clip": 0.01243011,
      "auxiliary_loss_mlp": 0.01042963,
      "balance_loss_clip": 1.06169748,
      "balance_loss_mlp": 1.03164423,
      "epoch": 0.13803883845367643,
      "flos": 35224166016000.0,
      "grad_norm": 2.0099419033994477,
      "language_loss": 0.77169085,
      "learning_rate": 3.878911148422496e-06,
      "loss": 0.79455054,
      "num_input_tokens_seen": 24414470,
      "step": 1148,
      "time_per_iteration": 2.7527682781219482
    },
    {
      "auxiliary_loss_clip": 0.01236516,
      "auxiliary_loss_mlp": 0.01036617,
      "balance_loss_clip": 1.06150758,
      "balance_loss_mlp": 1.02509534,
      "epoch": 0.1381590813443155,
      "flos": 32014542332160.0,
      "grad_norm": 2.2007120362614057,
      "language_loss": 0.70391941,
      "learning_rate": 3.878644075137364e-06,
      "loss": 0.72665077,
      "num_input_tokens_seen": 24435120,
      "step": 1149,
      "time_per_iteration": 2.6999382972717285
    },
    {
      "auxiliary_loss_clip": 0.01218987,
      "auxiliary_loss_mlp": 0.01036134,
      "balance_loss_clip": 1.05411005,
      "balance_loss_mlp": 1.02517819,
      "epoch": 0.13827932423495462,
      "flos": 17821855923840.0,
      "grad_norm": 2.68249221519675,
      "language_loss": 0.79268479,
      "learning_rate": 3.878376716863418e-06,
      "loss": 0.81523603,
      "num_input_tokens_seen": 24451420,
      "step": 1150,
      "time_per_iteration": 2.6846210956573486
    },
    {
      "auxiliary_loss_clip": 0.01233578,
      "auxiliary_loss_mlp": 0.01043383,
      "balance_loss_clip": 1.05622149,
      "balance_loss_mlp": 1.03121793,
      "epoch": 0.1383995671255937,
      "flos": 19427098728960.0,
      "grad_norm": 2.4395817344154347,
      "language_loss": 0.72079968,
      "learning_rate": 3.878109073641219e-06,
      "loss": 0.74356931,
      "num_input_tokens_seen": 24470450,
      "step": 1151,
      "time_per_iteration": 2.6762568950653076
    },
    {
      "auxiliary_loss_clip": 0.0123769,
      "auxiliary_loss_mlp": 0.01037991,
      "balance_loss_clip": 1.05742013,
      "balance_loss_mlp": 1.02702379,
      "epoch": 0.13851981001623279,
      "flos": 28296603331200.0,
      "grad_norm": 1.6685608614601877,
      "language_loss": 0.81209397,
      "learning_rate": 3.877841145511366e-06,
      "loss": 0.83485079,
      "num_input_tokens_seen": 24493190,
      "step": 1152,
      "time_per_iteration": 2.753849983215332
    },
    {
      "auxiliary_loss_clip": 0.01241352,
      "auxiliary_loss_mlp": 0.01041326,
      "balance_loss_clip": 1.06293559,
      "balance_loss_mlp": 1.03001332,
      "epoch": 0.13864005290687187,
      "flos": 21213079793280.0,
      "grad_norm": 2.0241288309860583,
      "language_loss": 0.82716835,
      "learning_rate": 3.8775729325145035e-06,
      "loss": 0.8499952,
      "num_input_tokens_seen": 24512425,
      "step": 1153,
      "time_per_iteration": 2.646841287612915
    },
    {
      "auxiliary_loss_clip": 0.01129256,
      "auxiliary_loss_mlp": 0.01007949,
      "balance_loss_clip": 1.02228963,
      "balance_loss_mlp": 1.00505209,
      "epoch": 0.13876029579751098,
      "flos": 71653389413760.0,
      "grad_norm": 0.7882873302209503,
      "language_loss": 0.64643008,
      "learning_rate": 3.877304434691321e-06,
      "loss": 0.66780216,
      "num_input_tokens_seen": 24579275,
      "step": 1154,
      "time_per_iteration": 3.3208045959472656
    },
    {
      "auxiliary_loss_clip": 0.01239214,
      "auxiliary_loss_mlp": 0.01032375,
      "balance_loss_clip": 1.06104851,
      "balance_loss_mlp": 1.02205729,
      "epoch": 0.13888053868815006,
      "flos": 21941348042880.0,
      "grad_norm": 3.141693288132136,
      "language_loss": 0.8003099,
      "learning_rate": 3.877035652082548e-06,
      "loss": 0.82302576,
      "num_input_tokens_seen": 24598720,
      "step": 1155,
      "time_per_iteration": 2.7152440547943115
    },
    {
      "auxiliary_loss_clip": 0.01228614,
      "auxiliary_loss_mlp": 0.01034476,
      "balance_loss_clip": 1.05924928,
      "balance_loss_mlp": 1.02302647,
      "epoch": 0.13900078157878915,
      "flos": 19608627087360.0,
      "grad_norm": 2.0526371335594704,
      "language_loss": 0.8508445,
      "learning_rate": 3.87676658472896e-06,
      "loss": 0.87347543,
      "num_input_tokens_seen": 24617530,
      "step": 1156,
      "time_per_iteration": 2.6621484756469727
    },
    {
      "auxiliary_loss_clip": 0.01235775,
      "auxiliary_loss_mlp": 0.01043354,
      "balance_loss_clip": 1.05937719,
      "balance_loss_mlp": 1.03225565,
      "epoch": 0.13912102446942826,
      "flos": 22638051216000.0,
      "grad_norm": 2.5899088004164104,
      "language_loss": 0.85418332,
      "learning_rate": 3.876497232671372e-06,
      "loss": 0.87697458,
      "num_input_tokens_seen": 24637485,
      "step": 1157,
      "time_per_iteration": 2.6422793865203857
    },
    {
      "auxiliary_loss_clip": 0.01243994,
      "auxiliary_loss_mlp": 0.0103796,
      "balance_loss_clip": 1.05593228,
      "balance_loss_mlp": 1.02728438,
      "epoch": 0.13924126736006734,
      "flos": 29643324975360.0,
      "grad_norm": 3.530714652350146,
      "language_loss": 0.83585811,
      "learning_rate": 3.876227595950647e-06,
      "loss": 0.85867763,
      "num_input_tokens_seen": 24656915,
      "step": 1158,
      "time_per_iteration": 2.7661361694335938
    },
    {
      "auxiliary_loss_clip": 0.01236619,
      "auxiliary_loss_mlp": 0.01042107,
      "balance_loss_clip": 1.0647862,
      "balance_loss_mlp": 1.03019249,
      "epoch": 0.13936151025070642,
      "flos": 27417653527680.0,
      "grad_norm": 5.3186596332967815,
      "language_loss": 0.79020572,
      "learning_rate": 3.875957674607686e-06,
      "loss": 0.81299299,
      "num_input_tokens_seen": 24679190,
      "step": 1159,
      "time_per_iteration": 2.635967493057251
    },
    {
      "auxiliary_loss_clip": 0.01231168,
      "auxiliary_loss_mlp": 0.0121046,
      "balance_loss_clip": 1.05944586,
      "balance_loss_mlp": 1.00124753,
      "epoch": 0.1394817531413455,
      "flos": 16399326625920.0,
      "grad_norm": 2.1388682000418306,
      "language_loss": 0.87973773,
      "learning_rate": 3.8756874686834386e-06,
      "loss": 0.90415406,
      "num_input_tokens_seen": 24697405,
      "step": 1160,
      "time_per_iteration": 2.6466634273529053
    },
    {
      "auxiliary_loss_clip": 0.01242327,
      "auxiliary_loss_mlp": 0.01210814,
      "balance_loss_clip": 1.06139779,
      "balance_loss_mlp": 1.00122035,
      "epoch": 0.13960199603198462,
      "flos": 30922319525760.0,
      "grad_norm": 2.058482693762898,
      "language_loss": 0.80148238,
      "learning_rate": 3.875416978218893e-06,
      "loss": 0.82601374,
      "num_input_tokens_seen": 24720600,
      "step": 1161,
      "time_per_iteration": 2.763411045074463
    },
    {
      "auxiliary_loss_clip": 0.01247338,
      "auxiliary_loss_mlp": 0.0103978,
      "balance_loss_clip": 1.05445814,
      "balance_loss_mlp": 1.02828193,
      "epoch": 0.1397222389226237,
      "flos": 18113773754880.0,
      "grad_norm": 2.449500522449049,
      "language_loss": 0.8270725,
      "learning_rate": 3.8751462032550835e-06,
      "loss": 0.8499437,
      "num_input_tokens_seen": 24737605,
      "step": 1162,
      "time_per_iteration": 2.75223708152771
    },
    {
      "auxiliary_loss_clip": 0.01236627,
      "auxiliary_loss_mlp": 0.01025561,
      "balance_loss_clip": 1.06384218,
      "balance_loss_mlp": 1.01499903,
      "epoch": 0.13984248181326278,
      "flos": 16872772815360.0,
      "grad_norm": 4.4732207225115515,
      "language_loss": 0.82620561,
      "learning_rate": 3.874875143833085e-06,
      "loss": 0.84882748,
      "num_input_tokens_seen": 24755845,
      "step": 1163,
      "time_per_iteration": 2.741142749786377
    },
    {
      "auxiliary_loss_clip": 0.01239492,
      "auxiliary_loss_mlp": 0.01044522,
      "balance_loss_clip": 1.06259322,
      "balance_loss_mlp": 1.0324347,
      "epoch": 0.1399627247039019,
      "flos": 54121401267840.0,
      "grad_norm": 2.004164124744088,
      "language_loss": 0.68837023,
      "learning_rate": 3.874603799994019e-06,
      "loss": 0.71121037,
      "num_input_tokens_seen": 24779380,
      "step": 1164,
      "time_per_iteration": 2.9321107864379883
    },
    {
      "auxiliary_loss_clip": 0.0123191,
      "auxiliary_loss_mlp": 0.01033533,
      "balance_loss_clip": 1.05597246,
      "balance_loss_mlp": 1.0232749,
      "epoch": 0.14008296759454097,
      "flos": 11765521618560.0,
      "grad_norm": 2.2764656663712413,
      "language_loss": 0.86733603,
      "learning_rate": 3.874332171779046e-06,
      "loss": 0.88999051,
      "num_input_tokens_seen": 24794260,
      "step": 1165,
      "time_per_iteration": 2.6385514736175537
    },
    {
      "auxiliary_loss_clip": 0.01234073,
      "auxiliary_loss_mlp": 0.01031861,
      "balance_loss_clip": 1.0549562,
      "balance_loss_mlp": 1.02082229,
      "epoch": 0.14020321048518006,
      "flos": 22017514832640.0,
      "grad_norm": 1.788423929937744,
      "language_loss": 0.75828391,
      "learning_rate": 3.874060259229373e-06,
      "loss": 0.78094327,
      "num_input_tokens_seen": 24815835,
      "step": 1166,
      "time_per_iteration": 2.766230583190918
    },
    {
      "auxiliary_loss_clip": 0.01241118,
      "auxiliary_loss_mlp": 0.01043356,
      "balance_loss_clip": 1.06472278,
      "balance_loss_mlp": 1.03204942,
      "epoch": 0.14032345337581917,
      "flos": 23404313076480.0,
      "grad_norm": 3.0847212111575857,
      "language_loss": 0.93369371,
      "learning_rate": 3.873788062386249e-06,
      "loss": 0.95653844,
      "num_input_tokens_seen": 24834095,
      "step": 1167,
      "time_per_iteration": 2.6478283405303955
    },
    {
      "auxiliary_loss_clip": 0.01245477,
      "auxiliary_loss_mlp": 0.01041412,
      "balance_loss_clip": 1.06281638,
      "balance_loss_mlp": 1.03051591,
      "epoch": 0.14044369626645825,
      "flos": 29645767100160.0,
      "grad_norm": 2.0146432006176993,
      "language_loss": 0.82111001,
      "learning_rate": 3.873515581290965e-06,
      "loss": 0.84397888,
      "num_input_tokens_seen": 24858900,
      "step": 1168,
      "time_per_iteration": 3.72257137298584
    },
    {
      "auxiliary_loss_clip": 0.01242985,
      "auxiliary_loss_mlp": 0.01036574,
      "balance_loss_clip": 1.06233776,
      "balance_loss_mlp": 1.02490377,
      "epoch": 0.14056393915709733,
      "flos": 18332972501760.0,
      "grad_norm": 2.3610265903883527,
      "language_loss": 0.75699443,
      "learning_rate": 3.8732428159848575e-06,
      "loss": 0.77978998,
      "num_input_tokens_seen": 24877875,
      "step": 1169,
      "time_per_iteration": 3.544726610183716
    },
    {
      "auxiliary_loss_clip": 0.01237401,
      "auxiliary_loss_mlp": 0.01038175,
      "balance_loss_clip": 1.06465793,
      "balance_loss_mlp": 1.0264926,
      "epoch": 0.14068418204773642,
      "flos": 26687517770880.0,
      "grad_norm": 2.871458463993046,
      "language_loss": 0.78236961,
      "learning_rate": 3.872969766509304e-06,
      "loss": 0.80512536,
      "num_input_tokens_seen": 24898430,
      "step": 1170,
      "time_per_iteration": 2.6917877197265625
    },
    {
      "auxiliary_loss_clip": 0.01135755,
      "auxiliary_loss_mlp": 0.01034457,
      "balance_loss_clip": 1.0250237,
      "balance_loss_mlp": 1.03114307,
      "epoch": 0.14080442493837553,
      "flos": 65259314501760.0,
      "grad_norm": 0.7742841393979581,
      "language_loss": 0.55621558,
      "learning_rate": 3.872696432905726e-06,
      "loss": 0.5779177,
      "num_input_tokens_seen": 24959250,
      "step": 1171,
      "time_per_iteration": 3.224048376083374
    },
    {
      "auxiliary_loss_clip": 0.01243676,
      "auxiliary_loss_mlp": 0.01048106,
      "balance_loss_clip": 1.06182528,
      "balance_loss_mlp": 1.03628659,
      "epoch": 0.1409246678290146,
      "flos": 25776715582080.0,
      "grad_norm": 4.747339261042205,
      "language_loss": 0.72079492,
      "learning_rate": 3.872422815215589e-06,
      "loss": 0.74371266,
      "num_input_tokens_seen": 24978330,
      "step": 1172,
      "time_per_iteration": 4.509531259536743
    },
    {
      "auxiliary_loss_clip": 0.01232128,
      "auxiliary_loss_mlp": 0.01039646,
      "balance_loss_clip": 1.05750966,
      "balance_loss_mlp": 1.02709305,
      "epoch": 0.1410449107196537,
      "flos": 21868521217920.0,
      "grad_norm": 1.9419992619883555,
      "language_loss": 0.74570239,
      "learning_rate": 3.8721489134803994e-06,
      "loss": 0.76842016,
      "num_input_tokens_seen": 24997120,
      "step": 1173,
      "time_per_iteration": 2.662771463394165
    },
    {
      "auxiliary_loss_clip": 0.0123735,
      "auxiliary_loss_mlp": 0.01047273,
      "balance_loss_clip": 1.06326127,
      "balance_loss_mlp": 1.0354538,
      "epoch": 0.1411651536102928,
      "flos": 16684133564160.0,
      "grad_norm": 2.1009096689908002,
      "language_loss": 0.72332501,
      "learning_rate": 3.871874727741707e-06,
      "loss": 0.74617124,
      "num_input_tokens_seen": 25014350,
      "step": 1174,
      "time_per_iteration": 2.586514472961426
    },
    {
      "auxiliary_loss_clip": 0.01234775,
      "auxiliary_loss_mlp": 0.0104105,
      "balance_loss_clip": 1.0636369,
      "balance_loss_mlp": 1.03026152,
      "epoch": 0.1412853965009319,
      "flos": 20992264934400.0,
      "grad_norm": 2.900299162658211,
      "language_loss": 0.96482974,
      "learning_rate": 3.871600258041108e-06,
      "loss": 0.98758799,
      "num_input_tokens_seen": 25033875,
      "step": 1175,
      "time_per_iteration": 2.6495583057403564
    },
    {
      "auxiliary_loss_clip": 0.01236034,
      "auxiliary_loss_mlp": 0.01042472,
      "balance_loss_clip": 1.05788934,
      "balance_loss_mlp": 1.03021169,
      "epoch": 0.14140563939157097,
      "flos": 20335279224960.0,
      "grad_norm": 2.3686985630345103,
      "language_loss": 0.86242086,
      "learning_rate": 3.871325504420238e-06,
      "loss": 0.88520592,
      "num_input_tokens_seen": 25052865,
      "step": 1176,
      "time_per_iteration": 2.6595420837402344
    },
    {
      "auxiliary_loss_clip": 0.01236986,
      "auxiliary_loss_mlp": 0.01034894,
      "balance_loss_clip": 1.0642767,
      "balance_loss_mlp": 1.02390337,
      "epoch": 0.14152588228221005,
      "flos": 21068826773760.0,
      "grad_norm": 2.1269858026599877,
      "language_loss": 0.81960839,
      "learning_rate": 3.871050466920776e-06,
      "loss": 0.84232718,
      "num_input_tokens_seen": 25072770,
      "step": 1177,
      "time_per_iteration": 2.6279475688934326
    },
    {
      "auxiliary_loss_clip": 0.01230726,
      "auxiliary_loss_mlp": 0.01030082,
      "balance_loss_clip": 1.05512667,
      "balance_loss_mlp": 1.01968729,
      "epoch": 0.14164612517284916,
      "flos": 18223157646720.0,
      "grad_norm": 2.1400322594243897,
      "language_loss": 0.79988694,
      "learning_rate": 3.870775145584447e-06,
      "loss": 0.82249504,
      "num_input_tokens_seen": 25090550,
      "step": 1178,
      "time_per_iteration": 2.6045727729797363
    },
    {
      "auxiliary_loss_clip": 0.01246064,
      "auxiliary_loss_mlp": 0.01046312,
      "balance_loss_clip": 1.06060004,
      "balance_loss_mlp": 1.03443909,
      "epoch": 0.14176636806348825,
      "flos": 22744454279040.0,
      "grad_norm": 2.9203954118657323,
      "language_loss": 0.64469814,
      "learning_rate": 3.8704995404530145e-06,
      "loss": 0.66762191,
      "num_input_tokens_seen": 25106175,
      "step": 1179,
      "time_per_iteration": 2.6740331649780273
    },
    {
      "auxiliary_loss_clip": 0.01234858,
      "auxiliary_loss_mlp": 0.01040883,
      "balance_loss_clip": 1.0641737,
      "balance_loss_mlp": 1.0302372,
      "epoch": 0.14188661095412733,
      "flos": 22091095843200.0,
      "grad_norm": 2.5270627756505326,
      "language_loss": 0.85273063,
      "learning_rate": 3.87022365156829e-06,
      "loss": 0.8754881,
      "num_input_tokens_seen": 25126890,
      "step": 1180,
      "time_per_iteration": 2.5976080894470215
    },
    {
      "auxiliary_loss_clip": 0.01242021,
      "auxiliary_loss_mlp": 0.01037897,
      "balance_loss_clip": 1.05235171,
      "balance_loss_mlp": 1.02685809,
      "epoch": 0.14200685384476644,
      "flos": 24352390604160.0,
      "grad_norm": 2.5213325379723748,
      "language_loss": 0.8115378,
      "learning_rate": 3.869947478972123e-06,
      "loss": 0.834337,
      "num_input_tokens_seen": 25147915,
      "step": 1181,
      "time_per_iteration": 2.7917301654815674
    },
    {
      "auxiliary_loss_clip": 0.01233593,
      "auxiliary_loss_mlp": 0.01040286,
      "balance_loss_clip": 1.06134462,
      "balance_loss_mlp": 1.02802563,
      "epoch": 0.14212709673540552,
      "flos": 24022048199040.0,
      "grad_norm": 3.0830924782885294,
      "language_loss": 0.82578874,
      "learning_rate": 3.869671022706412e-06,
      "loss": 0.84852755,
      "num_input_tokens_seen": 25166645,
      "step": 1182,
      "time_per_iteration": 2.641855478286743
    },
    {
      "auxiliary_loss_clip": 0.01226441,
      "auxiliary_loss_mlp": 0.01043906,
      "balance_loss_clip": 1.05205154,
      "balance_loss_mlp": 1.03255773,
      "epoch": 0.1422473396260446,
      "flos": 26431797870720.0,
      "grad_norm": 1.8869708966390764,
      "language_loss": 0.64928269,
      "learning_rate": 3.869394282813092e-06,
      "loss": 0.67198616,
      "num_input_tokens_seen": 25185845,
      "step": 1183,
      "time_per_iteration": 2.7510664463043213
    },
    {
      "auxiliary_loss_clip": 0.01246484,
      "auxiliary_loss_mlp": 0.01042798,
      "balance_loss_clip": 1.05707717,
      "balance_loss_mlp": 1.03173494,
      "epoch": 0.1423675825166837,
      "flos": 17055306754560.0,
      "grad_norm": 2.600031016627932,
      "language_loss": 0.89413953,
      "learning_rate": 3.869117259334147e-06,
      "loss": 0.91703236,
      "num_input_tokens_seen": 25203770,
      "step": 1184,
      "time_per_iteration": 2.7134056091308594
    },
    {
      "auxiliary_loss_clip": 0.01232569,
      "auxiliary_loss_mlp": 0.01039829,
      "balance_loss_clip": 1.05948353,
      "balance_loss_mlp": 1.02897525,
      "epoch": 0.1424878254073228,
      "flos": 17929480049280.0,
      "grad_norm": 2.139349323762505,
      "language_loss": 0.82275975,
      "learning_rate": 3.868839952311599e-06,
      "loss": 0.84548378,
      "num_input_tokens_seen": 25221725,
      "step": 1185,
      "time_per_iteration": 2.6414787769317627
    },
    {
      "auxiliary_loss_clip": 0.01232401,
      "auxiliary_loss_mlp": 0.01039124,
      "balance_loss_clip": 1.05839443,
      "balance_loss_mlp": 1.02797782,
      "epoch": 0.14260806829796188,
      "flos": 20303606407680.0,
      "grad_norm": 2.032106060354682,
      "language_loss": 0.80531359,
      "learning_rate": 3.868562361787516e-06,
      "loss": 0.8280288,
      "num_input_tokens_seen": 25240855,
      "step": 1186,
      "time_per_iteration": 2.6649701595306396
    },
    {
      "auxiliary_loss_clip": 0.012351,
      "auxiliary_loss_mlp": 0.01031295,
      "balance_loss_clip": 1.04879117,
      "balance_loss_mlp": 1.02022099,
      "epoch": 0.14272831118860096,
      "flos": 23185724860800.0,
      "grad_norm": 2.0672737102474783,
      "language_loss": 0.69141251,
      "learning_rate": 3.868284487804009e-06,
      "loss": 0.71407646,
      "num_input_tokens_seen": 25260085,
      "step": 1187,
      "time_per_iteration": 2.7558882236480713
    },
    {
      "auxiliary_loss_clip": 0.01242339,
      "auxiliary_loss_mlp": 0.01043639,
      "balance_loss_clip": 1.05799413,
      "balance_loss_mlp": 1.03275537,
      "epoch": 0.14284855407924008,
      "flos": 27232210586880.0,
      "grad_norm": 1.7846371058329775,
      "language_loss": 0.78275931,
      "learning_rate": 3.86800633040323e-06,
      "loss": 0.80561912,
      "num_input_tokens_seen": 25280675,
      "step": 1188,
      "time_per_iteration": 2.7238245010375977
    },
    {
      "auxiliary_loss_clip": 0.01236302,
      "auxiliary_loss_mlp": 0.01209916,
      "balance_loss_clip": 1.0614866,
      "balance_loss_mlp": 1.00088716,
      "epoch": 0.14296879696987916,
      "flos": 28184202696960.0,
      "grad_norm": 2.2695627365477273,
      "language_loss": 0.78159118,
      "learning_rate": 3.867727889627376e-06,
      "loss": 0.8060534,
      "num_input_tokens_seen": 25300290,
      "step": 1189,
      "time_per_iteration": 2.697890520095825
    },
    {
      "auxiliary_loss_clip": 0.01229207,
      "auxiliary_loss_mlp": 0.01039754,
      "balance_loss_clip": 1.05644846,
      "balance_loss_mlp": 1.02755904,
      "epoch": 0.14308903986051824,
      "flos": 19390290266880.0,
      "grad_norm": 2.3801156294732615,
      "language_loss": 0.78410161,
      "learning_rate": 3.867449165518687e-06,
      "loss": 0.80679119,
      "num_input_tokens_seen": 25316760,
      "step": 1190,
      "time_per_iteration": 2.6348423957824707
    },
    {
      "auxiliary_loss_clip": 0.01238786,
      "auxiliary_loss_mlp": 0.01210832,
      "balance_loss_clip": 1.06342077,
      "balance_loss_mlp": 1.00079846,
      "epoch": 0.14320928275115732,
      "flos": 17457506317440.0,
      "grad_norm": 1.9583034591356074,
      "language_loss": 0.71281189,
      "learning_rate": 3.867170158119444e-06,
      "loss": 0.73730808,
      "num_input_tokens_seen": 25335760,
      "step": 1191,
      "time_per_iteration": 2.6258697509765625
    },
    {
      "auxiliary_loss_clip": 0.01237683,
      "auxiliary_loss_mlp": 0.01039643,
      "balance_loss_clip": 1.06282616,
      "balance_loss_mlp": 1.02888405,
      "epoch": 0.14332952564179643,
      "flos": 21466070259840.0,
      "grad_norm": 2.1145949482153927,
      "language_loss": 0.75886989,
      "learning_rate": 3.866890867471972e-06,
      "loss": 0.78164315,
      "num_input_tokens_seen": 25354230,
      "step": 1192,
      "time_per_iteration": 2.611720085144043
    },
    {
      "auxiliary_loss_clip": 0.01233101,
      "auxiliary_loss_mlp": 0.01047174,
      "balance_loss_clip": 1.05359459,
      "balance_loss_mlp": 1.03521764,
      "epoch": 0.14344976853243552,
      "flos": 16396992241920.0,
      "grad_norm": 3.3567237240491403,
      "language_loss": 0.89573133,
      "learning_rate": 3.86661129361864e-06,
      "loss": 0.91853404,
      "num_input_tokens_seen": 25368720,
      "step": 1193,
      "time_per_iteration": 2.6162147521972656
    },
    {
      "auxiliary_loss_clip": 0.012356,
      "auxiliary_loss_mlp": 0.0104681,
      "balance_loss_clip": 1.05927896,
      "balance_loss_mlp": 1.03503811,
      "epoch": 0.1435700114230746,
      "flos": 18916736336640.0,
      "grad_norm": 4.910409631490808,
      "language_loss": 0.86232752,
      "learning_rate": 3.866331436601859e-06,
      "loss": 0.88515157,
      "num_input_tokens_seen": 25386715,
      "step": 1194,
      "time_per_iteration": 2.7549989223480225
    },
    {
      "auxiliary_loss_clip": 0.01236243,
      "auxiliary_loss_mlp": 0.01041109,
      "balance_loss_clip": 1.06326485,
      "balance_loss_mlp": 1.02984345,
      "epoch": 0.1436902543137137,
      "flos": 19755394058880.0,
      "grad_norm": 2.5158539515535017,
      "language_loss": 0.73318362,
      "learning_rate": 3.866051296464083e-06,
      "loss": 0.75595707,
      "num_input_tokens_seen": 25405550,
      "step": 1195,
      "time_per_iteration": 4.4665446281433105
    },
    {
      "auxiliary_loss_clip": 0.01235627,
      "auxiliary_loss_mlp": 0.01210153,
      "balance_loss_clip": 1.06010532,
      "balance_loss_mlp": 1.00082493,
      "epoch": 0.1438104972043528,
      "flos": 14684807669760.0,
      "grad_norm": 2.394056859391376,
      "language_loss": 0.85202873,
      "learning_rate": 3.86577087324781e-06,
      "loss": 0.87648654,
      "num_input_tokens_seen": 25422040,
      "step": 1196,
      "time_per_iteration": 2.5437324047088623
    },
    {
      "auxiliary_loss_clip": 0.01233493,
      "auxiliary_loss_mlp": 0.01036592,
      "balance_loss_clip": 1.06161404,
      "balance_loss_mlp": 1.02550554,
      "epoch": 0.14393074009499188,
      "flos": 17092330698240.0,
      "grad_norm": 2.355303137692086,
      "language_loss": 0.77261001,
      "learning_rate": 3.865490166995578e-06,
      "loss": 0.79531085,
      "num_input_tokens_seen": 25440270,
      "step": 1197,
      "time_per_iteration": 2.6348259449005127
    },
    {
      "auxiliary_loss_clip": 0.01236271,
      "auxiliary_loss_mlp": 0.01043029,
      "balance_loss_clip": 1.06114197,
      "balance_loss_mlp": 1.03153682,
      "epoch": 0.144050982985631,
      "flos": 30476200608000.0,
      "grad_norm": 4.310672693771425,
      "language_loss": 0.84173644,
      "learning_rate": 3.86520917774997e-06,
      "loss": 0.86452943,
      "num_input_tokens_seen": 25459705,
      "step": 1198,
      "time_per_iteration": 2.6558220386505127
    },
    {
      "auxiliary_loss_clip": 0.01232563,
      "auxiliary_loss_mlp": 0.01043967,
      "balance_loss_clip": 1.06068015,
      "balance_loss_mlp": 1.03339911,
      "epoch": 0.14417122587627007,
      "flos": 17858484817920.0,
      "grad_norm": 3.1702147863370786,
      "language_loss": 0.75345421,
      "learning_rate": 3.864927905553614e-06,
      "loss": 0.77621949,
      "num_input_tokens_seen": 25477615,
      "step": 1199,
      "time_per_iteration": 4.438794136047363
    },
    {
      "auxiliary_loss_clip": 0.01231763,
      "auxiliary_loss_mlp": 0.01042134,
      "balance_loss_clip": 1.05404234,
      "balance_loss_mlp": 1.03138685,
      "epoch": 0.14429146876690915,
      "flos": 21613914639360.0,
      "grad_norm": 1.9552760605554529,
      "language_loss": 0.89032137,
      "learning_rate": 3.8646463504491765e-06,
      "loss": 0.91306031,
      "num_input_tokens_seen": 25497750,
      "step": 1200,
      "time_per_iteration": 2.681645393371582
    },
    {
      "auxiliary_loss_clip": 0.01238329,
      "auxiliary_loss_mlp": 0.01039523,
      "balance_loss_clip": 1.06303942,
      "balance_loss_mlp": 1.02760172,
      "epoch": 0.14441171165754824,
      "flos": 23258120722560.0,
      "grad_norm": 2.6327429450708673,
      "language_loss": 0.83354783,
      "learning_rate": 3.8643645124793705e-06,
      "loss": 0.8563264,
      "num_input_tokens_seen": 25516650,
      "step": 1201,
      "time_per_iteration": 2.653702974319458
    },
    {
      "auxiliary_loss_clip": 0.01232807,
      "auxiliary_loss_mlp": 0.01035511,
      "balance_loss_clip": 1.05863261,
      "balance_loss_mlp": 1.02452028,
      "epoch": 0.14453195454818735,
      "flos": 42854213963520.0,
      "grad_norm": 1.8111924428733501,
      "language_loss": 0.74868077,
      "learning_rate": 3.8640823916869515e-06,
      "loss": 0.77136391,
      "num_input_tokens_seen": 25540960,
      "step": 1202,
      "time_per_iteration": 2.888051748275757
    },
    {
      "auxiliary_loss_clip": 0.01234687,
      "auxiliary_loss_mlp": 0.01037703,
      "balance_loss_clip": 1.06158447,
      "balance_loss_mlp": 1.02668262,
      "epoch": 0.14465219743882643,
      "flos": 27235873774080.0,
      "grad_norm": 3.063383204467658,
      "language_loss": 0.78617895,
      "learning_rate": 3.863799988114714e-06,
      "loss": 0.80890286,
      "num_input_tokens_seen": 25562990,
      "step": 1203,
      "time_per_iteration": 2.6338744163513184
    },
    {
      "auxiliary_loss_clip": 0.01239872,
      "auxiliary_loss_mlp": 0.01036019,
      "balance_loss_clip": 1.0635227,
      "balance_loss_mlp": 1.02451527,
      "epoch": 0.1447724403294655,
      "flos": 16690705752960.0,
      "grad_norm": 3.5168389642801734,
      "language_loss": 0.70764101,
      "learning_rate": 3.863517301805502e-06,
      "loss": 0.73039997,
      "num_input_tokens_seen": 25581380,
      "step": 1204,
      "time_per_iteration": 2.640465497970581
    },
    {
      "auxiliary_loss_clip": 0.01243495,
      "auxiliary_loss_mlp": 0.01042144,
      "balance_loss_clip": 1.06175518,
      "balance_loss_mlp": 1.03074121,
      "epoch": 0.14489268322010462,
      "flos": 20073741321600.0,
      "grad_norm": 2.388350212672608,
      "language_loss": 0.97039288,
      "learning_rate": 3.863234332802196e-06,
      "loss": 0.9932493,
      "num_input_tokens_seen": 25593585,
      "step": 1205,
      "time_per_iteration": 2.6534390449523926
    },
    {
      "auxiliary_loss_clip": 0.01233125,
      "auxiliary_loss_mlp": 0.01041353,
      "balance_loss_clip": 1.05701625,
      "balance_loss_mlp": 1.03089786,
      "epoch": 0.1450129261107437,
      "flos": 27125627955840.0,
      "grad_norm": 2.4823235165064834,
      "language_loss": 0.74419868,
      "learning_rate": 3.862951081147723e-06,
      "loss": 0.76694345,
      "num_input_tokens_seen": 25613750,
      "step": 1206,
      "time_per_iteration": 2.787179708480835
    },
    {
      "auxiliary_loss_clip": 0.01236681,
      "auxiliary_loss_mlp": 0.01038214,
      "balance_loss_clip": 1.06415594,
      "balance_loss_mlp": 1.02800989,
      "epoch": 0.1451331690013828,
      "flos": 25702344472320.0,
      "grad_norm": 2.4285952553043733,
      "language_loss": 0.78124148,
      "learning_rate": 3.862667546885053e-06,
      "loss": 0.80399042,
      "num_input_tokens_seen": 25632300,
      "step": 1207,
      "time_per_iteration": 2.6876718997955322
    },
    {
      "auxiliary_loss_clip": 0.01239455,
      "auxiliary_loss_mlp": 0.01038419,
      "balance_loss_clip": 1.05691648,
      "balance_loss_mlp": 1.02673697,
      "epoch": 0.14525341189202187,
      "flos": 25737393168000.0,
      "grad_norm": 1.9644221751451476,
      "language_loss": 0.73809826,
      "learning_rate": 3.8623837300571965e-06,
      "loss": 0.76087695,
      "num_input_tokens_seen": 25651285,
      "step": 1208,
      "time_per_iteration": 2.730414628982544
    },
    {
      "auxiliary_loss_clip": 0.01238236,
      "auxiliary_loss_mlp": 0.01037842,
      "balance_loss_clip": 1.06377864,
      "balance_loss_mlp": 1.02648115,
      "epoch": 0.14537365478266098,
      "flos": 23073898844160.0,
      "grad_norm": 3.010527856877145,
      "language_loss": 0.8412981,
      "learning_rate": 3.8620996307072085e-06,
      "loss": 0.86405885,
      "num_input_tokens_seen": 25671990,
      "step": 1209,
      "time_per_iteration": 2.688241958618164
    },
    {
      "auxiliary_loss_clip": 0.01240346,
      "auxiliary_loss_mlp": 0.01037824,
      "balance_loss_clip": 1.05441022,
      "balance_loss_mlp": 1.02614784,
      "epoch": 0.14549389767330007,
      "flos": 20595021448320.0,
      "grad_norm": 2.8139055812481675,
      "language_loss": 0.64963436,
      "learning_rate": 3.861815248878188e-06,
      "loss": 0.67241609,
      "num_input_tokens_seen": 25689475,
      "step": 1210,
      "time_per_iteration": 2.6935486793518066
    },
    {
      "auxiliary_loss_clip": 0.0123426,
      "auxiliary_loss_mlp": 0.01041604,
      "balance_loss_clip": 1.06068671,
      "balance_loss_mlp": 1.03113747,
      "epoch": 0.14561414056393915,
      "flos": 15121804533120.0,
      "grad_norm": 2.835190477464769,
      "language_loss": 0.79884982,
      "learning_rate": 3.861530584613274e-06,
      "loss": 0.82160848,
      "num_input_tokens_seen": 25707475,
      "step": 1211,
      "time_per_iteration": 2.6569979190826416
    },
    {
      "auxiliary_loss_clip": 0.0123981,
      "auxiliary_loss_mlp": 0.01209985,
      "balance_loss_clip": 1.06372523,
      "balance_loss_mlp": 1.00060701,
      "epoch": 0.14573438345457826,
      "flos": 19427493778560.0,
      "grad_norm": 2.4998258005131064,
      "language_loss": 0.82065731,
      "learning_rate": 3.86124563795565e-06,
      "loss": 0.84515524,
      "num_input_tokens_seen": 25726290,
      "step": 1212,
      "time_per_iteration": 2.662940740585327
    },
    {
      "auxiliary_loss_clip": 0.01235314,
      "auxiliary_loss_mlp": 0.01034697,
      "balance_loss_clip": 1.06485236,
      "balance_loss_mlp": 1.02396214,
      "epoch": 0.14585462634521734,
      "flos": 24828422572800.0,
      "grad_norm": 2.767592446885783,
      "language_loss": 0.70172745,
      "learning_rate": 3.860960408948543e-06,
      "loss": 0.72442758,
      "num_input_tokens_seen": 25748040,
      "step": 1213,
      "time_per_iteration": 2.68314528465271
    },
    {
      "auxiliary_loss_clip": 0.01226534,
      "auxiliary_loss_mlp": 0.01042314,
      "balance_loss_clip": 1.06097054,
      "balance_loss_mlp": 1.03201997,
      "epoch": 0.14597486923585642,
      "flos": 15448627405440.0,
      "grad_norm": 2.310714807202592,
      "language_loss": 0.89717388,
      "learning_rate": 3.860674897635222e-06,
      "loss": 0.91986239,
      "num_input_tokens_seen": 25764525,
      "step": 1214,
      "time_per_iteration": 2.595562696456909
    },
    {
      "auxiliary_loss_clip": 0.01235835,
      "auxiliary_loss_mlp": 0.01045559,
      "balance_loss_clip": 1.06315279,
      "balance_loss_mlp": 1.0341748,
      "epoch": 0.1460951121264955,
      "flos": 16655154266880.0,
      "grad_norm": 3.0585867821677315,
      "language_loss": 0.83750343,
      "learning_rate": 3.860389104058998e-06,
      "loss": 0.86031735,
      "num_input_tokens_seen": 25782755,
      "step": 1215,
      "time_per_iteration": 2.5866119861602783
    },
    {
      "auxiliary_loss_clip": 0.01235285,
      "auxiliary_loss_mlp": 0.01035582,
      "balance_loss_clip": 1.05866992,
      "balance_loss_mlp": 1.02487111,
      "epoch": 0.14621535501713462,
      "flos": 24863291700480.0,
      "grad_norm": 3.8807646861331793,
      "language_loss": 0.72401619,
      "learning_rate": 3.860103028263227e-06,
      "loss": 0.74672484,
      "num_input_tokens_seen": 25805860,
      "step": 1216,
      "time_per_iteration": 2.718752145767212
    },
    {
      "auxiliary_loss_clip": 0.01230208,
      "auxiliary_loss_mlp": 0.01032764,
      "balance_loss_clip": 1.04911888,
      "balance_loss_mlp": 1.02206516,
      "epoch": 0.1463355979077737,
      "flos": 25228000442880.0,
      "grad_norm": 2.411178288257319,
      "language_loss": 0.70550442,
      "learning_rate": 3.859816670291304e-06,
      "loss": 0.7281341,
      "num_input_tokens_seen": 25824955,
      "step": 1217,
      "time_per_iteration": 2.747896909713745
    },
    {
      "auxiliary_loss_clip": 0.01231328,
      "auxiliary_loss_mlp": 0.01029368,
      "balance_loss_clip": 1.05199182,
      "balance_loss_mlp": 1.01758409,
      "epoch": 0.14645584079841278,
      "flos": 22054143726720.0,
      "grad_norm": 3.281290774808008,
      "language_loss": 0.90304053,
      "learning_rate": 3.859530030186672e-06,
      "loss": 0.92564744,
      "num_input_tokens_seen": 25841965,
      "step": 1218,
      "time_per_iteration": 2.722980260848999
    },
    {
      "auxiliary_loss_clip": 0.01238847,
      "auxiliary_loss_mlp": 0.01036995,
      "balance_loss_clip": 1.06128192,
      "balance_loss_mlp": 1.02596855,
      "epoch": 0.1465760836890519,
      "flos": 23623870959360.0,
      "grad_norm": 2.4779566824064334,
      "language_loss": 0.82945555,
      "learning_rate": 3.859243107992813e-06,
      "loss": 0.85221398,
      "num_input_tokens_seen": 25860770,
      "step": 1219,
      "time_per_iteration": 2.7292797565460205
    },
    {
      "auxiliary_loss_clip": 0.01238797,
      "auxiliary_loss_mlp": 0.01043208,
      "balance_loss_clip": 1.053123,
      "balance_loss_mlp": 1.03132844,
      "epoch": 0.14669632657969098,
      "flos": 37407893356800.0,
      "grad_norm": 3.4754880998764435,
      "language_loss": 0.77970839,
      "learning_rate": 3.858955903753252e-06,
      "loss": 0.8025285,
      "num_input_tokens_seen": 25879410,
      "step": 1220,
      "time_per_iteration": 2.8185324668884277
    },
    {
      "auxiliary_loss_clip": 0.01233868,
      "auxiliary_loss_mlp": 0.01038384,
      "balance_loss_clip": 1.05826473,
      "balance_loss_mlp": 1.02861476,
      "epoch": 0.14681656947033006,
      "flos": 28365910623360.0,
      "grad_norm": 1.6781916312643281,
      "language_loss": 0.83616829,
      "learning_rate": 3.858668417511559e-06,
      "loss": 0.85889083,
      "num_input_tokens_seen": 25902160,
      "step": 1221,
      "time_per_iteration": 2.81652569770813
    },
    {
      "auxiliary_loss_clip": 0.01240301,
      "auxiliary_loss_mlp": 0.01034635,
      "balance_loss_clip": 1.06186771,
      "balance_loss_mlp": 1.02305365,
      "epoch": 0.14693681236096917,
      "flos": 18479488078080.0,
      "grad_norm": 2.9559267896149257,
      "language_loss": 0.76487851,
      "learning_rate": 3.8583806493113445e-06,
      "loss": 0.78762788,
      "num_input_tokens_seen": 25920505,
      "step": 1222,
      "time_per_iteration": 4.72874903678894
    },
    {
      "auxiliary_loss_clip": 0.01233418,
      "auxiliary_loss_mlp": 0.01043338,
      "balance_loss_clip": 1.06162083,
      "balance_loss_mlp": 1.03291905,
      "epoch": 0.14705705525160825,
      "flos": 20777806782720.0,
      "grad_norm": 2.8312862248001838,
      "language_loss": 0.82575822,
      "learning_rate": 3.858092599196263e-06,
      "loss": 0.84852576,
      "num_input_tokens_seen": 25938460,
      "step": 1223,
      "time_per_iteration": 2.820873260498047
    },
    {
      "auxiliary_loss_clip": 0.01236773,
      "auxiliary_loss_mlp": 0.0103178,
      "balance_loss_clip": 1.06182837,
      "balance_loss_mlp": 1.02102184,
      "epoch": 0.14717729814224734,
      "flos": 29932944336000.0,
      "grad_norm": 2.416444533222243,
      "language_loss": 0.82565033,
      "learning_rate": 3.857804267210012e-06,
      "loss": 0.84833586,
      "num_input_tokens_seen": 25957760,
      "step": 1224,
      "time_per_iteration": 2.879157781600952
    },
    {
      "auxiliary_loss_clip": 0.01223387,
      "auxiliary_loss_mlp": 0.01041795,
      "balance_loss_clip": 1.05180168,
      "balance_loss_mlp": 1.03087521,
      "epoch": 0.14729754103288642,
      "flos": 20047491457920.0,
      "grad_norm": 2.526515774593992,
      "language_loss": 0.8841989,
      "learning_rate": 3.857515653396331e-06,
      "loss": 0.9068507,
      "num_input_tokens_seen": 25974970,
      "step": 1225,
      "time_per_iteration": 3.862675666809082
    },
    {
      "auxiliary_loss_clip": 0.01240453,
      "auxiliary_loss_mlp": 0.01036912,
      "balance_loss_clip": 1.05598152,
      "balance_loss_mlp": 1.02677917,
      "epoch": 0.14741778392352553,
      "flos": 19281516906240.0,
      "grad_norm": 3.3482832327173138,
      "language_loss": 0.87389153,
      "learning_rate": 3.857226757799002e-06,
      "loss": 0.89666516,
      "num_input_tokens_seen": 25992525,
      "step": 1226,
      "time_per_iteration": 3.7988028526306152
    },
    {
      "auxiliary_loss_clip": 0.01235417,
      "auxiliary_loss_mlp": 0.01034892,
      "balance_loss_clip": 1.05822086,
      "balance_loss_mlp": 1.02419257,
      "epoch": 0.1475380268141646,
      "flos": 25411108999680.0,
      "grad_norm": 2.7703106454155346,
      "language_loss": 0.74451661,
      "learning_rate": 3.85693758046185e-06,
      "loss": 0.76721966,
      "num_input_tokens_seen": 26010815,
      "step": 1227,
      "time_per_iteration": 2.8158562183380127
    },
    {
      "auxiliary_loss_clip": 0.01236135,
      "auxiliary_loss_mlp": 0.01044513,
      "balance_loss_clip": 1.06642354,
      "balance_loss_mlp": 1.03402829,
      "epoch": 0.1476582697048037,
      "flos": 20847652778880.0,
      "grad_norm": 1.9397557238596554,
      "language_loss": 0.82934737,
      "learning_rate": 3.8566481214287435e-06,
      "loss": 0.85215378,
      "num_input_tokens_seen": 26028935,
      "step": 1228,
      "time_per_iteration": 2.825784206390381
    },
    {
      "auxiliary_loss_clip": 0.0122738,
      "auxiliary_loss_mlp": 0.01039527,
      "balance_loss_clip": 1.05216432,
      "balance_loss_mlp": 1.02845824,
      "epoch": 0.1477785125954428,
      "flos": 14028109269120.0,
      "grad_norm": 2.287387491905686,
      "language_loss": 0.90800542,
      "learning_rate": 3.8563583807435935e-06,
      "loss": 0.93067455,
      "num_input_tokens_seen": 26045080,
      "step": 1229,
      "time_per_iteration": 2.821336030960083
    },
    {
      "auxiliary_loss_clip": 0.0123908,
      "auxiliary_loss_mlp": 0.01209699,
      "balance_loss_clip": 1.06169176,
      "balance_loss_mlp": 1.00056911,
      "epoch": 0.1478987554860819,
      "flos": 20516699842560.0,
      "grad_norm": 2.1140784232322156,
      "language_loss": 0.77595037,
      "learning_rate": 3.856068358450353e-06,
      "loss": 0.80043817,
      "num_input_tokens_seen": 26065030,
      "step": 1230,
      "time_per_iteration": 2.867568016052246
    },
    {
      "auxiliary_loss_clip": 0.01234397,
      "auxiliary_loss_mlp": 0.01044344,
      "balance_loss_clip": 1.06390095,
      "balance_loss_mlp": 1.03332949,
      "epoch": 0.14801899837672097,
      "flos": 17857012360320.0,
      "grad_norm": 2.1052725291506267,
      "language_loss": 0.86090744,
      "learning_rate": 3.8557780545930186e-06,
      "loss": 0.88369489,
      "num_input_tokens_seen": 26083445,
      "step": 1231,
      "time_per_iteration": 2.88547420501709
    },
    {
      "auxiliary_loss_clip": 0.01236366,
      "auxiliary_loss_mlp": 0.01036627,
      "balance_loss_clip": 1.06138158,
      "balance_loss_mlp": 1.0264467,
      "epoch": 0.14813924126736006,
      "flos": 20881408584960.0,
      "grad_norm": 2.4084706970782275,
      "language_loss": 0.79101229,
      "learning_rate": 3.855487469215628e-06,
      "loss": 0.81374216,
      "num_input_tokens_seen": 26102375,
      "step": 1232,
      "time_per_iteration": 2.8272135257720947
    },
    {
      "auxiliary_loss_clip": 0.01238437,
      "auxiliary_loss_mlp": 0.01034069,
      "balance_loss_clip": 1.05840981,
      "balance_loss_mlp": 1.02364421,
      "epoch": 0.14825948415799917,
      "flos": 37414070496000.0,
      "grad_norm": 2.4788419270224904,
      "language_loss": 0.72566575,
      "learning_rate": 3.855196602362264e-06,
      "loss": 0.74839079,
      "num_input_tokens_seen": 26125295,
      "step": 1233,
      "time_per_iteration": 3.00359845161438
    },
    {
      "auxiliary_loss_clip": 0.01232793,
      "auxiliary_loss_mlp": 0.01032686,
      "balance_loss_clip": 1.05835474,
      "balance_loss_mlp": 1.02202845,
      "epoch": 0.14837972704863825,
      "flos": 22014641744640.0,
      "grad_norm": 2.9138631619506943,
      "language_loss": 0.94358712,
      "learning_rate": 3.854905454077051e-06,
      "loss": 0.96624196,
      "num_input_tokens_seen": 26142905,
      "step": 1234,
      "time_per_iteration": 2.805560827255249
    },
    {
      "auxiliary_loss_clip": 0.01237589,
      "auxiliary_loss_mlp": 0.01035904,
      "balance_loss_clip": 1.04990959,
      "balance_loss_mlp": 1.02520466,
      "epoch": 0.14849996993927733,
      "flos": 20996323171200.0,
      "grad_norm": 2.241164848262867,
      "language_loss": 0.88290894,
      "learning_rate": 3.854614024404155e-06,
      "loss": 0.90564388,
      "num_input_tokens_seen": 26161215,
      "step": 1235,
      "time_per_iteration": 2.9403188228607178
    },
    {
      "auxiliary_loss_clip": 0.01219918,
      "auxiliary_loss_mlp": 0.01031965,
      "balance_loss_clip": 1.05289495,
      "balance_loss_mlp": 1.02140284,
      "epoch": 0.14862021282991644,
      "flos": 20047994248320.0,
      "grad_norm": 2.1902806119543747,
      "language_loss": 0.89392531,
      "learning_rate": 3.8543223133877865e-06,
      "loss": 0.91644412,
      "num_input_tokens_seen": 26179810,
      "step": 1236,
      "time_per_iteration": 2.86954665184021
    },
    {
      "auxiliary_loss_clip": 0.01217409,
      "auxiliary_loss_mlp": 0.01041517,
      "balance_loss_clip": 1.05384052,
      "balance_loss_mlp": 1.02945268,
      "epoch": 0.14874045572055553,
      "flos": 22712027276160.0,
      "grad_norm": 6.337683607826455,
      "language_loss": 0.88386971,
      "learning_rate": 3.854030321072198e-06,
      "loss": 0.90645891,
      "num_input_tokens_seen": 26199715,
      "step": 1237,
      "time_per_iteration": 2.8672423362731934
    },
    {
      "auxiliary_loss_clip": 0.01241917,
      "auxiliary_loss_mlp": 0.01030354,
      "balance_loss_clip": 1.05643022,
      "balance_loss_mlp": 1.02004242,
      "epoch": 0.1488606986111946,
      "flos": 25411288567680.0,
      "grad_norm": 2.2027407587307457,
      "language_loss": 0.73501629,
      "learning_rate": 3.853738047501682e-06,
      "loss": 0.75773895,
      "num_input_tokens_seen": 26220275,
      "step": 1238,
      "time_per_iteration": 2.9445645809173584
    },
    {
      "auxiliary_loss_clip": 0.0123705,
      "auxiliary_loss_mlp": 0.01038344,
      "balance_loss_clip": 1.06376982,
      "balance_loss_mlp": 1.02711487,
      "epoch": 0.1489809415018337,
      "flos": 17018749687680.0,
      "grad_norm": 2.138199986568922,
      "language_loss": 0.7737844,
      "learning_rate": 3.85344549272058e-06,
      "loss": 0.79653835,
      "num_input_tokens_seen": 26238255,
      "step": 1239,
      "time_per_iteration": 2.819587469100952
    },
    {
      "auxiliary_loss_clip": 0.01230664,
      "auxiliary_loss_mlp": 0.01037051,
      "balance_loss_clip": 1.05913496,
      "balance_loss_mlp": 1.02591038,
      "epoch": 0.1491011843924728,
      "flos": 33659394860160.0,
      "grad_norm": 3.016229267812915,
      "language_loss": 0.8289212,
      "learning_rate": 3.853152656773269e-06,
      "loss": 0.85159838,
      "num_input_tokens_seen": 26259690,
      "step": 1240,
      "time_per_iteration": 2.960653781890869
    },
    {
      "auxiliary_loss_clip": 0.01232953,
      "auxiliary_loss_mlp": 0.01036043,
      "balance_loss_clip": 1.05895019,
      "balance_loss_mlp": 1.02534997,
      "epoch": 0.14922142728311188,
      "flos": 21179000764800.0,
      "grad_norm": 1.7691094357818524,
      "language_loss": 0.85020626,
      "learning_rate": 3.852859539704174e-06,
      "loss": 0.87289619,
      "num_input_tokens_seen": 26278990,
      "step": 1241,
      "time_per_iteration": 2.779057264328003
    },
    {
      "auxiliary_loss_clip": 0.01231096,
      "auxiliary_loss_mlp": 0.01037629,
      "balance_loss_clip": 1.04988718,
      "balance_loss_mlp": 1.02696013,
      "epoch": 0.14934167017375097,
      "flos": 29860548474240.0,
      "grad_norm": 2.1977647401821625,
      "language_loss": 0.76516926,
      "learning_rate": 3.85256614155776e-06,
      "loss": 0.78785652,
      "num_input_tokens_seen": 26299120,
      "step": 1242,
      "time_per_iteration": 2.969749689102173
    },
    {
      "auxiliary_loss_clip": 0.01229302,
      "auxiliary_loss_mlp": 0.01034523,
      "balance_loss_clip": 1.05559921,
      "balance_loss_mlp": 1.02416384,
      "epoch": 0.14946191306439008,
      "flos": 17019216564480.0,
      "grad_norm": 8.097925762341356,
      "language_loss": 0.7459743,
      "learning_rate": 3.852272462378535e-06,
      "loss": 0.7686125,
      "num_input_tokens_seen": 26316995,
      "step": 1243,
      "time_per_iteration": 2.7781686782836914
    },
    {
      "auxiliary_loss_clip": 0.01234774,
      "auxiliary_loss_mlp": 0.01046235,
      "balance_loss_clip": 1.0595566,
      "balance_loss_mlp": 1.03590524,
      "epoch": 0.14958215595502916,
      "flos": 15669047214720.0,
      "grad_norm": 3.7331984628420605,
      "language_loss": 0.78350157,
      "learning_rate": 3.85197850221105e-06,
      "loss": 0.80631167,
      "num_input_tokens_seen": 26333295,
      "step": 1244,
      "time_per_iteration": 2.72622013092041
    },
    {
      "auxiliary_loss_clip": 0.01232289,
      "auxiliary_loss_mlp": 0.01038985,
      "balance_loss_clip": 1.06232262,
      "balance_loss_mlp": 1.02861965,
      "epoch": 0.14970239884566824,
      "flos": 33108560818560.0,
      "grad_norm": 2.111043837221556,
      "language_loss": 0.76032537,
      "learning_rate": 3.851684261099899e-06,
      "loss": 0.78303814,
      "num_input_tokens_seen": 26355035,
      "step": 1245,
      "time_per_iteration": 2.8597803115844727
    },
    {
      "auxiliary_loss_clip": 0.01228633,
      "auxiliary_loss_mlp": 0.01041157,
      "balance_loss_clip": 1.05546355,
      "balance_loss_mlp": 1.02934337,
      "epoch": 0.14982264173630733,
      "flos": 17821245392640.0,
      "grad_norm": 1.9310594756387989,
      "language_loss": 0.86784023,
      "learning_rate": 3.851389739089718e-06,
      "loss": 0.8905381,
      "num_input_tokens_seen": 26371655,
      "step": 1246,
      "time_per_iteration": 2.7819485664367676
    },
    {
      "auxiliary_loss_clip": 0.01238234,
      "auxiliary_loss_mlp": 0.01039072,
      "balance_loss_clip": 1.0654732,
      "balance_loss_mlp": 1.02818251,
      "epoch": 0.14994288462694644,
      "flos": 32409559175040.0,
      "grad_norm": 2.1403861841614837,
      "language_loss": 0.80747169,
      "learning_rate": 3.851094936225186e-06,
      "loss": 0.83024472,
      "num_input_tokens_seen": 26392540,
      "step": 1247,
      "time_per_iteration": 2.846724271774292
    },
    {
      "auxiliary_loss_clip": 0.01231038,
      "auxiliary_loss_mlp": 0.01032617,
      "balance_loss_clip": 1.06083894,
      "balance_loss_mlp": 1.02200174,
      "epoch": 0.15006312751758552,
      "flos": 31794661226880.0,
      "grad_norm": 1.471990709968124,
      "language_loss": 0.76634061,
      "learning_rate": 3.850799852551024e-06,
      "loss": 0.78897715,
      "num_input_tokens_seen": 26414960,
      "step": 1248,
      "time_per_iteration": 3.7900521755218506
    },
    {
      "auxiliary_loss_clip": 0.01223729,
      "auxiliary_loss_mlp": 0.01039761,
      "balance_loss_clip": 1.05640864,
      "balance_loss_mlp": 1.028579,
      "epoch": 0.1501833704082246,
      "flos": 16618022582400.0,
      "grad_norm": 2.374665017124229,
      "language_loss": 0.86151624,
      "learning_rate": 3.850504488111995e-06,
      "loss": 0.8841511,
      "num_input_tokens_seen": 26431635,
      "step": 1249,
      "time_per_iteration": 2.797713041305542
    },
    {
      "auxiliary_loss_clip": 0.0122414,
      "auxiliary_loss_mlp": 0.01029146,
      "balance_loss_clip": 1.05403399,
      "balance_loss_mlp": 1.01883399,
      "epoch": 0.15030361329886371,
      "flos": 23471178243840.0,
      "grad_norm": 1.9492512688227595,
      "language_loss": 0.83035362,
      "learning_rate": 3.850208842952907e-06,
      "loss": 0.85288656,
      "num_input_tokens_seen": 26450440,
      "step": 1250,
      "time_per_iteration": 2.8361754417419434
    },
    {
      "auxiliary_loss_clip": 0.01245157,
      "auxiliary_loss_mlp": 0.0104387,
      "balance_loss_clip": 1.05398202,
      "balance_loss_mlp": 1.03327239,
      "epoch": 0.1504238561895028,
      "flos": 25629409906560.0,
      "grad_norm": 5.061984236064091,
      "language_loss": 0.79312468,
      "learning_rate": 3.849912917118608e-06,
      "loss": 0.81601501,
      "num_input_tokens_seen": 26471480,
      "step": 1251,
      "time_per_iteration": 2.8752431869506836
    },
    {
      "auxiliary_loss_clip": 0.01144411,
      "auxiliary_loss_mlp": 0.01007067,
      "balance_loss_clip": 1.03416109,
      "balance_loss_mlp": 1.00387216,
      "epoch": 0.15054409908014188,
      "flos": 52095146129280.0,
      "grad_norm": 0.8743791956826895,
      "language_loss": 0.59200108,
      "learning_rate": 3.849616710653992e-06,
      "loss": 0.61351585,
      "num_input_tokens_seen": 26532950,
      "step": 1252,
      "time_per_iteration": 4.219804525375366
    },
    {
      "auxiliary_loss_clip": 0.01231248,
      "auxiliary_loss_mlp": 0.01038132,
      "balance_loss_clip": 1.06011307,
      "balance_loss_mlp": 1.0269978,
      "epoch": 0.150664341970781,
      "flos": 18880251096960.0,
      "grad_norm": 1.6036912504400034,
      "language_loss": 0.75284469,
      "learning_rate": 3.84932022360399e-06,
      "loss": 0.77553856,
      "num_input_tokens_seen": 26551615,
      "step": 1253,
      "time_per_iteration": 2.9161250591278076
    },
    {
      "auxiliary_loss_clip": 0.01233688,
      "auxiliary_loss_mlp": 0.01044964,
      "balance_loss_clip": 1.06124473,
      "balance_loss_mlp": 1.03322768,
      "epoch": 0.15078458486142007,
      "flos": 22163240309760.0,
      "grad_norm": 2.8517861323304516,
      "language_loss": 0.84353268,
      "learning_rate": 3.849023456013581e-06,
      "loss": 0.86631924,
      "num_input_tokens_seen": 26569175,
      "step": 1254,
      "time_per_iteration": 2.799237012863159
    },
    {
      "auxiliary_loss_clip": 0.01239035,
      "auxiliary_loss_mlp": 0.0104031,
      "balance_loss_clip": 1.06165636,
      "balance_loss_mlp": 1.02881241,
      "epoch": 0.15090482775205916,
      "flos": 26651894457600.0,
      "grad_norm": 3.781774173820574,
      "language_loss": 0.62537402,
      "learning_rate": 3.848726407927784e-06,
      "loss": 0.64816749,
      "num_input_tokens_seen": 26589560,
      "step": 1255,
      "time_per_iteration": 2.7710766792297363
    },
    {
      "auxiliary_loss_clip": 0.01236773,
      "auxiliary_loss_mlp": 0.01037755,
      "balance_loss_clip": 1.0614239,
      "balance_loss_mlp": 1.02743745,
      "epoch": 0.15102507064269824,
      "flos": 21798998444160.0,
      "grad_norm": 5.770067347916118,
      "language_loss": 0.86555111,
      "learning_rate": 3.84842907939166e-06,
      "loss": 0.88829643,
      "num_input_tokens_seen": 26608785,
      "step": 1256,
      "time_per_iteration": 2.7633347511291504
    },
    {
      "auxiliary_loss_clip": 0.01228627,
      "auxiliary_loss_mlp": 0.01038098,
      "balance_loss_clip": 1.05542624,
      "balance_loss_mlp": 1.02773261,
      "epoch": 0.15114531353333735,
      "flos": 22820908377600.0,
      "grad_norm": 2.3422997318922767,
      "language_loss": 0.71116149,
      "learning_rate": 3.8481314704503146e-06,
      "loss": 0.73382878,
      "num_input_tokens_seen": 26628615,
      "step": 1257,
      "time_per_iteration": 2.837172508239746
    },
    {
      "auxiliary_loss_clip": 0.01233786,
      "auxiliary_loss_mlp": 0.01037211,
      "balance_loss_clip": 1.06366134,
      "balance_loss_mlp": 1.02712584,
      "epoch": 0.15126555642397643,
      "flos": 19682674974720.0,
      "grad_norm": 3.073922389245649,
      "language_loss": 0.87720227,
      "learning_rate": 3.847833581148895e-06,
      "loss": 0.89991224,
      "num_input_tokens_seen": 26647525,
      "step": 1258,
      "time_per_iteration": 2.7822790145874023
    },
    {
      "auxiliary_loss_clip": 0.01230434,
      "auxiliary_loss_mlp": 0.0103178,
      "balance_loss_clip": 1.05921125,
      "balance_loss_mlp": 1.02085435,
      "epoch": 0.15138579931461552,
      "flos": 28726022424960.0,
      "grad_norm": 6.330002361695914,
      "language_loss": 0.81121463,
      "learning_rate": 3.84753541153259e-06,
      "loss": 0.83383679,
      "num_input_tokens_seen": 26667095,
      "step": 1259,
      "time_per_iteration": 2.824324369430542
    },
    {
      "auxiliary_loss_clip": 0.01234374,
      "auxiliary_loss_mlp": 0.01037308,
      "balance_loss_clip": 1.06269324,
      "balance_loss_mlp": 1.02699065,
      "epoch": 0.15150604220525463,
      "flos": 22127006465280.0,
      "grad_norm": 1.939457179899818,
      "language_loss": 0.83520389,
      "learning_rate": 3.847236961646633e-06,
      "loss": 0.85792065,
      "num_input_tokens_seen": 26686075,
      "step": 1260,
      "time_per_iteration": 2.7797322273254395
    },
    {
      "auxiliary_loss_clip": 0.01227988,
      "auxiliary_loss_mlp": 0.01038035,
      "balance_loss_clip": 1.05697894,
      "balance_loss_mlp": 1.0267638,
      "epoch": 0.1516262850958937,
      "flos": 12968708515200.0,
      "grad_norm": 2.6907633776156805,
      "language_loss": 0.77842921,
      "learning_rate": 3.846938231536296e-06,
      "loss": 0.80108941,
      "num_input_tokens_seen": 26701695,
      "step": 1261,
      "time_per_iteration": 2.787019968032837
    },
    {
      "auxiliary_loss_clip": 0.0123854,
      "auxiliary_loss_mlp": 0.01032432,
      "balance_loss_clip": 1.06356645,
      "balance_loss_mlp": 1.02185822,
      "epoch": 0.1517465279865328,
      "flos": 21797130936960.0,
      "grad_norm": 2.320688185721974,
      "language_loss": 0.80977607,
      "learning_rate": 3.8466392212468995e-06,
      "loss": 0.8324858,
      "num_input_tokens_seen": 26721885,
      "step": 1262,
      "time_per_iteration": 2.939328670501709
    },
    {
      "auxiliary_loss_clip": 0.01150133,
      "auxiliary_loss_mlp": 0.01008149,
      "balance_loss_clip": 1.02584791,
      "balance_loss_mlp": 1.0050261,
      "epoch": 0.15186677087717187,
      "flos": 58174569901440.0,
      "grad_norm": 0.8217892138739745,
      "language_loss": 0.61911923,
      "learning_rate": 3.8463399308238e-06,
      "loss": 0.64070201,
      "num_input_tokens_seen": 26780990,
      "step": 1263,
      "time_per_iteration": 3.249678373336792
    },
    {
      "auxiliary_loss_clip": 0.01233318,
      "auxiliary_loss_mlp": 0.01040527,
      "balance_loss_clip": 1.06271148,
      "balance_loss_mlp": 1.02791429,
      "epoch": 0.15198701376781099,
      "flos": 32669696448000.0,
      "grad_norm": 1.998022826997656,
      "language_loss": 0.64280438,
      "learning_rate": 3.846040360312402e-06,
      "loss": 0.66554284,
      "num_input_tokens_seen": 26804250,
      "step": 1264,
      "time_per_iteration": 2.870706081390381
    },
    {
      "auxiliary_loss_clip": 0.01230046,
      "auxiliary_loss_mlp": 0.0104031,
      "balance_loss_clip": 1.05975819,
      "balance_loss_mlp": 1.02873492,
      "epoch": 0.15210725665845007,
      "flos": 28402575431040.0,
      "grad_norm": 2.600207618833039,
      "language_loss": 0.81289536,
      "learning_rate": 3.8457405097581485e-06,
      "loss": 0.83559889,
      "num_input_tokens_seen": 26823240,
      "step": 1265,
      "time_per_iteration": 2.7812368869781494
    },
    {
      "auxiliary_loss_clip": 0.01239763,
      "auxiliary_loss_mlp": 0.01037159,
      "balance_loss_clip": 1.05258489,
      "balance_loss_mlp": 1.02572107,
      "epoch": 0.15222749954908915,
      "flos": 19938179393280.0,
      "grad_norm": 1.9888937513434122,
      "language_loss": 0.77754176,
      "learning_rate": 3.8454403792065275e-06,
      "loss": 0.80031097,
      "num_input_tokens_seen": 26842060,
      "step": 1266,
      "time_per_iteration": 2.8126232624053955
    },
    {
      "auxiliary_loss_clip": 0.01223996,
      "auxiliary_loss_mlp": 0.01049801,
      "balance_loss_clip": 1.05343032,
      "balance_loss_mlp": 1.03853559,
      "epoch": 0.15234774243972826,
      "flos": 21324223451520.0,
      "grad_norm": 2.192486152544506,
      "language_loss": 0.85553974,
      "learning_rate": 3.845139968703068e-06,
      "loss": 0.87827766,
      "num_input_tokens_seen": 26859580,
      "step": 1267,
      "time_per_iteration": 2.8264970779418945
    },
    {
      "auxiliary_loss_clip": 0.01237691,
      "auxiliary_loss_mlp": 0.01043473,
      "balance_loss_clip": 1.05456948,
      "balance_loss_mlp": 1.03144479,
      "epoch": 0.15246798533036734,
      "flos": 25957812977280.0,
      "grad_norm": 1.9869843220392485,
      "language_loss": 0.82698625,
      "learning_rate": 3.844839278293342e-06,
      "loss": 0.84979784,
      "num_input_tokens_seen": 26880430,
      "step": 1268,
      "time_per_iteration": 2.8245327472686768
    },
    {
      "auxiliary_loss_clip": 0.01236174,
      "auxiliary_loss_mlp": 0.01038859,
      "balance_loss_clip": 1.06511867,
      "balance_loss_mlp": 1.0271225,
      "epoch": 0.15258822822100643,
      "flos": 25811907932160.0,
      "grad_norm": 3.6062658919575954,
      "language_loss": 0.76709414,
      "learning_rate": 3.8445383080229654e-06,
      "loss": 0.78984445,
      "num_input_tokens_seen": 26896445,
      "step": 1269,
      "time_per_iteration": 2.7279584407806396
    },
    {
      "auxiliary_loss_clip": 0.01227774,
      "auxiliary_loss_mlp": 0.01038412,
      "balance_loss_clip": 1.05533242,
      "balance_loss_mlp": 1.02663422,
      "epoch": 0.1527084711116455,
      "flos": 25265455349760.0,
      "grad_norm": 8.358456212173365,
      "language_loss": 0.73281807,
      "learning_rate": 3.844237057937593e-06,
      "loss": 0.75547993,
      "num_input_tokens_seen": 26915450,
      "step": 1270,
      "time_per_iteration": 2.7815895080566406
    },
    {
      "auxiliary_loss_clip": 0.01240668,
      "auxiliary_loss_mlp": 0.01034179,
      "balance_loss_clip": 1.06084824,
      "balance_loss_mlp": 1.02255583,
      "epoch": 0.15282871400228462,
      "flos": 29240227572480.0,
      "grad_norm": 2.5152524357909387,
      "language_loss": 0.77936733,
      "learning_rate": 3.843935528082926e-06,
      "loss": 0.8021158,
      "num_input_tokens_seen": 26936475,
      "step": 1271,
      "time_per_iteration": 2.8722615242004395
    },
    {
      "auxiliary_loss_clip": 0.01237241,
      "auxiliary_loss_mlp": 0.0103341,
      "balance_loss_clip": 1.06281066,
      "balance_loss_mlp": 1.02250862,
      "epoch": 0.1529489568929237,
      "flos": 20882952869760.0,
      "grad_norm": 2.329787469114607,
      "language_loss": 0.85234737,
      "learning_rate": 3.843633718504704e-06,
      "loss": 0.87505388,
      "num_input_tokens_seen": 26954920,
      "step": 1272,
      "time_per_iteration": 2.747101306915283
    },
    {
      "auxiliary_loss_clip": 0.01237867,
      "auxiliary_loss_mlp": 0.01039432,
      "balance_loss_clip": 1.05815053,
      "balance_loss_mlp": 1.02830398,
      "epoch": 0.1530691997835628,
      "flos": 20083833043200.0,
      "grad_norm": 8.318512272874425,
      "language_loss": 0.9042685,
      "learning_rate": 3.843331629248715e-06,
      "loss": 0.92704141,
      "num_input_tokens_seen": 26972520,
      "step": 1273,
      "time_per_iteration": 2.797433376312256
    },
    {
      "auxiliary_loss_clip": 0.01236343,
      "auxiliary_loss_mlp": 0.01036461,
      "balance_loss_clip": 1.064762,
      "balance_loss_mlp": 1.02558327,
      "epoch": 0.1531894426742019,
      "flos": 28759814144640.0,
      "grad_norm": 2.446650731019594,
      "language_loss": 0.77053964,
      "learning_rate": 3.843029260360782e-06,
      "loss": 0.79326767,
      "num_input_tokens_seen": 26990890,
      "step": 1274,
      "time_per_iteration": 2.8075220584869385
    },
    {
      "auxiliary_loss_clip": 0.0123484,
      "auxiliary_loss_mlp": 0.01043006,
      "balance_loss_clip": 1.06226993,
      "balance_loss_mlp": 1.03244972,
      "epoch": 0.15330968556484098,
      "flos": 22236282616320.0,
      "grad_norm": 2.0877994864840845,
      "language_loss": 0.79047114,
      "learning_rate": 3.8427266118867755e-06,
      "loss": 0.81324959,
      "num_input_tokens_seen": 27010640,
      "step": 1275,
      "time_per_iteration": 3.7768049240112305
    },
    {
      "auxiliary_loss_clip": 0.01235296,
      "auxiliary_loss_mlp": 0.01034362,
      "balance_loss_clip": 1.06027424,
      "balance_loss_mlp": 1.02292991,
      "epoch": 0.15342992845548006,
      "flos": 27527504296320.0,
      "grad_norm": 2.9946550368624987,
      "language_loss": 0.82838929,
      "learning_rate": 3.842423683872608e-06,
      "loss": 0.8510859,
      "num_input_tokens_seen": 27031215,
      "step": 1276,
      "time_per_iteration": 2.8391377925872803
    },
    {
      "auxiliary_loss_clip": 0.01235633,
      "auxiliary_loss_mlp": 0.01039337,
      "balance_loss_clip": 1.06200194,
      "balance_loss_mlp": 1.02833366,
      "epoch": 0.15355017134611917,
      "flos": 19609596754560.0,
      "grad_norm": 2.867496016020226,
      "language_loss": 0.78401774,
      "learning_rate": 3.842120476364232e-06,
      "loss": 0.80676746,
      "num_input_tokens_seen": 27049665,
      "step": 1277,
      "time_per_iteration": 2.7494468688964844
    },
    {
      "auxiliary_loss_clip": 0.01241911,
      "auxiliary_loss_mlp": 0.01036629,
      "balance_loss_clip": 1.06123281,
      "balance_loss_mlp": 1.02551842,
      "epoch": 0.15367041423675826,
      "flos": 18478590238080.0,
      "grad_norm": 2.2864226628969884,
      "language_loss": 0.84012997,
      "learning_rate": 3.841816989407644e-06,
      "loss": 0.86291534,
      "num_input_tokens_seen": 27065155,
      "step": 1278,
      "time_per_iteration": 2.8204567432403564
    },
    {
      "auxiliary_loss_clip": 0.0123416,
      "auxiliary_loss_mlp": 0.0104198,
      "balance_loss_clip": 1.05757272,
      "balance_loss_mlp": 1.03082156,
      "epoch": 0.15379065712739734,
      "flos": 41427662342400.0,
      "grad_norm": 2.888436803022329,
      "language_loss": 0.77207899,
      "learning_rate": 3.841513223048884e-06,
      "loss": 0.7948404,
      "num_input_tokens_seen": 27085840,
      "step": 1279,
      "time_per_iteration": 4.820150375366211
    },
    {
      "auxiliary_loss_clip": 0.01231711,
      "auxiliary_loss_mlp": 0.01038486,
      "balance_loss_clip": 1.05339432,
      "balance_loss_mlp": 1.02736938,
      "epoch": 0.15391090001803642,
      "flos": 22054215553920.0,
      "grad_norm": 3.426552144144738,
      "language_loss": 0.79120314,
      "learning_rate": 3.841209177334031e-06,
      "loss": 0.813905,
      "num_input_tokens_seen": 27104200,
      "step": 1280,
      "time_per_iteration": 2.8773295879364014
    },
    {
      "auxiliary_loss_clip": 0.01232352,
      "auxiliary_loss_mlp": 0.0103805,
      "balance_loss_clip": 1.06192076,
      "balance_loss_mlp": 1.0270412,
      "epoch": 0.15403114290867553,
      "flos": 15450351258240.0,
      "grad_norm": 2.5108616749445054,
      "language_loss": 0.74849927,
      "learning_rate": 3.84090485230921e-06,
      "loss": 0.7712034,
      "num_input_tokens_seen": 27122440,
      "step": 1281,
      "time_per_iteration": 2.8450043201446533
    },
    {
      "auxiliary_loss_clip": 0.01234097,
      "auxiliary_loss_mlp": 0.01035631,
      "balance_loss_clip": 1.06311846,
      "balance_loss_mlp": 1.02443075,
      "epoch": 0.15415138579931462,
      "flos": 17929156826880.0,
      "grad_norm": 2.757650318583595,
      "language_loss": 0.76517463,
      "learning_rate": 3.840600248020588e-06,
      "loss": 0.78787184,
      "num_input_tokens_seen": 27139380,
      "step": 1282,
      "time_per_iteration": 2.76802396774292
    },
    {
      "auxiliary_loss_clip": 0.01242125,
      "auxiliary_loss_mlp": 0.01048621,
      "balance_loss_clip": 1.05818701,
      "balance_loss_mlp": 1.03696203,
      "epoch": 0.1542716286899537,
      "flos": 11429325296640.0,
      "grad_norm": 2.220868387152902,
      "language_loss": 0.79934287,
      "learning_rate": 3.840295364514371e-06,
      "loss": 0.82225037,
      "num_input_tokens_seen": 27156760,
      "step": 1283,
      "time_per_iteration": 2.7792813777923584
    },
    {
      "auxiliary_loss_clip": 0.01236498,
      "auxiliary_loss_mlp": 0.01035182,
      "balance_loss_clip": 1.05889261,
      "balance_loss_mlp": 1.02394664,
      "epoch": 0.1543918715805928,
      "flos": 17420338719360.0,
      "grad_norm": 2.9369728545356355,
      "language_loss": 0.78972363,
      "learning_rate": 3.83999020183681e-06,
      "loss": 0.81244051,
      "num_input_tokens_seen": 27175455,
      "step": 1284,
      "time_per_iteration": 2.8449580669403076
    },
    {
      "auxiliary_loss_clip": 0.01234445,
      "auxiliary_loss_mlp": 0.01043655,
      "balance_loss_clip": 1.05432129,
      "balance_loss_mlp": 1.03265202,
      "epoch": 0.1545121144712319,
      "flos": 17786376264960.0,
      "grad_norm": 2.0308880244486955,
      "language_loss": 0.78523457,
      "learning_rate": 3.839684760034199e-06,
      "loss": 0.80801558,
      "num_input_tokens_seen": 27193660,
      "step": 1285,
      "time_per_iteration": 2.8376219272613525
    },
    {
      "auxiliary_loss_clip": 0.01229467,
      "auxiliary_loss_mlp": 0.0103999,
      "balance_loss_clip": 1.05490923,
      "balance_loss_mlp": 1.02814639,
      "epoch": 0.15463235736187098,
      "flos": 28220185146240.0,
      "grad_norm": 3.51630868326249,
      "language_loss": 0.65044898,
      "learning_rate": 3.8393790391528716e-06,
      "loss": 0.67314351,
      "num_input_tokens_seen": 27214355,
      "step": 1286,
      "time_per_iteration": 2.815380334854126
    },
    {
      "auxiliary_loss_clip": 0.01233539,
      "auxiliary_loss_mlp": 0.01041617,
      "balance_loss_clip": 1.05684018,
      "balance_loss_mlp": 1.03028011,
      "epoch": 0.15475260025251006,
      "flos": 22856890826880.0,
      "grad_norm": 2.0472308143916225,
      "language_loss": 0.89077747,
      "learning_rate": 3.8390730392392075e-06,
      "loss": 0.91352904,
      "num_input_tokens_seen": 27234335,
      "step": 1287,
      "time_per_iteration": 2.7263312339782715
    },
    {
      "auxiliary_loss_clip": 0.01235789,
      "auxiliary_loss_mlp": 0.01039252,
      "balance_loss_clip": 1.0650481,
      "balance_loss_mlp": 1.02842212,
      "epoch": 0.15487284314314917,
      "flos": 17602872658560.0,
      "grad_norm": 2.417215208256116,
      "language_loss": 0.79296172,
      "learning_rate": 3.838766760339626e-06,
      "loss": 0.81571209,
      "num_input_tokens_seen": 27252860,
      "step": 1288,
      "time_per_iteration": 2.671912670135498
    },
    {
      "auxiliary_loss_clip": 0.01217223,
      "auxiliary_loss_mlp": 0.01038658,
      "balance_loss_clip": 1.05303717,
      "balance_loss_mlp": 1.02664804,
      "epoch": 0.15499308603378825,
      "flos": 20082037363200.0,
      "grad_norm": 4.026147151580156,
      "language_loss": 0.79445112,
      "learning_rate": 3.838460202500587e-06,
      "loss": 0.81700993,
      "num_input_tokens_seen": 27268650,
      "step": 1289,
      "time_per_iteration": 2.693455696105957
    },
    {
      "auxiliary_loss_clip": 0.01232784,
      "auxiliary_loss_mlp": 0.01035202,
      "balance_loss_clip": 1.06093514,
      "balance_loss_mlp": 1.02291107,
      "epoch": 0.15511332892442733,
      "flos": 15918051271680.0,
      "grad_norm": 2.2334135954371486,
      "language_loss": 0.73689705,
      "learning_rate": 3.838153365768599e-06,
      "loss": 0.75957692,
      "num_input_tokens_seen": 27285160,
      "step": 1290,
      "time_per_iteration": 2.7944140434265137
    },
    {
      "auxiliary_loss_clip": 0.01236353,
      "auxiliary_loss_mlp": 0.01047612,
      "balance_loss_clip": 1.06283164,
      "balance_loss_mlp": 1.03591156,
      "epoch": 0.15523357181506645,
      "flos": 41282475569280.0,
      "grad_norm": 2.247356161886768,
      "language_loss": 0.75972819,
      "learning_rate": 3.837846250190206e-06,
      "loss": 0.78256786,
      "num_input_tokens_seen": 27308025,
      "step": 1291,
      "time_per_iteration": 3.0404231548309326
    },
    {
      "auxiliary_loss_clip": 0.01231138,
      "auxiliary_loss_mlp": 0.01210504,
      "balance_loss_clip": 1.054317,
      "balance_loss_mlp": 1.00079155,
      "epoch": 0.15535381470570553,
      "flos": 18478769806080.0,
      "grad_norm": 2.031005344572813,
      "language_loss": 0.77393055,
      "learning_rate": 3.837538855811998e-06,
      "loss": 0.798347,
      "num_input_tokens_seen": 27326200,
      "step": 1292,
      "time_per_iteration": 2.8026270866394043
    },
    {
      "auxiliary_loss_clip": 0.01244466,
      "auxiliary_loss_mlp": 0.0103822,
      "balance_loss_clip": 1.06351709,
      "balance_loss_mlp": 1.02719259,
      "epoch": 0.1554740575963446,
      "flos": 13918150759680.0,
      "grad_norm": 2.2005576208200464,
      "language_loss": 0.70372969,
      "learning_rate": 3.837231182680606e-06,
      "loss": 0.72655654,
      "num_input_tokens_seen": 27344165,
      "step": 1293,
      "time_per_iteration": 2.7799124717712402
    },
    {
      "auxiliary_loss_clip": 0.01239738,
      "auxiliary_loss_mlp": 0.01040436,
      "balance_loss_clip": 1.06159544,
      "balance_loss_mlp": 1.02896798,
      "epoch": 0.1555943004869837,
      "flos": 20847078161280.0,
      "grad_norm": 1.7858796671100838,
      "language_loss": 0.76218224,
      "learning_rate": 3.836923230842706e-06,
      "loss": 0.78498399,
      "num_input_tokens_seen": 27363280,
      "step": 1294,
      "time_per_iteration": 2.836747169494629
    },
    {
      "auxiliary_loss_clip": 0.01239727,
      "auxiliary_loss_mlp": 0.01041089,
      "balance_loss_clip": 1.05224061,
      "balance_loss_mlp": 1.02926385,
      "epoch": 0.1557145433776228,
      "flos": 22085888371200.0,
      "grad_norm": 3.5439744335525383,
      "language_loss": 0.80653882,
      "learning_rate": 3.836615000345011e-06,
      "loss": 0.82934695,
      "num_input_tokens_seen": 27381460,
      "step": 1295,
      "time_per_iteration": 2.880133628845215
    },
    {
      "auxiliary_loss_clip": 0.01232012,
      "auxiliary_loss_mlp": 0.01034459,
      "balance_loss_clip": 1.06244504,
      "balance_loss_mlp": 1.02408767,
      "epoch": 0.1558347862682619,
      "flos": 19791987039360.0,
      "grad_norm": 6.187830301935622,
      "language_loss": 0.77677858,
      "learning_rate": 3.836306491234282e-06,
      "loss": 0.79944324,
      "num_input_tokens_seen": 27399310,
      "step": 1296,
      "time_per_iteration": 2.7850589752197266
    },
    {
      "auxiliary_loss_clip": 0.01232095,
      "auxiliary_loss_mlp": 0.01034918,
      "balance_loss_clip": 1.06318235,
      "balance_loss_mlp": 1.02456474,
      "epoch": 0.15595502915890097,
      "flos": 17237086508160.0,
      "grad_norm": 45.15282493944296,
      "language_loss": 0.75640678,
      "learning_rate": 3.835997703557317e-06,
      "loss": 0.77907687,
      "num_input_tokens_seen": 27416050,
      "step": 1297,
      "time_per_iteration": 2.7974929809570312
    },
    {
      "auxiliary_loss_clip": 0.01236886,
      "auxiliary_loss_mlp": 0.01040938,
      "balance_loss_clip": 1.05091023,
      "balance_loss_mlp": 1.02988744,
      "epoch": 0.15607527204954008,
      "flos": 19719519350400.0,
      "grad_norm": 1.8372238774800245,
      "language_loss": 0.80172002,
      "learning_rate": 3.83568863736096e-06,
      "loss": 0.8244983,
      "num_input_tokens_seen": 27434920,
      "step": 1298,
      "time_per_iteration": 2.8865461349487305
    },
    {
      "auxiliary_loss_clip": 0.0123845,
      "auxiliary_loss_mlp": 0.01036772,
      "balance_loss_clip": 1.05548596,
      "balance_loss_mlp": 1.02612066,
      "epoch": 0.15619551494017916,
      "flos": 18515650095360.0,
      "grad_norm": 2.4063991694722153,
      "language_loss": 0.89164859,
      "learning_rate": 3.8353792926920975e-06,
      "loss": 0.91440082,
      "num_input_tokens_seen": 27453570,
      "step": 1299,
      "time_per_iteration": 2.783961296081543
    },
    {
      "auxiliary_loss_clip": 0.0124293,
      "auxiliary_loss_mlp": 0.01043413,
      "balance_loss_clip": 1.06540608,
      "balance_loss_mlp": 1.03181434,
      "epoch": 0.15631575783081825,
      "flos": 19902125116800.0,
      "grad_norm": 2.8141052482525586,
      "language_loss": 0.82440752,
      "learning_rate": 3.835069669597655e-06,
      "loss": 0.84727097,
      "num_input_tokens_seen": 27471960,
      "step": 1300,
      "time_per_iteration": 2.790207862854004
    },
    {
      "auxiliary_loss_clip": 0.01238045,
      "auxiliary_loss_mlp": 0.0121089,
      "balance_loss_clip": 1.06099665,
      "balance_loss_mlp": 1.00062442,
      "epoch": 0.15643600072145733,
      "flos": 20777663128320.0,
      "grad_norm": 6.101904631305568,
      "language_loss": 0.80273056,
      "learning_rate": 3.834759768124603e-06,
      "loss": 0.82721996,
      "num_input_tokens_seen": 27490835,
      "step": 1301,
      "time_per_iteration": 3.7259931564331055
    },
    {
      "auxiliary_loss_clip": 0.01240857,
      "auxiliary_loss_mlp": 0.01035927,
      "balance_loss_clip": 1.06056786,
      "balance_loss_mlp": 1.02470982,
      "epoch": 0.15655624361209644,
      "flos": 18546389159040.0,
      "grad_norm": 3.284558342207731,
      "language_loss": 0.76472265,
      "learning_rate": 3.834449588319953e-06,
      "loss": 0.78749049,
      "num_input_tokens_seen": 27508870,
      "step": 1302,
      "time_per_iteration": 3.8427958488464355
    },
    {
      "auxiliary_loss_clip": 0.01235866,
      "auxiliary_loss_mlp": 0.01044266,
      "balance_loss_clip": 1.0659349,
      "balance_loss_mlp": 1.03320336,
      "epoch": 0.15667648650273552,
      "flos": 25229544727680.0,
      "grad_norm": 5.991599314056963,
      "language_loss": 0.85075027,
      "learning_rate": 3.834139130230758e-06,
      "loss": 0.87355161,
      "num_input_tokens_seen": 27528175,
      "step": 1303,
      "time_per_iteration": 2.815666437149048
    },
    {
      "auxiliary_loss_clip": 0.01238834,
      "auxiliary_loss_mlp": 0.01039864,
      "balance_loss_clip": 1.05764771,
      "balance_loss_mlp": 1.02858639,
      "epoch": 0.1567967293933746,
      "flos": 24827093769600.0,
      "grad_norm": 3.310614687451746,
      "language_loss": 0.81278598,
      "learning_rate": 3.833828393904117e-06,
      "loss": 0.83557296,
      "num_input_tokens_seen": 27548455,
      "step": 1304,
      "time_per_iteration": 2.8427071571350098
    },
    {
      "auxiliary_loss_clip": 0.01231705,
      "auxiliary_loss_mlp": 0.01033274,
      "balance_loss_clip": 1.05181122,
      "balance_loss_mlp": 1.02165723,
      "epoch": 0.15691697228401372,
      "flos": 19164555244800.0,
      "grad_norm": 3.187806887839263,
      "language_loss": 0.77839661,
      "learning_rate": 3.833517379387165e-06,
      "loss": 0.80104637,
      "num_input_tokens_seen": 27564910,
      "step": 1305,
      "time_per_iteration": 2.800363063812256
    },
    {
      "auxiliary_loss_clip": 0.01237826,
      "auxiliary_loss_mlp": 0.01040363,
      "balance_loss_clip": 1.06197059,
      "balance_loss_mlp": 1.02899063,
      "epoch": 0.1570372151746528,
      "flos": 24790931752320.0,
      "grad_norm": 2.7326335837209306,
      "language_loss": 0.89129472,
      "learning_rate": 3.833206086727085e-06,
      "loss": 0.91407657,
      "num_input_tokens_seen": 27584260,
      "step": 1306,
      "time_per_iteration": 3.8023293018341064
    },
    {
      "auxiliary_loss_clip": 0.01240424,
      "auxiliary_loss_mlp": 0.01035589,
      "balance_loss_clip": 1.05453789,
      "balance_loss_mlp": 1.0245924,
      "epoch": 0.15715745806529188,
      "flos": 24863650836480.0,
      "grad_norm": 2.1674841609535163,
      "language_loss": 0.70943356,
      "learning_rate": 3.8328945159710994e-06,
      "loss": 0.73219371,
      "num_input_tokens_seen": 27604440,
      "step": 1307,
      "time_per_iteration": 2.911123752593994
    },
    {
      "auxiliary_loss_clip": 0.01243863,
      "auxiliary_loss_mlp": 0.01209502,
      "balance_loss_clip": 1.06592011,
      "balance_loss_mlp": 1.00053012,
      "epoch": 0.157277700955931,
      "flos": 21872148491520.0,
      "grad_norm": 3.2111789156077735,
      "language_loss": 0.8881129,
      "learning_rate": 3.832582667166473e-06,
      "loss": 0.91264659,
      "num_input_tokens_seen": 27624250,
      "step": 1308,
      "time_per_iteration": 2.8135011196136475
    },
    {
      "auxiliary_loss_clip": 0.01235119,
      "auxiliary_loss_mlp": 0.01039778,
      "balance_loss_clip": 1.05773377,
      "balance_loss_mlp": 1.0273025,
      "epoch": 0.15739794384657008,
      "flos": 24533344344960.0,
      "grad_norm": 1.79495671408863,
      "language_loss": 0.81551236,
      "learning_rate": 3.8322705403605125e-06,
      "loss": 0.83826137,
      "num_input_tokens_seen": 27644595,
      "step": 1309,
      "time_per_iteration": 2.7762463092803955
    },
    {
      "auxiliary_loss_clip": 0.01228793,
      "auxiliary_loss_mlp": 0.01038262,
      "balance_loss_clip": 1.05822301,
      "balance_loss_mlp": 1.02784312,
      "epoch": 0.15751818673720916,
      "flos": 17745329998080.0,
      "grad_norm": 2.0461125909439697,
      "language_loss": 0.81022692,
      "learning_rate": 3.831958135600568e-06,
      "loss": 0.83289742,
      "num_input_tokens_seen": 27662145,
      "step": 1310,
      "time_per_iteration": 2.7273595333099365
    },
    {
      "auxiliary_loss_clip": 0.01234991,
      "auxiliary_loss_mlp": 0.01034283,
      "balance_loss_clip": 1.06155503,
      "balance_loss_mlp": 1.0243175,
      "epoch": 0.15763842962784824,
      "flos": 17858520731520.0,
      "grad_norm": 1.9735851530298212,
      "language_loss": 0.79887241,
      "learning_rate": 3.831645452934032e-06,
      "loss": 0.82156515,
      "num_input_tokens_seen": 27680575,
      "step": 1311,
      "time_per_iteration": 2.780900478363037
    },
    {
      "auxiliary_loss_clip": 0.01235233,
      "auxiliary_loss_mlp": 0.01044963,
      "balance_loss_clip": 1.06525886,
      "balance_loss_mlp": 1.03419888,
      "epoch": 0.15775867251848735,
      "flos": 26980908059520.0,
      "grad_norm": 2.8020588448230557,
      "language_loss": 0.80390894,
      "learning_rate": 3.831332492408336e-06,
      "loss": 0.82671088,
      "num_input_tokens_seen": 27701985,
      "step": 1312,
      "time_per_iteration": 2.923234701156616
    },
    {
      "auxiliary_loss_clip": 0.01229437,
      "auxiliary_loss_mlp": 0.01031217,
      "balance_loss_clip": 1.05557227,
      "balance_loss_mlp": 1.01991034,
      "epoch": 0.15787891540912644,
      "flos": 19240398812160.0,
      "grad_norm": 2.0150879553512735,
      "language_loss": 0.68859446,
      "learning_rate": 3.831019254070957e-06,
      "loss": 0.71120095,
      "num_input_tokens_seen": 27719770,
      "step": 1313,
      "time_per_iteration": 2.878654718399048
    },
    {
      "auxiliary_loss_clip": 0.01241926,
      "auxiliary_loss_mlp": 0.01035962,
      "balance_loss_clip": 1.05513906,
      "balance_loss_mlp": 1.02479196,
      "epoch": 0.15799915829976552,
      "flos": 27271102037760.0,
      "grad_norm": 4.446378984232253,
      "language_loss": 0.94790262,
      "learning_rate": 3.8307057379694135e-06,
      "loss": 0.97068143,
      "num_input_tokens_seen": 27739105,
      "step": 1314,
      "time_per_iteration": 2.919945001602173
    },
    {
      "auxiliary_loss_clip": 0.01233554,
      "auxiliary_loss_mlp": 0.01043217,
      "balance_loss_clip": 1.06203413,
      "balance_loss_mlp": 1.03199375,
      "epoch": 0.15811940119040463,
      "flos": 20405520270720.0,
      "grad_norm": 2.5741123181622156,
      "language_loss": 0.82677639,
      "learning_rate": 3.830391944151264e-06,
      "loss": 0.84954411,
      "num_input_tokens_seen": 27754985,
      "step": 1315,
      "time_per_iteration": 2.7507736682891846
    },
    {
      "auxiliary_loss_clip": 0.01233162,
      "auxiliary_loss_mlp": 0.01039283,
      "balance_loss_clip": 1.05552995,
      "balance_loss_mlp": 1.02845263,
      "epoch": 0.1582396440810437,
      "flos": 32599347661440.0,
      "grad_norm": 7.130559730875644,
      "language_loss": 0.67781937,
      "learning_rate": 3.830077872664114e-06,
      "loss": 0.70054382,
      "num_input_tokens_seen": 27776110,
      "step": 1316,
      "time_per_iteration": 2.7488856315612793
    },
    {
      "auxiliary_loss_clip": 0.01236988,
      "auxiliary_loss_mlp": 0.01036373,
      "balance_loss_clip": 1.05042386,
      "balance_loss_mlp": 1.02584672,
      "epoch": 0.1583598869716828,
      "flos": 33800559310080.0,
      "grad_norm": 1.9764371625259813,
      "language_loss": 0.72848916,
      "learning_rate": 3.829763523555604e-06,
      "loss": 0.75122285,
      "num_input_tokens_seen": 27796510,
      "step": 1317,
      "time_per_iteration": 2.788443088531494
    },
    {
      "auxiliary_loss_clip": 0.01228101,
      "auxiliary_loss_mlp": 0.01035941,
      "balance_loss_clip": 1.06384897,
      "balance_loss_mlp": 1.0259397,
      "epoch": 0.15848012986232188,
      "flos": 24681332378880.0,
      "grad_norm": 2.3915782101719447,
      "language_loss": 0.78343976,
      "learning_rate": 3.829448896873423e-06,
      "loss": 0.80608016,
      "num_input_tokens_seen": 27815610,
      "step": 1318,
      "time_per_iteration": 2.5627520084381104
    },
    {
      "auxiliary_loss_clip": 0.0122768,
      "auxiliary_loss_mlp": 0.01210145,
      "balance_loss_clip": 1.0566771,
      "balance_loss_mlp": 1.00057948,
      "epoch": 0.158600372752961,
      "flos": 22602068766720.0,
      "grad_norm": 1.8992354312649182,
      "language_loss": 0.79104227,
      "learning_rate": 3.829133992665299e-06,
      "loss": 0.81542051,
      "num_input_tokens_seen": 27834735,
      "step": 1319,
      "time_per_iteration": 2.6370689868927
    },
    {
      "auxiliary_loss_clip": 0.01218772,
      "auxiliary_loss_mlp": 0.01034266,
      "balance_loss_clip": 1.05734849,
      "balance_loss_mlp": 1.02388906,
      "epoch": 0.15872061564360007,
      "flos": 27927944092800.0,
      "grad_norm": 2.7259094586136254,
      "language_loss": 0.89096332,
      "learning_rate": 3.828818810979002e-06,
      "loss": 0.91349369,
      "num_input_tokens_seen": 27853065,
      "step": 1320,
      "time_per_iteration": 2.5977532863616943
    },
    {
      "auxiliary_loss_clip": 0.01229263,
      "auxiliary_loss_mlp": 0.01038379,
      "balance_loss_clip": 1.06374121,
      "balance_loss_mlp": 1.02762032,
      "epoch": 0.15884085853423915,
      "flos": 23696805525120.0,
      "grad_norm": 2.5033892246343963,
      "language_loss": 0.80460787,
      "learning_rate": 3.8285033518623454e-06,
      "loss": 0.82728434,
      "num_input_tokens_seen": 27873315,
      "step": 1321,
      "time_per_iteration": 2.6769912242889404
    },
    {
      "auxiliary_loss_clip": 0.01236584,
      "auxiliary_loss_mlp": 0.01042031,
      "balance_loss_clip": 1.06181991,
      "balance_loss_mlp": 1.02987802,
      "epoch": 0.15896110142487826,
      "flos": 23112359331840.0,
      "grad_norm": 3.424109295118474,
      "language_loss": 0.81243509,
      "learning_rate": 3.8281876153631845e-06,
      "loss": 0.83522129,
      "num_input_tokens_seen": 27890070,
      "step": 1322,
      "time_per_iteration": 2.6486692428588867
    },
    {
      "auxiliary_loss_clip": 0.01230988,
      "auxiliary_loss_mlp": 0.01041005,
      "balance_loss_clip": 1.05219042,
      "balance_loss_mlp": 1.02867317,
      "epoch": 0.15908134431551735,
      "flos": 14685238632960.0,
      "grad_norm": 2.2256520438732945,
      "language_loss": 0.64826798,
      "learning_rate": 3.827871601529416e-06,
      "loss": 0.67098796,
      "num_input_tokens_seen": 27908590,
      "step": 1323,
      "time_per_iteration": 2.729562759399414
    },
    {
      "auxiliary_loss_clip": 0.0122787,
      "auxiliary_loss_mlp": 0.01037844,
      "balance_loss_clip": 1.05359006,
      "balance_loss_mlp": 1.02700233,
      "epoch": 0.15920158720615643,
      "flos": 20193611984640.0,
      "grad_norm": 1.8621455631747952,
      "language_loss": 0.80734456,
      "learning_rate": 3.827555310408979e-06,
      "loss": 0.83000177,
      "num_input_tokens_seen": 27927985,
      "step": 1324,
      "time_per_iteration": 2.7722270488739014
    },
    {
      "auxiliary_loss_clip": 0.01231758,
      "auxiliary_loss_mlp": 0.01037536,
      "balance_loss_clip": 1.05965447,
      "balance_loss_mlp": 1.02633047,
      "epoch": 0.1593218300967955,
      "flos": 24826626892800.0,
      "grad_norm": 2.3048527488944903,
      "language_loss": 0.82998395,
      "learning_rate": 3.827238742049854e-06,
      "loss": 0.85267687,
      "num_input_tokens_seen": 27948280,
      "step": 1325,
      "time_per_iteration": 2.7419886589050293
    },
    {
      "auxiliary_loss_clip": 0.01229227,
      "auxiliary_loss_mlp": 0.01036476,
      "balance_loss_clip": 1.06067252,
      "balance_loss_mlp": 1.02540135,
      "epoch": 0.15944207298743462,
      "flos": 28328707111680.0,
      "grad_norm": 1.9191835645324642,
      "language_loss": 0.51829606,
      "learning_rate": 3.826921896500066e-06,
      "loss": 0.54095316,
      "num_input_tokens_seen": 27969565,
      "step": 1326,
      "time_per_iteration": 2.69081449508667
    },
    {
      "auxiliary_loss_clip": 0.01240407,
      "auxiliary_loss_mlp": 0.01035924,
      "balance_loss_clip": 1.05941916,
      "balance_loss_mlp": 1.02375281,
      "epoch": 0.1595623158780737,
      "flos": 22964838174720.0,
      "grad_norm": 1.9484136591495995,
      "language_loss": 0.78150713,
      "learning_rate": 3.826604773807678e-06,
      "loss": 0.80427051,
      "num_input_tokens_seen": 27987540,
      "step": 1327,
      "time_per_iteration": 2.735210418701172
    },
    {
      "auxiliary_loss_clip": 0.01230231,
      "auxiliary_loss_mlp": 0.01033692,
      "balance_loss_clip": 1.05404544,
      "balance_loss_mlp": 1.0220753,
      "epoch": 0.1596825587687128,
      "flos": 19710540950400.0,
      "grad_norm": 9.357176051864576,
      "language_loss": 0.73419988,
      "learning_rate": 3.826287374020798e-06,
      "loss": 0.7568391,
      "num_input_tokens_seen": 28002345,
      "step": 1328,
      "time_per_iteration": 3.5712969303131104
    },
    {
      "auxiliary_loss_clip": 0.01232142,
      "auxiliary_loss_mlp": 0.01037332,
      "balance_loss_clip": 1.06425738,
      "balance_loss_mlp": 1.02690744,
      "epoch": 0.1598028016593519,
      "flos": 22637727993600.0,
      "grad_norm": 2.032214917869746,
      "language_loss": 0.82392645,
      "learning_rate": 3.825969697187575e-06,
      "loss": 0.84662116,
      "num_input_tokens_seen": 28021675,
      "step": 1329,
      "time_per_iteration": 3.513540029525757
    },
    {
      "auxiliary_loss_clip": 0.01233866,
      "auxiliary_loss_mlp": 0.01034724,
      "balance_loss_clip": 1.05623078,
      "balance_loss_mlp": 1.0237267,
      "epoch": 0.15992304454999098,
      "flos": 20482908122880.0,
      "grad_norm": 1.90256447863637,
      "language_loss": 0.69319725,
      "learning_rate": 3.8256517433562015e-06,
      "loss": 0.71588302,
      "num_input_tokens_seen": 28039615,
      "step": 1330,
      "time_per_iteration": 2.6513447761535645
    },
    {
      "auxiliary_loss_clip": 0.01228004,
      "auxiliary_loss_mlp": 0.01033899,
      "balance_loss_clip": 1.0615406,
      "balance_loss_mlp": 1.02443981,
      "epoch": 0.16004328744063007,
      "flos": 17676094533120.0,
      "grad_norm": 2.3587264817883815,
      "language_loss": 0.91626418,
      "learning_rate": 3.82533351257491e-06,
      "loss": 0.93888319,
      "num_input_tokens_seen": 28057565,
      "step": 1331,
      "time_per_iteration": 2.575136661529541
    },
    {
      "auxiliary_loss_clip": 0.01231632,
      "auxiliary_loss_mlp": 0.01035566,
      "balance_loss_clip": 1.06351459,
      "balance_loss_mlp": 1.02516532,
      "epoch": 0.16016353033126918,
      "flos": 24098717779200.0,
      "grad_norm": 15.60100106599204,
      "language_loss": 0.88682306,
      "learning_rate": 3.825015004891975e-06,
      "loss": 0.909495,
      "num_input_tokens_seen": 28076305,
      "step": 1332,
      "time_per_iteration": 3.5430006980895996
    },
    {
      "auxiliary_loss_clip": 0.01229542,
      "auxiliary_loss_mlp": 0.01033452,
      "balance_loss_clip": 1.06134737,
      "balance_loss_mlp": 1.02284217,
      "epoch": 0.16028377322190826,
      "flos": 27634841112960.0,
      "grad_norm": 4.660578287898821,
      "language_loss": 0.75846422,
      "learning_rate": 3.824696220355716e-06,
      "loss": 0.78109419,
      "num_input_tokens_seen": 28097895,
      "step": 1333,
      "time_per_iteration": 2.6666665077209473
    },
    {
      "auxiliary_loss_clip": 0.01231117,
      "auxiliary_loss_mlp": 0.01041639,
      "balance_loss_clip": 1.05918431,
      "balance_loss_mlp": 1.03082097,
      "epoch": 0.16040401611254734,
      "flos": 20961202648320.0,
      "grad_norm": 2.0319329276445868,
      "language_loss": 0.7890619,
      "learning_rate": 3.824377159014491e-06,
      "loss": 0.81178939,
      "num_input_tokens_seen": 28118790,
      "step": 1334,
      "time_per_iteration": 2.6382217407226562
    },
    {
      "auxiliary_loss_clip": 0.01227884,
      "auxiliary_loss_mlp": 0.01038617,
      "balance_loss_clip": 1.06071067,
      "balance_loss_mlp": 1.02788782,
      "epoch": 0.16052425900318643,
      "flos": 21247051080960.0,
      "grad_norm": 1.8170414792344092,
      "language_loss": 0.8478263,
      "learning_rate": 3.824057820916702e-06,
      "loss": 0.87049127,
      "num_input_tokens_seen": 28135995,
      "step": 1335,
      "time_per_iteration": 2.652881622314453
    },
    {
      "auxiliary_loss_clip": 0.01234895,
      "auxiliary_loss_mlp": 0.01032667,
      "balance_loss_clip": 1.05980229,
      "balance_loss_mlp": 1.02104402,
      "epoch": 0.16064450189382554,
      "flos": 15524004096000.0,
      "grad_norm": 2.2169042117123388,
      "language_loss": 0.71660912,
      "learning_rate": 3.8237382061107904e-06,
      "loss": 0.73928475,
      "num_input_tokens_seen": 28152715,
      "step": 1336,
      "time_per_iteration": 2.615878105163574
    },
    {
      "auxiliary_loss_clip": 0.01225228,
      "auxiliary_loss_mlp": 0.01037841,
      "balance_loss_clip": 1.04550934,
      "balance_loss_mlp": 1.02763665,
      "epoch": 0.16076474478446462,
      "flos": 21178497974400.0,
      "grad_norm": 2.2477221495087787,
      "language_loss": 0.78846681,
      "learning_rate": 3.823418314645243e-06,
      "loss": 0.8110975,
      "num_input_tokens_seen": 28171590,
      "step": 1337,
      "time_per_iteration": 2.7186501026153564
    },
    {
      "auxiliary_loss_clip": 0.01219731,
      "auxiliary_loss_mlp": 0.01043559,
      "balance_loss_clip": 1.05459166,
      "balance_loss_mlp": 1.03348029,
      "epoch": 0.1608849876751037,
      "flos": 18366476912640.0,
      "grad_norm": 2.127212574474433,
      "language_loss": 0.75752938,
      "learning_rate": 3.823098146568588e-06,
      "loss": 0.78016233,
      "num_input_tokens_seen": 28191295,
      "step": 1338,
      "time_per_iteration": 2.69736909866333
    },
    {
      "auxiliary_loss_clip": 0.0123015,
      "auxiliary_loss_mlp": 0.01038275,
      "balance_loss_clip": 1.061252,
      "balance_loss_mlp": 1.02864313,
      "epoch": 0.1610052305657428,
      "flos": 29497024880640.0,
      "grad_norm": 2.000297492454486,
      "language_loss": 0.71844947,
      "learning_rate": 3.822777701929394e-06,
      "loss": 0.74113369,
      "num_input_tokens_seen": 28213120,
      "step": 1339,
      "time_per_iteration": 2.6749792098999023
    },
    {
      "auxiliary_loss_clip": 0.01218311,
      "auxiliary_loss_mlp": 0.01039998,
      "balance_loss_clip": 1.05696344,
      "balance_loss_mlp": 1.02897739,
      "epoch": 0.1611254734563819,
      "flos": 26797871329920.0,
      "grad_norm": 2.0619941143809304,
      "language_loss": 0.73809278,
      "learning_rate": 3.8224569807762714e-06,
      "loss": 0.76067591,
      "num_input_tokens_seen": 28232440,
      "step": 1340,
      "time_per_iteration": 2.7122223377227783
    },
    {
      "auxiliary_loss_clip": 0.01214293,
      "auxiliary_loss_mlp": 0.01038979,
      "balance_loss_clip": 1.04796338,
      "balance_loss_mlp": 1.02789259,
      "epoch": 0.16124571634702098,
      "flos": 22419570741120.0,
      "grad_norm": 9.421474067706464,
      "language_loss": 0.76486707,
      "learning_rate": 3.822135983157873e-06,
      "loss": 0.78739989,
      "num_input_tokens_seen": 28251715,
      "step": 1341,
      "time_per_iteration": 2.717628002166748
    },
    {
      "auxiliary_loss_clip": 0.01223881,
      "auxiliary_loss_mlp": 0.01209606,
      "balance_loss_clip": 1.05991042,
      "balance_loss_mlp": 1.00050306,
      "epoch": 0.16136595923766006,
      "flos": 10999116103680.0,
      "grad_norm": 2.6190179967489073,
      "language_loss": 0.84431159,
      "learning_rate": 3.821814709122896e-06,
      "loss": 0.86864644,
      "num_input_tokens_seen": 28269765,
      "step": 1342,
      "time_per_iteration": 2.593756914138794
    },
    {
      "auxiliary_loss_clip": 0.01227217,
      "auxiliary_loss_mlp": 0.01032887,
      "balance_loss_clip": 1.05673265,
      "balance_loss_mlp": 1.0230521,
      "epoch": 0.16148620212829917,
      "flos": 21214983214080.0,
      "grad_norm": 2.2204227260029863,
      "language_loss": 0.85125113,
      "learning_rate": 3.821493158720076e-06,
      "loss": 0.87385213,
      "num_input_tokens_seen": 28288870,
      "step": 1343,
      "time_per_iteration": 2.642725706100464
    },
    {
      "auxiliary_loss_clip": 0.01231962,
      "auxiliary_loss_mlp": 0.01031181,
      "balance_loss_clip": 1.05351973,
      "balance_loss_mlp": 1.02010083,
      "epoch": 0.16160644501893826,
      "flos": 16758468760320.0,
      "grad_norm": 2.7538587639655128,
      "language_loss": 0.72898126,
      "learning_rate": 3.821171331998191e-06,
      "loss": 0.75161266,
      "num_input_tokens_seen": 28305400,
      "step": 1344,
      "time_per_iteration": 2.697662830352783
    },
    {
      "auxiliary_loss_clip": 0.01149076,
      "auxiliary_loss_mlp": 0.01004596,
      "balance_loss_clip": 1.03137994,
      "balance_loss_mlp": 1.00142503,
      "epoch": 0.16172668790957734,
      "flos": 64444967308800.0,
      "grad_norm": 0.7309382065353885,
      "language_loss": 0.54469061,
      "learning_rate": 3.820849229006064e-06,
      "loss": 0.56622732,
      "num_input_tokens_seen": 28373150,
      "step": 1345,
      "time_per_iteration": 3.324277877807617
    },
    {
      "auxiliary_loss_clip": 0.01231232,
      "auxiliary_loss_mlp": 0.01033022,
      "balance_loss_clip": 1.06387758,
      "balance_loss_mlp": 1.02282357,
      "epoch": 0.16184693080021645,
      "flos": 23257689759360.0,
      "grad_norm": 2.1298343889364224,
      "language_loss": 0.70836461,
      "learning_rate": 3.8205268497925564e-06,
      "loss": 0.7310071,
      "num_input_tokens_seen": 28393620,
      "step": 1346,
      "time_per_iteration": 2.632805585861206
    },
    {
      "auxiliary_loss_clip": 0.01228572,
      "auxiliary_loss_mlp": 0.01035008,
      "balance_loss_clip": 1.06333661,
      "balance_loss_mlp": 1.02481008,
      "epoch": 0.16196717369085553,
      "flos": 17451113696640.0,
      "grad_norm": 2.4527415441776026,
      "language_loss": 0.78697574,
      "learning_rate": 3.8202041944065725e-06,
      "loss": 0.80961156,
      "num_input_tokens_seen": 28409440,
      "step": 1347,
      "time_per_iteration": 2.546416759490967
    },
    {
      "auxiliary_loss_clip": 0.01226526,
      "auxiliary_loss_mlp": 0.01038398,
      "balance_loss_clip": 1.0629313,
      "balance_loss_mlp": 1.02797341,
      "epoch": 0.16208741658149461,
      "flos": 23873377806720.0,
      "grad_norm": 2.212478316082295,
      "language_loss": 0.74083877,
      "learning_rate": 3.819881262897061e-06,
      "loss": 0.76348799,
      "num_input_tokens_seen": 28427575,
      "step": 1348,
      "time_per_iteration": 2.617990255355835
    },
    {
      "auxiliary_loss_clip": 0.0123416,
      "auxiliary_loss_mlp": 0.01035058,
      "balance_loss_clip": 1.06108356,
      "balance_loss_mlp": 1.02383494,
      "epoch": 0.1622076594721337,
      "flos": 25884806584320.0,
      "grad_norm": 2.0318107307831936,
      "language_loss": 0.73359811,
      "learning_rate": 3.819558055313008e-06,
      "loss": 0.75629032,
      "num_input_tokens_seen": 28448260,
      "step": 1349,
      "time_per_iteration": 2.6763665676116943
    },
    {
      "auxiliary_loss_clip": 0.01234919,
      "auxiliary_loss_mlp": 0.01039691,
      "balance_loss_clip": 1.06269109,
      "balance_loss_mlp": 1.02990353,
      "epoch": 0.1623279023627728,
      "flos": 21539759011200.0,
      "grad_norm": 2.180550628712443,
      "language_loss": 0.7753855,
      "learning_rate": 3.819234571703444e-06,
      "loss": 0.79813159,
      "num_input_tokens_seen": 28467085,
      "step": 1350,
      "time_per_iteration": 2.6147971153259277
    },
    {
      "auxiliary_loss_clip": 0.01221656,
      "auxiliary_loss_mlp": 0.01035872,
      "balance_loss_clip": 1.05747247,
      "balance_loss_mlp": 1.02531052,
      "epoch": 0.1624481452534119,
      "flos": 22085421494400.0,
      "grad_norm": 2.3222510057680865,
      "language_loss": 0.85357642,
      "learning_rate": 3.8189108121174435e-06,
      "loss": 0.8761518,
      "num_input_tokens_seen": 28486850,
      "step": 1351,
      "time_per_iteration": 2.629777193069458
    },
    {
      "auxiliary_loss_clip": 0.01227701,
      "auxiliary_loss_mlp": 0.01036855,
      "balance_loss_clip": 1.05984437,
      "balance_loss_mlp": 1.02692509,
      "epoch": 0.16256838814405097,
      "flos": 27087490690560.0,
      "grad_norm": 2.1851183409271644,
      "language_loss": 0.83606052,
      "learning_rate": 3.818586776604118e-06,
      "loss": 0.85870612,
      "num_input_tokens_seen": 28507490,
      "step": 1352,
      "time_per_iteration": 2.6850099563598633
    },
    {
      "auxiliary_loss_clip": 0.01226427,
      "auxiliary_loss_mlp": 0.01038784,
      "balance_loss_clip": 1.05777633,
      "balance_loss_mlp": 1.02897894,
      "epoch": 0.16268863103469008,
      "flos": 20120354196480.0,
      "grad_norm": 2.048142097810969,
      "language_loss": 0.61604321,
      "learning_rate": 3.818262465212625e-06,
      "loss": 0.63869524,
      "num_input_tokens_seen": 28527615,
      "step": 1353,
      "time_per_iteration": 2.6874189376831055
    },
    {
      "auxiliary_loss_clip": 0.01221323,
      "auxiliary_loss_mlp": 0.01048599,
      "balance_loss_clip": 1.06140089,
      "balance_loss_mlp": 1.03739917,
      "epoch": 0.16280887392532917,
      "flos": 18332792933760.0,
      "grad_norm": 2.1976432514423716,
      "language_loss": 0.77485323,
      "learning_rate": 3.817937877992161e-06,
      "loss": 0.79755235,
      "num_input_tokens_seen": 28544910,
      "step": 1354,
      "time_per_iteration": 3.4869916439056396
    },
    {
      "auxiliary_loss_clip": 0.0122987,
      "auxiliary_loss_mlp": 0.0120988,
      "balance_loss_clip": 1.05292773,
      "balance_loss_mlp": 1.00026751,
      "epoch": 0.16292911681596825,
      "flos": 11874330892800.0,
      "grad_norm": 2.369696758031917,
      "language_loss": 0.85822988,
      "learning_rate": 3.817613014991967e-06,
      "loss": 0.88262737,
      "num_input_tokens_seen": 28561050,
      "step": 1355,
      "time_per_iteration": 2.6927342414855957
    },
    {
      "auxiliary_loss_clip": 0.01222805,
      "auxiliary_loss_mlp": 0.01031979,
      "balance_loss_clip": 1.05383837,
      "balance_loss_mlp": 1.02172744,
      "epoch": 0.16304935970660733,
      "flos": 26103466627200.0,
      "grad_norm": 2.2418973308182077,
      "language_loss": 0.76692206,
      "learning_rate": 3.817287876261323e-06,
      "loss": 0.7894699,
      "num_input_tokens_seen": 28581385,
      "step": 1356,
      "time_per_iteration": 3.629509449005127
    },
    {
      "auxiliary_loss_clip": 0.01225409,
      "auxiliary_loss_mlp": 0.01033936,
      "balance_loss_clip": 1.0604527,
      "balance_loss_mlp": 1.02281976,
      "epoch": 0.16316960259724644,
      "flos": 29351945848320.0,
      "grad_norm": 1.9432791981167112,
      "language_loss": 0.80071175,
      "learning_rate": 3.816962461849553e-06,
      "loss": 0.82330525,
      "num_input_tokens_seen": 28603255,
      "step": 1357,
      "time_per_iteration": 2.73067569732666
    },
    {
      "auxiliary_loss_clip": 0.01223344,
      "auxiliary_loss_mlp": 0.01037553,
      "balance_loss_clip": 1.0592438,
      "balance_loss_mlp": 1.02702641,
      "epoch": 0.16328984548788553,
      "flos": 20886759711360.0,
      "grad_norm": 1.831104541604065,
      "language_loss": 0.84532845,
      "learning_rate": 3.8166367718060235e-06,
      "loss": 0.86793739,
      "num_input_tokens_seen": 28623145,
      "step": 1358,
      "time_per_iteration": 2.6769139766693115
    },
    {
      "auxiliary_loss_clip": 0.01222884,
      "auxiliary_loss_mlp": 0.010325,
      "balance_loss_clip": 1.05751991,
      "balance_loss_mlp": 1.02255189,
      "epoch": 0.1634100883785246,
      "flos": 18041090584320.0,
      "grad_norm": 4.488463532113664,
      "language_loss": 0.76261413,
      "learning_rate": 3.816310806180139e-06,
      "loss": 0.78516793,
      "num_input_tokens_seen": 28641555,
      "step": 1359,
      "time_per_iteration": 4.398776292800903
    },
    {
      "auxiliary_loss_clip": 0.01222528,
      "auxiliary_loss_mlp": 0.01034818,
      "balance_loss_clip": 1.05715287,
      "balance_loss_mlp": 1.02473307,
      "epoch": 0.16353033126916372,
      "flos": 24572128055040.0,
      "grad_norm": 1.909777219612215,
      "language_loss": 0.80841506,
      "learning_rate": 3.81598456502135e-06,
      "loss": 0.83098847,
      "num_input_tokens_seen": 28661575,
      "step": 1360,
      "time_per_iteration": 2.680424690246582
    },
    {
      "auxiliary_loss_clip": 0.01223944,
      "auxiliary_loss_mlp": 0.01036374,
      "balance_loss_clip": 1.05878997,
      "balance_loss_mlp": 1.0256393,
      "epoch": 0.1636505741598028,
      "flos": 19892895321600.0,
      "grad_norm": 1.9932743092766192,
      "language_loss": 0.86980581,
      "learning_rate": 3.8156580483791455e-06,
      "loss": 0.89240903,
      "num_input_tokens_seen": 28676765,
      "step": 1361,
      "time_per_iteration": 2.6371724605560303
    },
    {
      "auxiliary_loss_clip": 0.01228589,
      "auxiliary_loss_mlp": 0.01031988,
      "balance_loss_clip": 1.0626235,
      "balance_loss_mlp": 1.02195084,
      "epoch": 0.16377081705044189,
      "flos": 28402611344640.0,
      "grad_norm": 3.266835851372975,
      "language_loss": 0.77156281,
      "learning_rate": 3.815331256303059e-06,
      "loss": 0.79416859,
      "num_input_tokens_seen": 28696795,
      "step": 1362,
      "time_per_iteration": 2.653125286102295
    },
    {
      "auxiliary_loss_clip": 0.01224786,
      "auxiliary_loss_mlp": 0.01035095,
      "balance_loss_clip": 1.05770838,
      "balance_loss_mlp": 1.0246644,
      "epoch": 0.163891059941081,
      "flos": 21908059113600.0,
      "grad_norm": 2.435381639633159,
      "language_loss": 0.77296364,
      "learning_rate": 3.815004188842665e-06,
      "loss": 0.79556245,
      "num_input_tokens_seen": 28714835,
      "step": 1363,
      "time_per_iteration": 2.657667398452759
    },
    {
      "auxiliary_loss_clip": 0.01220461,
      "auxiliary_loss_mlp": 0.01032285,
      "balance_loss_clip": 1.05213118,
      "balance_loss_mlp": 1.02163923,
      "epoch": 0.16401130283172008,
      "flos": 26797619934720.0,
      "grad_norm": 1.9703997076393156,
      "language_loss": 0.79520535,
      "learning_rate": 3.814676846047578e-06,
      "loss": 0.81773281,
      "num_input_tokens_seen": 28735710,
      "step": 1364,
      "time_per_iteration": 2.6620066165924072
    },
    {
      "auxiliary_loss_clip": 0.01225031,
      "auxiliary_loss_mlp": 0.01036593,
      "balance_loss_clip": 1.06012487,
      "balance_loss_mlp": 1.02619815,
      "epoch": 0.16413154572235916,
      "flos": 32997417160320.0,
      "grad_norm": 1.8798922308001325,
      "language_loss": 0.70051914,
      "learning_rate": 3.8143492279674565e-06,
      "loss": 0.72313535,
      "num_input_tokens_seen": 28758405,
      "step": 1365,
      "time_per_iteration": 2.7262625694274902
    },
    {
      "auxiliary_loss_clip": 0.01146737,
      "auxiliary_loss_mlp": 0.01006129,
      "balance_loss_clip": 1.03110623,
      "balance_loss_mlp": 1.00313687,
      "epoch": 0.16425178861299825,
      "flos": 40113622074240.0,
      "grad_norm": 1.3420253768468202,
      "language_loss": 0.58391869,
      "learning_rate": 3.8140213346519997e-06,
      "loss": 0.60544729,
      "num_input_tokens_seen": 28809000,
      "step": 1366,
      "time_per_iteration": 2.921640396118164
    },
    {
      "auxiliary_loss_clip": 0.01213407,
      "auxiliary_loss_mlp": 0.0103275,
      "balance_loss_clip": 1.05138493,
      "balance_loss_mlp": 1.02246237,
      "epoch": 0.16437203150363736,
      "flos": 25447486498560.0,
      "grad_norm": 3.3550591602524076,
      "language_loss": 0.77394748,
      "learning_rate": 3.813693166150948e-06,
      "loss": 0.79640901,
      "num_input_tokens_seen": 28829210,
      "step": 1367,
      "time_per_iteration": 2.744572639465332
    },
    {
      "auxiliary_loss_clip": 0.01221167,
      "auxiliary_loss_mlp": 0.01033711,
      "balance_loss_clip": 1.05566478,
      "balance_loss_mlp": 1.02330446,
      "epoch": 0.16449227439427644,
      "flos": 23476888506240.0,
      "grad_norm": 3.126755920502524,
      "language_loss": 0.85391867,
      "learning_rate": 3.813364722514086e-06,
      "loss": 0.87646747,
      "num_input_tokens_seen": 28847545,
      "step": 1368,
      "time_per_iteration": 2.7133476734161377
    },
    {
      "auxiliary_loss_clip": 0.01224654,
      "auxiliary_loss_mlp": 0.01034484,
      "balance_loss_clip": 1.05832386,
      "balance_loss_mlp": 1.02429771,
      "epoch": 0.16461251728491552,
      "flos": 13545217802880.0,
      "grad_norm": 2.6007361773155333,
      "language_loss": 0.80696154,
      "learning_rate": 3.8130360037912368e-06,
      "loss": 0.82955289,
      "num_input_tokens_seen": 28863990,
      "step": 1369,
      "time_per_iteration": 2.5974462032318115
    },
    {
      "auxiliary_loss_clip": 0.01224742,
      "auxiliary_loss_mlp": 0.01033177,
      "balance_loss_clip": 1.05828011,
      "balance_loss_mlp": 1.02194118,
      "epoch": 0.16473276017555463,
      "flos": 23003298662400.0,
      "grad_norm": 3.930640749024751,
      "language_loss": 0.82006323,
      "learning_rate": 3.812707010032268e-06,
      "loss": 0.84264243,
      "num_input_tokens_seen": 28883045,
      "step": 1370,
      "time_per_iteration": 2.6293532848358154
    },
    {
      "auxiliary_loss_clip": 0.01231766,
      "auxiliary_loss_mlp": 0.01040052,
      "balance_loss_clip": 1.06335592,
      "balance_loss_mlp": 1.02972865,
      "epoch": 0.16485300306619372,
      "flos": 24790680357120.0,
      "grad_norm": 2.3705599750797828,
      "language_loss": 0.79492962,
      "learning_rate": 3.8123777412870863e-06,
      "loss": 0.81764781,
      "num_input_tokens_seen": 28902545,
      "step": 1371,
      "time_per_iteration": 2.665738105773926
    },
    {
      "auxiliary_loss_clip": 0.01230277,
      "auxiliary_loss_mlp": 0.01037289,
      "balance_loss_clip": 1.0560652,
      "balance_loss_mlp": 1.0271858,
      "epoch": 0.1649732459568328,
      "flos": 21106497162240.0,
      "grad_norm": 5.938264203961955,
      "language_loss": 0.78523397,
      "learning_rate": 3.812048197605643e-06,
      "loss": 0.80790961,
      "num_input_tokens_seen": 28921440,
      "step": 1372,
      "time_per_iteration": 2.6705572605133057
    },
    {
      "auxiliary_loss_clip": 0.01226903,
      "auxiliary_loss_mlp": 0.01027911,
      "balance_loss_clip": 1.06047618,
      "balance_loss_mlp": 1.01774836,
      "epoch": 0.16509348884747188,
      "flos": 20266726118400.0,
      "grad_norm": 2.02389895204754,
      "language_loss": 0.8167302,
      "learning_rate": 3.8117183790379277e-06,
      "loss": 0.83927834,
      "num_input_tokens_seen": 28939890,
      "step": 1373,
      "time_per_iteration": 2.6705100536346436
    },
    {
      "auxiliary_loss_clip": 0.01224203,
      "auxiliary_loss_mlp": 0.01033635,
      "balance_loss_clip": 1.06067443,
      "balance_loss_mlp": 1.02325773,
      "epoch": 0.165213731738111,
      "flos": 11035493602560.0,
      "grad_norm": 9.321003558054901,
      "language_loss": 0.93757367,
      "learning_rate": 3.811388285633976e-06,
      "loss": 0.96015203,
      "num_input_tokens_seen": 28955875,
      "step": 1374,
      "time_per_iteration": 2.616791009902954
    },
    {
      "auxiliary_loss_clip": 0.01233465,
      "auxiliary_loss_mlp": 0.01038184,
      "balance_loss_clip": 1.05354619,
      "balance_loss_mlp": 1.02761579,
      "epoch": 0.16533397462875007,
      "flos": 29972051268480.0,
      "grad_norm": 2.283511038593715,
      "language_loss": 0.6209451,
      "learning_rate": 3.811057917443861e-06,
      "loss": 0.64366156,
      "num_input_tokens_seen": 28975140,
      "step": 1375,
      "time_per_iteration": 2.762145519256592
    },
    {
      "auxiliary_loss_clip": 0.01141342,
      "auxiliary_loss_mlp": 0.01003311,
      "balance_loss_clip": 1.03019154,
      "balance_loss_mlp": 1.00072372,
      "epoch": 0.16545421751938916,
      "flos": 65556763027200.0,
      "grad_norm": 0.8658461808742262,
      "language_loss": 0.68237251,
      "learning_rate": 3.8107272745177e-06,
      "loss": 0.70381904,
      "num_input_tokens_seen": 29047470,
      "step": 1376,
      "time_per_iteration": 3.3721554279327393
    },
    {
      "auxiliary_loss_clip": 0.0123041,
      "auxiliary_loss_mlp": 0.01032078,
      "balance_loss_clip": 1.05780435,
      "balance_loss_mlp": 1.02224922,
      "epoch": 0.16557446041002827,
      "flos": 22492361652480.0,
      "grad_norm": 1.895093928263085,
      "language_loss": 0.78999031,
      "learning_rate": 3.8103963569056513e-06,
      "loss": 0.81261516,
      "num_input_tokens_seen": 29066605,
      "step": 1377,
      "time_per_iteration": 2.662893772125244
    },
    {
      "auxiliary_loss_clip": 0.01220591,
      "auxiliary_loss_mlp": 0.01034589,
      "balance_loss_clip": 1.05422401,
      "balance_loss_mlp": 1.02424717,
      "epoch": 0.16569470330066735,
      "flos": 24602723464320.0,
      "grad_norm": 1.8678351388265575,
      "language_loss": 0.88226587,
      "learning_rate": 3.8100651646579146e-06,
      "loss": 0.9048177,
      "num_input_tokens_seen": 29085815,
      "step": 1378,
      "time_per_iteration": 2.687232494354248
    },
    {
      "auxiliary_loss_clip": 0.01221144,
      "auxiliary_loss_mlp": 0.01035898,
      "balance_loss_clip": 1.05318594,
      "balance_loss_mlp": 1.02562857,
      "epoch": 0.16581494619130643,
      "flos": 15006207588480.0,
      "grad_norm": 2.3426466075661834,
      "language_loss": 0.92361987,
      "learning_rate": 3.8097336978247317e-06,
      "loss": 0.94619036,
      "num_input_tokens_seen": 29102520,
      "step": 1379,
      "time_per_iteration": 2.6824018955230713
    },
    {
      "auxiliary_loss_clip": 0.01209763,
      "auxiliary_loss_mlp": 0.01025678,
      "balance_loss_clip": 1.05232668,
      "balance_loss_mlp": 1.01495481,
      "epoch": 0.16593518908194552,
      "flos": 17420338719360.0,
      "grad_norm": 8.588334760514174,
      "language_loss": 0.88865024,
      "learning_rate": 3.8094019564563854e-06,
      "loss": 0.91100466,
      "num_input_tokens_seen": 29119450,
      "step": 1380,
      "time_per_iteration": 2.5967514514923096
    },
    {
      "auxiliary_loss_clip": 0.01222453,
      "auxiliary_loss_mlp": 0.012098,
      "balance_loss_clip": 1.0595696,
      "balance_loss_mlp": 1.00028992,
      "epoch": 0.16605543197258463,
      "flos": 20412631163520.0,
      "grad_norm": 2.4091508778975745,
      "language_loss": 0.75264162,
      "learning_rate": 3.809069940603201e-06,
      "loss": 0.77696413,
      "num_input_tokens_seen": 29137405,
      "step": 1381,
      "time_per_iteration": 3.551093578338623
    },
    {
      "auxiliary_loss_clip": 0.01216533,
      "auxiliary_loss_mlp": 0.01033062,
      "balance_loss_clip": 1.05636263,
      "balance_loss_mlp": 1.02306032,
      "epoch": 0.1661756748632237,
      "flos": 14209745368320.0,
      "grad_norm": 2.1417443791577,
      "language_loss": 0.78015554,
      "learning_rate": 3.8087376503155452e-06,
      "loss": 0.80265152,
      "num_input_tokens_seen": 29154890,
      "step": 1382,
      "time_per_iteration": 3.5380561351776123
    },
    {
      "auxiliary_loss_clip": 0.01135243,
      "auxiliary_loss_mlp": 0.0100545,
      "balance_loss_clip": 1.02773523,
      "balance_loss_mlp": 1.00280404,
      "epoch": 0.1662959177538628,
      "flos": 66080877350400.0,
      "grad_norm": 0.9525768924744044,
      "language_loss": 0.56237286,
      "learning_rate": 3.808405085643826e-06,
      "loss": 0.58377981,
      "num_input_tokens_seen": 29219770,
      "step": 1383,
      "time_per_iteration": 3.2504231929779053
    },
    {
      "auxiliary_loss_clip": 0.01225181,
      "auxiliary_loss_mlp": 0.01208742,
      "balance_loss_clip": 1.06127834,
      "balance_loss_mlp": 1.00018382,
      "epoch": 0.1664161606445019,
      "flos": 20740567357440.0,
      "grad_norm": 6.292293262893507,
      "language_loss": 0.89441806,
      "learning_rate": 3.8080722466384925e-06,
      "loss": 0.91875732,
      "num_input_tokens_seen": 29237620,
      "step": 1384,
      "time_per_iteration": 2.6558120250701904
    },
    {
      "auxiliary_loss_clip": 0.01225733,
      "auxiliary_loss_mlp": 0.01036328,
      "balance_loss_clip": 1.05922151,
      "balance_loss_mlp": 1.02500319,
      "epoch": 0.166536403535141,
      "flos": 25260930236160.0,
      "grad_norm": 2.1942675804413945,
      "language_loss": 0.71357208,
      "learning_rate": 3.8077391333500376e-06,
      "loss": 0.7361927,
      "num_input_tokens_seen": 29256760,
      "step": 1385,
      "time_per_iteration": 3.5312440395355225
    },
    {
      "auxiliary_loss_clip": 0.01228414,
      "auxiliary_loss_mlp": 0.01031765,
      "balance_loss_clip": 1.06093872,
      "balance_loss_mlp": 1.02224612,
      "epoch": 0.16665664642578007,
      "flos": 25447450584960.0,
      "grad_norm": 2.242219846935983,
      "language_loss": 0.76443642,
      "learning_rate": 3.8074057458289934e-06,
      "loss": 0.78703821,
      "num_input_tokens_seen": 29277450,
      "step": 1386,
      "time_per_iteration": 3.55735182762146
    },
    {
      "auxiliary_loss_clip": 0.01229583,
      "auxiliary_loss_mlp": 0.01031355,
      "balance_loss_clip": 1.0571475,
      "balance_loss_mlp": 1.02153802,
      "epoch": 0.16677688931641918,
      "flos": 22200767043840.0,
      "grad_norm": 2.568905307593911,
      "language_loss": 0.82286072,
      "learning_rate": 3.807072084125934e-06,
      "loss": 0.84547007,
      "num_input_tokens_seen": 29299300,
      "step": 1387,
      "time_per_iteration": 2.688310146331787
    },
    {
      "auxiliary_loss_clip": 0.01222889,
      "auxiliary_loss_mlp": 0.01035026,
      "balance_loss_clip": 1.05794346,
      "balance_loss_mlp": 1.02438092,
      "epoch": 0.16689713220705826,
      "flos": 16945958776320.0,
      "grad_norm": 3.267273395289391,
      "language_loss": 0.80559409,
      "learning_rate": 3.806738148291477e-06,
      "loss": 0.82817322,
      "num_input_tokens_seen": 29316125,
      "step": 1388,
      "time_per_iteration": 2.6440553665161133
    },
    {
      "auxiliary_loss_clip": 0.01233383,
      "auxiliary_loss_mlp": 0.01034716,
      "balance_loss_clip": 1.04980206,
      "balance_loss_mlp": 1.02367175,
      "epoch": 0.16701737509769735,
      "flos": 36244423923840.0,
      "grad_norm": 2.5161549302549964,
      "language_loss": 0.71539307,
      "learning_rate": 3.8064039383762793e-06,
      "loss": 0.73807406,
      "num_input_tokens_seen": 29338490,
      "step": 1389,
      "time_per_iteration": 2.9398083686828613
    },
    {
      "auxiliary_loss_clip": 0.0122483,
      "auxiliary_loss_mlp": 0.01034325,
      "balance_loss_clip": 1.06048679,
      "balance_loss_mlp": 1.02421653,
      "epoch": 0.16713761798833643,
      "flos": 23258659426560.0,
      "grad_norm": 2.1022491248691146,
      "language_loss": 0.77190137,
      "learning_rate": 3.8060694544310396e-06,
      "loss": 0.79449284,
      "num_input_tokens_seen": 29357000,
      "step": 1390,
      "time_per_iteration": 2.681379795074463
    },
    {
      "auxiliary_loss_clip": 0.01223895,
      "auxiliary_loss_mlp": 0.01041301,
      "balance_loss_clip": 1.05946875,
      "balance_loss_mlp": 1.03001213,
      "epoch": 0.16725786087897554,
      "flos": 25302515207040.0,
      "grad_norm": 2.2121419543707392,
      "language_loss": 0.78899449,
      "learning_rate": 3.8057346965065006e-06,
      "loss": 0.81164646,
      "num_input_tokens_seen": 29378230,
      "step": 1391,
      "time_per_iteration": 2.66225004196167
    },
    {
      "auxiliary_loss_clip": 0.01223441,
      "auxiliary_loss_mlp": 0.01032835,
      "balance_loss_clip": 1.05798745,
      "balance_loss_mlp": 1.02268982,
      "epoch": 0.16737810376961462,
      "flos": 31831541516160.0,
      "grad_norm": 1.757100287584233,
      "language_loss": 0.84432018,
      "learning_rate": 3.805399664653443e-06,
      "loss": 0.86688292,
      "num_input_tokens_seen": 29400370,
      "step": 1392,
      "time_per_iteration": 2.7353899478912354
    },
    {
      "auxiliary_loss_clip": 0.01225876,
      "auxiliary_loss_mlp": 0.01034351,
      "balance_loss_clip": 1.0604465,
      "balance_loss_mlp": 1.02372968,
      "epoch": 0.1674983466602537,
      "flos": 27961843553280.0,
      "grad_norm": 3.173975031756122,
      "language_loss": 0.74600828,
      "learning_rate": 3.805064358922692e-06,
      "loss": 0.7686106,
      "num_input_tokens_seen": 29418660,
      "step": 1393,
      "time_per_iteration": 2.6289494037628174
    },
    {
      "auxiliary_loss_clip": 0.01229539,
      "auxiliary_loss_mlp": 0.01032709,
      "balance_loss_clip": 1.05918443,
      "balance_loss_mlp": 1.02202821,
      "epoch": 0.16761858955089282,
      "flos": 21762656858880.0,
      "grad_norm": 1.9849919162597383,
      "language_loss": 0.80870312,
      "learning_rate": 3.8047287793651136e-06,
      "loss": 0.83132565,
      "num_input_tokens_seen": 29440105,
      "step": 1394,
      "time_per_iteration": 2.6204206943511963
    },
    {
      "auxiliary_loss_clip": 0.01229283,
      "auxiliary_loss_mlp": 0.01039145,
      "balance_loss_clip": 1.05561423,
      "balance_loss_mlp": 1.02912569,
      "epoch": 0.1677388324415319,
      "flos": 23805507058560.0,
      "grad_norm": 2.515244874646044,
      "language_loss": 0.88732374,
      "learning_rate": 3.8043929260316137e-06,
      "loss": 0.91000801,
      "num_input_tokens_seen": 29458260,
      "step": 1395,
      "time_per_iteration": 2.738677501678467
    },
    {
      "auxiliary_loss_clip": 0.01232624,
      "auxiliary_loss_mlp": 0.01038973,
      "balance_loss_clip": 1.06387806,
      "balance_loss_mlp": 1.02838159,
      "epoch": 0.16785907533217098,
      "flos": 20558859431040.0,
      "grad_norm": 9.46243447863102,
      "language_loss": 0.83467633,
      "learning_rate": 3.8040567989731417e-06,
      "loss": 0.85739231,
      "num_input_tokens_seen": 29476205,
      "step": 1396,
      "time_per_iteration": 2.6262567043304443
    },
    {
      "auxiliary_loss_clip": 0.01220814,
      "auxiliary_loss_mlp": 0.01029223,
      "balance_loss_clip": 1.05998194,
      "balance_loss_mlp": 1.01932287,
      "epoch": 0.16797931822281006,
      "flos": 15669657745920.0,
      "grad_norm": 2.194846584705994,
      "language_loss": 0.79714161,
      "learning_rate": 3.8037203982406876e-06,
      "loss": 0.81964207,
      "num_input_tokens_seen": 29494370,
      "step": 1397,
      "time_per_iteration": 2.637071132659912
    },
    {
      "auxiliary_loss_clip": 0.01224632,
      "auxiliary_loss_mlp": 0.0103536,
      "balance_loss_clip": 1.06234193,
      "balance_loss_mlp": 1.02449405,
      "epoch": 0.16809956111344918,
      "flos": 16541101607040.0,
      "grad_norm": 1.838923993833643,
      "language_loss": 0.73165727,
      "learning_rate": 3.8033837238852835e-06,
      "loss": 0.7542572,
      "num_input_tokens_seen": 29511070,
      "step": 1398,
      "time_per_iteration": 2.5825815200805664
    },
    {
      "auxiliary_loss_clip": 0.01216518,
      "auxiliary_loss_mlp": 0.01033498,
      "balance_loss_clip": 1.05286872,
      "balance_loss_mlp": 1.02356172,
      "epoch": 0.16821980400408826,
      "flos": 23258084808960.0,
      "grad_norm": 1.96986802431549,
      "language_loss": 0.69537783,
      "learning_rate": 3.8030467759580017e-06,
      "loss": 0.71787798,
      "num_input_tokens_seen": 29531990,
      "step": 1399,
      "time_per_iteration": 2.6798224449157715
    },
    {
      "auxiliary_loss_clip": 0.01227231,
      "auxiliary_loss_mlp": 0.01035751,
      "balance_loss_clip": 1.05746508,
      "balance_loss_mlp": 1.0248673,
      "epoch": 0.16834004689472734,
      "flos": 20774754126720.0,
      "grad_norm": 2.430810145511587,
      "language_loss": 0.87230313,
      "learning_rate": 3.802709554509958e-06,
      "loss": 0.89493299,
      "num_input_tokens_seen": 29549790,
      "step": 1400,
      "time_per_iteration": 2.6373069286346436
    },
    {
      "auxiliary_loss_clip": 0.01224829,
      "auxiliary_loss_mlp": 0.01031024,
      "balance_loss_clip": 1.05474174,
      "balance_loss_mlp": 1.02133799,
      "epoch": 0.16846028978536645,
      "flos": 26687302289280.0,
      "grad_norm": 2.737751854797037,
      "language_loss": 0.79755855,
      "learning_rate": 3.8023720595923083e-06,
      "loss": 0.82011712,
      "num_input_tokens_seen": 29569045,
      "step": 1401,
      "time_per_iteration": 2.6911461353302
    },
    {
      "auxiliary_loss_clip": 0.01225461,
      "auxiliary_loss_mlp": 0.01034853,
      "balance_loss_clip": 1.05118656,
      "balance_loss_mlp": 1.02428532,
      "epoch": 0.16858053267600553,
      "flos": 18843298980480.0,
      "grad_norm": 2.3341731091674913,
      "language_loss": 0.87681937,
      "learning_rate": 3.80203429125625e-06,
      "loss": 0.89942253,
      "num_input_tokens_seen": 29587220,
      "step": 1402,
      "time_per_iteration": 2.700289487838745
    },
    {
      "auxiliary_loss_clip": 0.01222046,
      "auxiliary_loss_mlp": 0.01030944,
      "balance_loss_clip": 1.05122709,
      "balance_loss_mlp": 1.02068663,
      "epoch": 0.16870077556664462,
      "flos": 27744548227200.0,
      "grad_norm": 2.5086620403760427,
      "language_loss": 0.70133168,
      "learning_rate": 3.8016962495530225e-06,
      "loss": 0.72386158,
      "num_input_tokens_seen": 29606410,
      "step": 1403,
      "time_per_iteration": 2.8050341606140137
    },
    {
      "auxiliary_loss_clip": 0.01224719,
      "auxiliary_loss_mlp": 0.01035038,
      "balance_loss_clip": 1.05960071,
      "balance_loss_mlp": 1.02528119,
      "epoch": 0.1688210184572837,
      "flos": 13730768484480.0,
      "grad_norm": 3.0481347534200722,
      "language_loss": 0.76821786,
      "learning_rate": 3.8013579345339063e-06,
      "loss": 0.79081547,
      "num_input_tokens_seen": 29621275,
      "step": 1404,
      "time_per_iteration": 2.617830276489258
    },
    {
      "auxiliary_loss_clip": 0.01231874,
      "auxiliary_loss_mlp": 0.01029873,
      "balance_loss_clip": 1.05509353,
      "balance_loss_mlp": 1.01928723,
      "epoch": 0.1689412613479228,
      "flos": 26468785900800.0,
      "grad_norm": 2.565373651817197,
      "language_loss": 0.69413412,
      "learning_rate": 3.801019346250224e-06,
      "loss": 0.71675158,
      "num_input_tokens_seen": 29641420,
      "step": 1405,
      "time_per_iteration": 2.801877975463867
    },
    {
      "auxiliary_loss_clip": 0.01224629,
      "auxiliary_loss_mlp": 0.01033637,
      "balance_loss_clip": 1.06001961,
      "balance_loss_mlp": 1.02324796,
      "epoch": 0.1690615042385619,
      "flos": 21138852337920.0,
      "grad_norm": 2.290518948290069,
      "language_loss": 0.83621359,
      "learning_rate": 3.8006804847533395e-06,
      "loss": 0.85879624,
      "num_input_tokens_seen": 29660935,
      "step": 1406,
      "time_per_iteration": 2.6641931533813477
    },
    {
      "auxiliary_loss_clip": 0.01226343,
      "auxiliary_loss_mlp": 0.01037291,
      "balance_loss_clip": 1.0605458,
      "balance_loss_mlp": 1.02796888,
      "epoch": 0.16918174712920098,
      "flos": 20849340718080.0,
      "grad_norm": 2.809336036310197,
      "language_loss": 0.85730857,
      "learning_rate": 3.8003413500946556e-06,
      "loss": 0.87994492,
      "num_input_tokens_seen": 29681045,
      "step": 1407,
      "time_per_iteration": 3.501736640930176
    },
    {
      "auxiliary_loss_clip": 0.01228916,
      "auxiliary_loss_mlp": 0.01036396,
      "balance_loss_clip": 1.05869973,
      "balance_loss_mlp": 1.02556586,
      "epoch": 0.1693019900198401,
      "flos": 16983270028800.0,
      "grad_norm": 2.616028309506093,
      "language_loss": 0.83883035,
      "learning_rate": 3.8000019423256216e-06,
      "loss": 0.86148345,
      "num_input_tokens_seen": 29698810,
      "step": 1408,
      "time_per_iteration": 2.617536783218384
    },
    {
      "auxiliary_loss_clip": 0.01217186,
      "auxiliary_loss_mlp": 0.01044579,
      "balance_loss_clip": 1.05641305,
      "balance_loss_mlp": 1.03459537,
      "epoch": 0.16942223291047917,
      "flos": 26796901662720.0,
      "grad_norm": 2.022701637519543,
      "language_loss": 0.88036543,
      "learning_rate": 3.7996622614977234e-06,
      "loss": 0.90298307,
      "num_input_tokens_seen": 29720000,
      "step": 1409,
      "time_per_iteration": 3.6677005290985107
    },
    {
      "auxiliary_loss_clip": 0.01226479,
      "auxiliary_loss_mlp": 0.01034154,
      "balance_loss_clip": 1.05904281,
      "balance_loss_mlp": 1.02460575,
      "epoch": 0.16954247580111825,
      "flos": 18583700411520.0,
      "grad_norm": 1.928634311334262,
      "language_loss": 0.78767508,
      "learning_rate": 3.799322307662492e-06,
      "loss": 0.8102814,
      "num_input_tokens_seen": 29737820,
      "step": 1410,
      "time_per_iteration": 2.641578197479248
    },
    {
      "auxiliary_loss_clip": 0.01234572,
      "auxiliary_loss_mlp": 0.01033481,
      "balance_loss_clip": 1.05277729,
      "balance_loss_mlp": 1.02298462,
      "epoch": 0.16966271869175734,
      "flos": 13983651210240.0,
      "grad_norm": 2.526852457800832,
      "language_loss": 0.83886695,
      "learning_rate": 3.798982080871496e-06,
      "loss": 0.86154747,
      "num_input_tokens_seen": 29752960,
      "step": 1411,
      "time_per_iteration": 2.7168009281158447
    },
    {
      "auxiliary_loss_clip": 0.01228741,
      "auxiliary_loss_mlp": 0.0103609,
      "balance_loss_clip": 1.06311429,
      "balance_loss_mlp": 1.02555203,
      "epoch": 0.16978296158239645,
      "flos": 37487328284160.0,
      "grad_norm": 2.1678570368330705,
      "language_loss": 0.67772281,
      "learning_rate": 3.798641581176349e-06,
      "loss": 0.70037115,
      "num_input_tokens_seen": 29775240,
      "step": 1412,
      "time_per_iteration": 4.554715871810913
    },
    {
      "auxiliary_loss_clip": 0.01228944,
      "auxiliary_loss_mlp": 0.01041468,
      "balance_loss_clip": 1.05634117,
      "balance_loss_mlp": 1.03048301,
      "epoch": 0.16990320447303553,
      "flos": 28328958506880.0,
      "grad_norm": 1.7855166038069086,
      "language_loss": 0.74846363,
      "learning_rate": 3.7983008086287044e-06,
      "loss": 0.77116776,
      "num_input_tokens_seen": 29796560,
      "step": 1413,
      "time_per_iteration": 2.690077543258667
    },
    {
      "auxiliary_loss_clip": 0.01223021,
      "auxiliary_loss_mlp": 0.0103627,
      "balance_loss_clip": 1.05472088,
      "balance_loss_mlp": 1.0252912,
      "epoch": 0.1700234473636746,
      "flos": 20188189031040.0,
      "grad_norm": 2.7350342561709966,
      "language_loss": 0.79346716,
      "learning_rate": 3.797959763280257e-06,
      "loss": 0.81606013,
      "num_input_tokens_seen": 29815245,
      "step": 1414,
      "time_per_iteration": 2.656980276107788
    },
    {
      "auxiliary_loss_clip": 0.01230697,
      "auxiliary_loss_mlp": 0.01042926,
      "balance_loss_clip": 1.05972505,
      "balance_loss_mlp": 1.03328812,
      "epoch": 0.17014369025431372,
      "flos": 24858658846080.0,
      "grad_norm": 1.80874507615427,
      "language_loss": 0.79086435,
      "learning_rate": 3.797618445182743e-06,
      "loss": 0.8136006,
      "num_input_tokens_seen": 29836640,
      "step": 1415,
      "time_per_iteration": 2.673178195953369
    },
    {
      "auxiliary_loss_clip": 0.01227562,
      "auxiliary_loss_mlp": 0.01033387,
      "balance_loss_clip": 1.0521642,
      "balance_loss_mlp": 1.02271807,
      "epoch": 0.1702639331449528,
      "flos": 16467233287680.0,
      "grad_norm": 2.8372399806552075,
      "language_loss": 0.84706271,
      "learning_rate": 3.79727685438794e-06,
      "loss": 0.86967224,
      "num_input_tokens_seen": 29850830,
      "step": 1416,
      "time_per_iteration": 2.7079267501831055
    },
    {
      "auxiliary_loss_clip": 0.0113796,
      "auxiliary_loss_mlp": 0.01004881,
      "balance_loss_clip": 1.03155625,
      "balance_loss_mlp": 1.00223446,
      "epoch": 0.1703841760355919,
      "flos": 52508870979840.0,
      "grad_norm": 0.859339918010511,
      "language_loss": 0.61614001,
      "learning_rate": 3.796934990947667e-06,
      "loss": 0.63756847,
      "num_input_tokens_seen": 29912515,
      "step": 1417,
      "time_per_iteration": 3.221064329147339
    },
    {
      "auxiliary_loss_clip": 0.01134873,
      "auxiliary_loss_mlp": 0.01001676,
      "balance_loss_clip": 1.02854431,
      "balance_loss_mlp": 0.99897003,
      "epoch": 0.170504418926231,
      "flos": 49370637576960.0,
      "grad_norm": 0.881279394685571,
      "language_loss": 0.62390292,
      "learning_rate": 3.7965928549137854e-06,
      "loss": 0.64526838,
      "num_input_tokens_seen": 29969330,
      "step": 1418,
      "time_per_iteration": 3.075809955596924
    },
    {
      "auxiliary_loss_clip": 0.01232618,
      "auxiliary_loss_mlp": 0.01036341,
      "balance_loss_clip": 1.05119383,
      "balance_loss_mlp": 1.02525437,
      "epoch": 0.17062466181687008,
      "flos": 25849219184640.0,
      "grad_norm": 2.4905254827010954,
      "language_loss": 0.77565247,
      "learning_rate": 3.7962504463381953e-06,
      "loss": 0.79834211,
      "num_input_tokens_seen": 29990820,
      "step": 1419,
      "time_per_iteration": 2.7516989707946777
    },
    {
      "auxiliary_loss_clip": 0.0122673,
      "auxiliary_loss_mlp": 0.01210217,
      "balance_loss_clip": 1.06157696,
      "balance_loss_mlp": 1.00024271,
      "epoch": 0.17074490470750917,
      "flos": 20960412549120.0,
      "grad_norm": 2.3792269166046074,
      "language_loss": 0.78805631,
      "learning_rate": 3.7959077652728412e-06,
      "loss": 0.81242579,
      "num_input_tokens_seen": 30009275,
      "step": 1420,
      "time_per_iteration": 2.637701988220215
    },
    {
      "auxiliary_loss_clip": 0.01227237,
      "auxiliary_loss_mlp": 0.0104022,
      "balance_loss_clip": 1.05585396,
      "balance_loss_mlp": 1.0299859,
      "epoch": 0.17086514759814825,
      "flos": 20959766104320.0,
      "grad_norm": 2.1131928724941917,
      "language_loss": 0.77808183,
      "learning_rate": 3.795564811769707e-06,
      "loss": 0.80075639,
      "num_input_tokens_seen": 30027630,
      "step": 1421,
      "time_per_iteration": 2.6798622608184814
    },
    {
      "auxiliary_loss_clip": 0.01227176,
      "auxiliary_loss_mlp": 0.01035789,
      "balance_loss_clip": 1.05987263,
      "balance_loss_mlp": 1.02448773,
      "epoch": 0.17098539048878736,
      "flos": 28474073452800.0,
      "grad_norm": 2.075999594328629,
      "language_loss": 0.78417182,
      "learning_rate": 3.795221585880818e-06,
      "loss": 0.80680144,
      "num_input_tokens_seen": 30048310,
      "step": 1422,
      "time_per_iteration": 2.7162790298461914
    },
    {
      "auxiliary_loss_clip": 0.01230944,
      "auxiliary_loss_mlp": 0.01036817,
      "balance_loss_clip": 1.06035185,
      "balance_loss_mlp": 1.02746558,
      "epoch": 0.17110563337942644,
      "flos": 16290014561280.0,
      "grad_norm": 2.0143600713331256,
      "language_loss": 0.91334081,
      "learning_rate": 3.794878087658242e-06,
      "loss": 0.93601841,
      "num_input_tokens_seen": 30066080,
      "step": 1423,
      "time_per_iteration": 2.836254119873047
    },
    {
      "auxiliary_loss_clip": 0.01229002,
      "auxiliary_loss_mlp": 0.01037712,
      "balance_loss_clip": 1.05934715,
      "balance_loss_mlp": 1.02778816,
      "epoch": 0.17122587627006552,
      "flos": 29674207693440.0,
      "grad_norm": 1.8228638347978732,
      "language_loss": 0.78641605,
      "learning_rate": 3.7945343171540873e-06,
      "loss": 0.8090831,
      "num_input_tokens_seen": 30086955,
      "step": 1424,
      "time_per_iteration": 2.7024757862091064
    },
    {
      "auxiliary_loss_clip": 0.01225507,
      "auxiliary_loss_mlp": 0.01041676,
      "balance_loss_clip": 1.05993199,
      "balance_loss_mlp": 1.0308156,
      "epoch": 0.17134611916070464,
      "flos": 25338389915520.0,
      "grad_norm": 2.0710909001992817,
      "language_loss": 0.78876197,
      "learning_rate": 3.7941902744205033e-06,
      "loss": 0.81143379,
      "num_input_tokens_seen": 30107990,
      "step": 1425,
      "time_per_iteration": 2.6807079315185547
    },
    {
      "auxiliary_loss_clip": 0.01232425,
      "auxiliary_loss_mlp": 0.01033435,
      "balance_loss_clip": 1.05703425,
      "balance_loss_mlp": 1.02186632,
      "epoch": 0.17146636205134372,
      "flos": 13953845900160.0,
      "grad_norm": 1.8814730738471037,
      "language_loss": 0.83817172,
      "learning_rate": 3.7938459595096817e-06,
      "loss": 0.86083025,
      "num_input_tokens_seen": 30126535,
      "step": 1426,
      "time_per_iteration": 2.801478385925293
    },
    {
      "auxiliary_loss_clip": 0.01232821,
      "auxiliary_loss_mlp": 0.01034242,
      "balance_loss_clip": 1.05947948,
      "balance_loss_mlp": 1.02372146,
      "epoch": 0.1715866049419828,
      "flos": 23915214172800.0,
      "grad_norm": 2.3308045494890397,
      "language_loss": 0.86552793,
      "learning_rate": 3.7935013724738545e-06,
      "loss": 0.88819849,
      "num_input_tokens_seen": 30147035,
      "step": 1427,
      "time_per_iteration": 2.660015106201172
    },
    {
      "auxiliary_loss_clip": 0.01220501,
      "auxiliary_loss_mlp": 0.01041233,
      "balance_loss_clip": 1.05778432,
      "balance_loss_mlp": 1.03161883,
      "epoch": 0.17170684783262188,
      "flos": 22709369669760.0,
      "grad_norm": 2.4330356975754226,
      "language_loss": 0.77721828,
      "learning_rate": 3.7931565133652945e-06,
      "loss": 0.79983562,
      "num_input_tokens_seen": 30167110,
      "step": 1428,
      "time_per_iteration": 2.684544801712036
    },
    {
      "auxiliary_loss_clip": 0.01224213,
      "auxiliary_loss_mlp": 0.01036577,
      "balance_loss_clip": 1.05946946,
      "balance_loss_mlp": 1.02652216,
      "epoch": 0.171827090723261,
      "flos": 26613290315520.0,
      "grad_norm": 2.2380459844243754,
      "language_loss": 0.67775059,
      "learning_rate": 3.792811382236317e-06,
      "loss": 0.70035845,
      "num_input_tokens_seen": 30185620,
      "step": 1429,
      "time_per_iteration": 2.623973846435547
    },
    {
      "auxiliary_loss_clip": 0.01229814,
      "auxiliary_loss_mlp": 0.0103402,
      "balance_loss_clip": 1.05828631,
      "balance_loss_mlp": 1.02349377,
      "epoch": 0.17194733361390008,
      "flos": 28148507556480.0,
      "grad_norm": 2.6271528002139006,
      "language_loss": 0.78006649,
      "learning_rate": 3.792465979139279e-06,
      "loss": 0.80270481,
      "num_input_tokens_seen": 30208225,
      "step": 1430,
      "time_per_iteration": 2.6998777389526367
    },
    {
      "auxiliary_loss_clip": 0.01140698,
      "auxiliary_loss_mlp": 0.01006924,
      "balance_loss_clip": 1.02573967,
      "balance_loss_mlp": 1.00403881,
      "epoch": 0.17206757650453916,
      "flos": 65530689753600.0,
      "grad_norm": 0.9251386308956856,
      "language_loss": 0.65614545,
      "learning_rate": 3.792120304126576e-06,
      "loss": 0.67762172,
      "num_input_tokens_seen": 30271600,
      "step": 1431,
      "time_per_iteration": 3.2099080085754395
    },
    {
      "auxiliary_loss_clip": 0.01234499,
      "auxiliary_loss_mlp": 0.01026411,
      "balance_loss_clip": 1.05019736,
      "balance_loss_mlp": 1.01708281,
      "epoch": 0.17218781939517827,
      "flos": 22273486128000.0,
      "grad_norm": 2.390252954004473,
      "language_loss": 0.8394599,
      "learning_rate": 3.791774357250649e-06,
      "loss": 0.86206901,
      "num_input_tokens_seen": 30290430,
      "step": 1432,
      "time_per_iteration": 2.790588855743408
    },
    {
      "auxiliary_loss_clip": 0.01222682,
      "auxiliary_loss_mlp": 0.01043129,
      "balance_loss_clip": 1.05439234,
      "balance_loss_mlp": 1.0322932,
      "epoch": 0.17230806228581735,
      "flos": 14137313592960.0,
      "grad_norm": 2.737539466075772,
      "language_loss": 0.79112113,
      "learning_rate": 3.7914281385639757e-06,
      "loss": 0.81377923,
      "num_input_tokens_seen": 30308305,
      "step": 1433,
      "time_per_iteration": 2.638777256011963
    },
    {
      "auxiliary_loss_clip": 0.01223174,
      "auxiliary_loss_mlp": 0.01030025,
      "balance_loss_clip": 1.0541141,
      "balance_loss_mlp": 1.01962388,
      "epoch": 0.17242830517645644,
      "flos": 20704836303360.0,
      "grad_norm": 2.2015697169497392,
      "language_loss": 0.80066109,
      "learning_rate": 3.7910816481190784e-06,
      "loss": 0.82319307,
      "num_input_tokens_seen": 30328120,
      "step": 1434,
      "time_per_iteration": 3.5944695472717285
    },
    {
      "auxiliary_loss_clip": 0.01214194,
      "auxiliary_loss_mlp": 0.01033009,
      "balance_loss_clip": 1.05200195,
      "balance_loss_mlp": 1.02264392,
      "epoch": 0.17254854806709552,
      "flos": 30774582887040.0,
      "grad_norm": 2.633665139256438,
      "language_loss": 0.74649346,
      "learning_rate": 3.7907348859685193e-06,
      "loss": 0.76896548,
      "num_input_tokens_seen": 30349825,
      "step": 1435,
      "time_per_iteration": 3.623326301574707
    },
    {
      "auxiliary_loss_clip": 0.01217844,
      "auxiliary_loss_mlp": 0.0102914,
      "balance_loss_clip": 1.05737567,
      "balance_loss_mlp": 1.01861358,
      "epoch": 0.17266879095773463,
      "flos": 26614726859520.0,
      "grad_norm": 2.2372112936066544,
      "language_loss": 0.81228745,
      "learning_rate": 3.790387852164902e-06,
      "loss": 0.83475721,
      "num_input_tokens_seen": 30370555,
      "step": 1436,
      "time_per_iteration": 2.7536559104919434
    },
    {
      "auxiliary_loss_clip": 0.01223737,
      "auxiliary_loss_mlp": 0.01034154,
      "balance_loss_clip": 1.056458,
      "balance_loss_mlp": 1.02409911,
      "epoch": 0.1727890338483737,
      "flos": 20266295155200.0,
      "grad_norm": 3.8703371059924936,
      "language_loss": 0.7692591,
      "learning_rate": 3.7900405467608707e-06,
      "loss": 0.79183799,
      "num_input_tokens_seen": 30390100,
      "step": 1437,
      "time_per_iteration": 2.611616849899292
    },
    {
      "auxiliary_loss_clip": 0.01214947,
      "auxiliary_loss_mlp": 0.01032504,
      "balance_loss_clip": 1.04755783,
      "balance_loss_mlp": 1.0221386,
      "epoch": 0.1729092767390128,
      "flos": 18179812909440.0,
      "grad_norm": 2.8495024866704544,
      "language_loss": 0.79513705,
      "learning_rate": 3.7896929698091114e-06,
      "loss": 0.81761158,
      "num_input_tokens_seen": 30402915,
      "step": 1438,
      "time_per_iteration": 2.6423420906066895
    },
    {
      "auxiliary_loss_clip": 0.01228279,
      "auxiliary_loss_mlp": 0.0103796,
      "balance_loss_clip": 1.06403518,
      "balance_loss_mlp": 1.02735639,
      "epoch": 0.1730295196296519,
      "flos": 26759518583040.0,
      "grad_norm": 2.7691329481633584,
      "language_loss": 0.6849488,
      "learning_rate": 3.7893451213623518e-06,
      "loss": 0.7076112,
      "num_input_tokens_seen": 30420145,
      "step": 1439,
      "time_per_iteration": 4.531384229660034
    },
    {
      "auxiliary_loss_clip": 0.01226529,
      "auxiliary_loss_mlp": 0.01209591,
      "balance_loss_clip": 1.06072509,
      "balance_loss_mlp": 1.00017953,
      "epoch": 0.173149762520291,
      "flos": 23842531002240.0,
      "grad_norm": 2.1159146639131525,
      "language_loss": 0.82265973,
      "learning_rate": 3.7889970014733606e-06,
      "loss": 0.84702098,
      "num_input_tokens_seen": 30439250,
      "step": 1440,
      "time_per_iteration": 2.662580728530884
    },
    {
      "auxiliary_loss_clip": 0.01212112,
      "auxiliary_loss_mlp": 0.010342,
      "balance_loss_clip": 1.04786229,
      "balance_loss_mlp": 1.02333975,
      "epoch": 0.17327000541093007,
      "flos": 23368186972800.0,
      "grad_norm": 6.176167232764488,
      "language_loss": 0.78164363,
      "learning_rate": 3.7886486101949463e-06,
      "loss": 0.80410683,
      "num_input_tokens_seen": 30460430,
      "step": 1441,
      "time_per_iteration": 2.7337851524353027
    },
    {
      "auxiliary_loss_clip": 0.01218282,
      "auxiliary_loss_mlp": 0.01043017,
      "balance_loss_clip": 1.04928946,
      "balance_loss_mlp": 1.03246129,
      "epoch": 0.17339024830156918,
      "flos": 18221290139520.0,
      "grad_norm": 2.233040701425794,
      "language_loss": 0.88400495,
      "learning_rate": 3.7882999475799594e-06,
      "loss": 0.90661788,
      "num_input_tokens_seen": 30478465,
      "step": 1442,
      "time_per_iteration": 2.702308177947998
    },
    {
      "auxiliary_loss_clip": 0.01213739,
      "auxiliary_loss_mlp": 0.01039316,
      "balance_loss_clip": 1.05288446,
      "balance_loss_mlp": 1.02918315,
      "epoch": 0.17351049119220827,
      "flos": 23332024955520.0,
      "grad_norm": 2.5332759412595363,
      "language_loss": 0.81840396,
      "learning_rate": 3.787951013681293e-06,
      "loss": 0.84093451,
      "num_input_tokens_seen": 30496510,
      "step": 1443,
      "time_per_iteration": 2.7836458683013916
    },
    {
      "auxiliary_loss_clip": 0.01222209,
      "auxiliary_loss_mlp": 0.01038143,
      "balance_loss_clip": 1.05521679,
      "balance_loss_mlp": 1.02683008,
      "epoch": 0.17363073408284735,
      "flos": 23803495896960.0,
      "grad_norm": 2.1694605769076314,
      "language_loss": 0.77790684,
      "learning_rate": 3.787601808551879e-06,
      "loss": 0.80051041,
      "num_input_tokens_seen": 30516325,
      "step": 1444,
      "time_per_iteration": 2.6615941524505615
    },
    {
      "auxiliary_loss_clip": 0.01227783,
      "auxiliary_loss_mlp": 0.01038977,
      "balance_loss_clip": 1.05409265,
      "balance_loss_mlp": 1.02856421,
      "epoch": 0.17375097697348643,
      "flos": 18515290959360.0,
      "grad_norm": 2.3920079130914127,
      "language_loss": 0.84563756,
      "learning_rate": 3.7872523322446926e-06,
      "loss": 0.86830515,
      "num_input_tokens_seen": 30535210,
      "step": 1445,
      "time_per_iteration": 2.671069860458374
    },
    {
      "auxiliary_loss_clip": 0.012333,
      "auxiliary_loss_mlp": 0.01030128,
      "balance_loss_clip": 1.04851151,
      "balance_loss_mlp": 1.02054322,
      "epoch": 0.17387121986412554,
      "flos": 38877897456000.0,
      "grad_norm": 2.2938615342244604,
      "language_loss": 0.60196316,
      "learning_rate": 3.7869025848127478e-06,
      "loss": 0.62459737,
      "num_input_tokens_seen": 30559405,
      "step": 1446,
      "time_per_iteration": 2.8720996379852295
    },
    {
      "auxiliary_loss_clip": 0.01223357,
      "auxiliary_loss_mlp": 0.01033137,
      "balance_loss_clip": 1.05470824,
      "balance_loss_mlp": 1.023314,
      "epoch": 0.17399146275476463,
      "flos": 20375714960640.0,
      "grad_norm": 2.4872126077332513,
      "language_loss": 0.80885398,
      "learning_rate": 3.786552566309102e-06,
      "loss": 0.83141893,
      "num_input_tokens_seen": 30577615,
      "step": 1447,
      "time_per_iteration": 2.6721205711364746
    },
    {
      "auxiliary_loss_clip": 0.01222607,
      "auxiliary_loss_mlp": 0.01208929,
      "balance_loss_clip": 1.05725002,
      "balance_loss_mlp": 1.00009358,
      "epoch": 0.1741117056454037,
      "flos": 19164339763200.0,
      "grad_norm": 2.166267938437742,
      "language_loss": 0.85970116,
      "learning_rate": 3.7862022767868517e-06,
      "loss": 0.88401651,
      "num_input_tokens_seen": 30595205,
      "step": 1448,
      "time_per_iteration": 2.626910924911499
    },
    {
      "auxiliary_loss_clip": 0.01225644,
      "auxiliary_loss_mlp": 0.01038293,
      "balance_loss_clip": 1.05934882,
      "balance_loss_mlp": 1.02837467,
      "epoch": 0.17423194853604282,
      "flos": 25374300537600.0,
      "grad_norm": 2.4884035966454903,
      "language_loss": 0.84885556,
      "learning_rate": 3.7858517162991367e-06,
      "loss": 0.87149489,
      "num_input_tokens_seen": 30615280,
      "step": 1449,
      "time_per_iteration": 2.721564531326294
    },
    {
      "auxiliary_loss_clip": 0.01225604,
      "auxiliary_loss_mlp": 0.01034633,
      "balance_loss_clip": 1.05157042,
      "balance_loss_mlp": 1.02432156,
      "epoch": 0.1743521914266819,
      "flos": 25191874339200.0,
      "grad_norm": 4.19055379993629,
      "language_loss": 0.60813171,
      "learning_rate": 3.7855008848991363e-06,
      "loss": 0.63073409,
      "num_input_tokens_seen": 30633485,
      "step": 1450,
      "time_per_iteration": 2.693681001663208
    },
    {
      "auxiliary_loss_clip": 0.01223013,
      "auxiliary_loss_mlp": 0.01034535,
      "balance_loss_clip": 1.05688775,
      "balance_loss_mlp": 1.02492023,
      "epoch": 0.17447243431732098,
      "flos": 25666577504640.0,
      "grad_norm": 3.30774924792555,
      "language_loss": 0.77685642,
      "learning_rate": 3.7851497826400714e-06,
      "loss": 0.79943192,
      "num_input_tokens_seen": 30653625,
      "step": 1451,
      "time_per_iteration": 2.734485387802124
    },
    {
      "auxiliary_loss_clip": 0.0122486,
      "auxiliary_loss_mlp": 0.0104018,
      "balance_loss_clip": 1.06066072,
      "balance_loss_mlp": 1.0295105,
      "epoch": 0.17459267720796007,
      "flos": 36281950657920.0,
      "grad_norm": 1.838663700367744,
      "language_loss": 0.76098382,
      "learning_rate": 3.7847984095752034e-06,
      "loss": 0.78363419,
      "num_input_tokens_seen": 30677080,
      "step": 1452,
      "time_per_iteration": 2.709620714187622
    },
    {
      "auxiliary_loss_clip": 0.01222126,
      "auxiliary_loss_mlp": 0.01026093,
      "balance_loss_clip": 1.05801153,
      "balance_loss_mlp": 1.01635981,
      "epoch": 0.17471292009859918,
      "flos": 20011113959040.0,
      "grad_norm": 2.484563758856121,
      "language_loss": 0.80794966,
      "learning_rate": 3.784446765757836e-06,
      "loss": 0.83043182,
      "num_input_tokens_seen": 30695725,
      "step": 1453,
      "time_per_iteration": 2.6739706993103027
    },
    {
      "auxiliary_loss_clip": 0.01209239,
      "auxiliary_loss_mlp": 0.01031926,
      "balance_loss_clip": 1.05083847,
      "balance_loss_mlp": 1.02170372,
      "epoch": 0.17483316298923826,
      "flos": 27819242559360.0,
      "grad_norm": 3.308350763591608,
      "language_loss": 0.77749014,
      "learning_rate": 3.7840948512413133e-06,
      "loss": 0.79990184,
      "num_input_tokens_seen": 30713310,
      "step": 1454,
      "time_per_iteration": 2.7171661853790283
    },
    {
      "auxiliary_loss_clip": 0.01223519,
      "auxiliary_loss_mlp": 0.01033588,
      "balance_loss_clip": 1.0555141,
      "balance_loss_mlp": 1.02254927,
      "epoch": 0.17495340587987734,
      "flos": 44017934791680.0,
      "grad_norm": 1.7199633329969748,
      "language_loss": 0.79223269,
      "learning_rate": 3.7837426660790196e-06,
      "loss": 0.81480378,
      "num_input_tokens_seen": 30734725,
      "step": 1455,
      "time_per_iteration": 2.866258144378662
    },
    {
      "auxiliary_loss_clip": 0.01218846,
      "auxiliary_loss_mlp": 0.01043455,
      "balance_loss_clip": 1.05756044,
      "balance_loss_mlp": 1.0336802,
      "epoch": 0.17507364877051645,
      "flos": 20885825957760.0,
      "grad_norm": 12.398770585140586,
      "language_loss": 0.82265222,
      "learning_rate": 3.783390210324382e-06,
      "loss": 0.84527528,
      "num_input_tokens_seen": 30754450,
      "step": 1456,
      "time_per_iteration": 2.6537094116210938
    },
    {
      "auxiliary_loss_clip": 0.01225248,
      "auxiliary_loss_mlp": 0.01034251,
      "balance_loss_clip": 1.05523634,
      "balance_loss_mlp": 1.02442801,
      "epoch": 0.17519389166115554,
      "flos": 24717602136960.0,
      "grad_norm": 1.7800958909428386,
      "language_loss": 0.7292459,
      "learning_rate": 3.7830374840308676e-06,
      "loss": 0.75184083,
      "num_input_tokens_seen": 30774605,
      "step": 1457,
      "time_per_iteration": 2.6572656631469727
    },
    {
      "auxiliary_loss_clip": 0.0122491,
      "auxiliary_loss_mlp": 0.01038055,
      "balance_loss_clip": 1.05810487,
      "balance_loss_mlp": 1.02760005,
      "epoch": 0.17531413455179462,
      "flos": 23798144770560.0,
      "grad_norm": 2.304073841282013,
      "language_loss": 0.82864302,
      "learning_rate": 3.7826844872519842e-06,
      "loss": 0.85127264,
      "num_input_tokens_seen": 30792460,
      "step": 1458,
      "time_per_iteration": 2.668769121170044
    },
    {
      "auxiliary_loss_clip": 0.01222708,
      "auxiliary_loss_mlp": 0.01032984,
      "balance_loss_clip": 1.05665207,
      "balance_loss_mlp": 1.02370369,
      "epoch": 0.1754343774424337,
      "flos": 24572379450240.0,
      "grad_norm": 2.0807693287063653,
      "language_loss": 0.72365773,
      "learning_rate": 3.782331220041282e-06,
      "loss": 0.74621463,
      "num_input_tokens_seen": 30812525,
      "step": 1459,
      "time_per_iteration": 2.701683521270752
    },
    {
      "auxiliary_loss_clip": 0.01233772,
      "auxiliary_loss_mlp": 0.01036311,
      "balance_loss_clip": 1.05363286,
      "balance_loss_mlp": 1.02592778,
      "epoch": 0.17555462033307281,
      "flos": 18114599767680.0,
      "grad_norm": 2.0490230575878234,
      "language_loss": 0.82535142,
      "learning_rate": 3.7819776824523504e-06,
      "loss": 0.84805226,
      "num_input_tokens_seen": 30830390,
      "step": 1460,
      "time_per_iteration": 3.5582082271575928
    },
    {
      "auxiliary_loss_clip": 0.0123275,
      "auxiliary_loss_mlp": 0.01036534,
      "balance_loss_clip": 1.0552994,
      "balance_loss_mlp": 1.02665746,
      "epoch": 0.1756748632237119,
      "flos": 28366018364160.0,
      "grad_norm": 2.525008880779874,
      "language_loss": 0.84211516,
      "learning_rate": 3.7816238745388213e-06,
      "loss": 0.86480796,
      "num_input_tokens_seen": 30849935,
      "step": 1461,
      "time_per_iteration": 2.708664894104004
    },
    {
      "auxiliary_loss_clip": 0.0122978,
      "auxiliary_loss_mlp": 0.01029605,
      "balance_loss_clip": 1.05277848,
      "balance_loss_mlp": 1.0199492,
      "epoch": 0.17579510611435098,
      "flos": 25732939881600.0,
      "grad_norm": 2.781154369805199,
      "language_loss": 0.87068939,
      "learning_rate": 3.781269796354367e-06,
      "loss": 0.89328325,
      "num_input_tokens_seen": 30869555,
      "step": 1462,
      "time_per_iteration": 3.6268205642700195
    },
    {
      "auxiliary_loss_clip": 0.01226553,
      "auxiliary_loss_mlp": 0.0103714,
      "balance_loss_clip": 1.05606067,
      "balance_loss_mlp": 1.02703714,
      "epoch": 0.1759153490049901,
      "flos": 18588081870720.0,
      "grad_norm": 2.1265164012764872,
      "language_loss": 0.86423403,
      "learning_rate": 3.7809154479527006e-06,
      "loss": 0.88687098,
      "num_input_tokens_seen": 30888760,
      "step": 1463,
      "time_per_iteration": 2.6926820278167725
    },
    {
      "auxiliary_loss_clip": 0.01217244,
      "auxiliary_loss_mlp": 0.01029349,
      "balance_loss_clip": 1.05217814,
      "balance_loss_mlp": 1.01990795,
      "epoch": 0.17603559189562917,
      "flos": 18619323724800.0,
      "grad_norm": 2.6589097348984705,
      "language_loss": 0.84413409,
      "learning_rate": 3.780560829387577e-06,
      "loss": 0.86660004,
      "num_input_tokens_seen": 30907260,
      "step": 1464,
      "time_per_iteration": 2.6608150005340576
    },
    {
      "auxiliary_loss_clip": 0.01127732,
      "auxiliary_loss_mlp": 0.01002318,
      "balance_loss_clip": 1.02492929,
      "balance_loss_mlp": 0.99955237,
      "epoch": 0.17615583478626826,
      "flos": 60530775373440.0,
      "grad_norm": 0.9050328529775434,
      "language_loss": 0.57954031,
      "learning_rate": 3.7802059407127915e-06,
      "loss": 0.60084081,
      "num_input_tokens_seen": 30965810,
      "step": 1465,
      "time_per_iteration": 4.071656942367554
    },
    {
      "auxiliary_loss_clip": 0.01218378,
      "auxiliary_loss_mlp": 0.01041184,
      "balance_loss_clip": 1.05172467,
      "balance_loss_mlp": 1.03127182,
      "epoch": 0.17627607767690734,
      "flos": 23616221362560.0,
      "grad_norm": 2.0350990201693486,
      "language_loss": 0.86306524,
      "learning_rate": 3.7798507819821797e-06,
      "loss": 0.88566089,
      "num_input_tokens_seen": 30982935,
      "step": 1466,
      "time_per_iteration": 3.7637882232666016
    },
    {
      "auxiliary_loss_clip": 0.01221812,
      "auxiliary_loss_mlp": 0.01037939,
      "balance_loss_clip": 1.05497956,
      "balance_loss_mlp": 1.0275079,
      "epoch": 0.17639632056754645,
      "flos": 17639070589440.0,
      "grad_norm": 7.358318463373974,
      "language_loss": 0.7914809,
      "learning_rate": 3.7794953532496197e-06,
      "loss": 0.81407845,
      "num_input_tokens_seen": 30998840,
      "step": 1467,
      "time_per_iteration": 2.6590840816497803
    },
    {
      "auxiliary_loss_clip": 0.01134076,
      "auxiliary_loss_mlp": 0.01201096,
      "balance_loss_clip": 1.02484059,
      "balance_loss_mlp": 0.99983519,
      "epoch": 0.17651656345818553,
      "flos": 57932604910080.0,
      "grad_norm": 0.8948071401507216,
      "language_loss": 0.57914662,
      "learning_rate": 3.7791396545690295e-06,
      "loss": 0.60249835,
      "num_input_tokens_seen": 31060075,
      "step": 1468,
      "time_per_iteration": 3.3174526691436768
    },
    {
      "auxiliary_loss_clip": 0.01226711,
      "auxiliary_loss_mlp": 0.01034615,
      "balance_loss_clip": 1.06123054,
      "balance_loss_mlp": 1.02462482,
      "epoch": 0.17663680634882462,
      "flos": 22929502170240.0,
      "grad_norm": 4.0998054874194985,
      "language_loss": 0.80731833,
      "learning_rate": 3.7787836859943685e-06,
      "loss": 0.82993162,
      "num_input_tokens_seen": 31078800,
      "step": 1469,
      "time_per_iteration": 2.8421380519866943
    },
    {
      "auxiliary_loss_clip": 0.01225341,
      "auxiliary_loss_mlp": 0.01033129,
      "balance_loss_clip": 1.05860448,
      "balance_loss_mlp": 1.02263856,
      "epoch": 0.17675704923946373,
      "flos": 22637979388800.0,
      "grad_norm": 2.8039066834560833,
      "language_loss": 0.7956391,
      "learning_rate": 3.7784274475796363e-06,
      "loss": 0.81822383,
      "num_input_tokens_seen": 31097430,
      "step": 1470,
      "time_per_iteration": 2.623544454574585
    },
    {
      "auxiliary_loss_clip": 0.01228664,
      "auxiliary_loss_mlp": 0.01035628,
      "balance_loss_clip": 1.05320787,
      "balance_loss_mlp": 1.02551866,
      "epoch": 0.1768772921301028,
      "flos": 27126525795840.0,
      "grad_norm": 2.015580934936408,
      "language_loss": 0.76263011,
      "learning_rate": 3.7780709393788745e-06,
      "loss": 0.78527308,
      "num_input_tokens_seen": 31117905,
      "step": 1471,
      "time_per_iteration": 2.7637977600097656
    },
    {
      "auxiliary_loss_clip": 0.01222352,
      "auxiliary_loss_mlp": 0.01032632,
      "balance_loss_clip": 1.05959451,
      "balance_loss_mlp": 1.02191496,
      "epoch": 0.1769975350207419,
      "flos": 19172133014400.0,
      "grad_norm": 1.9663547913746904,
      "language_loss": 0.75672543,
      "learning_rate": 3.777714161446165e-06,
      "loss": 0.77927518,
      "num_input_tokens_seen": 31137610,
      "step": 1472,
      "time_per_iteration": 2.609264612197876
    },
    {
      "auxiliary_loss_clip": 0.01225215,
      "auxiliary_loss_mlp": 0.0103003,
      "balance_loss_clip": 1.05889595,
      "balance_loss_mlp": 1.01952207,
      "epoch": 0.177117777911381,
      "flos": 36134932291200.0,
      "grad_norm": 2.2849651037213947,
      "language_loss": 0.69272447,
      "learning_rate": 3.7773571138356304e-06,
      "loss": 0.7152769,
      "num_input_tokens_seen": 31157780,
      "step": 1473,
      "time_per_iteration": 2.770561456680298
    },
    {
      "auxiliary_loss_clip": 0.01213589,
      "auxiliary_loss_mlp": 0.0102675,
      "balance_loss_clip": 1.04938364,
      "balance_loss_mlp": 1.01715946,
      "epoch": 0.17723802080202009,
      "flos": 22090593052800.0,
      "grad_norm": 5.32406007699771,
      "language_loss": 0.89489436,
      "learning_rate": 3.776999796601435e-06,
      "loss": 0.91729772,
      "num_input_tokens_seen": 31176540,
      "step": 1474,
      "time_per_iteration": 2.7589545249938965
    },
    {
      "auxiliary_loss_clip": 0.01230214,
      "auxiliary_loss_mlp": 0.01034775,
      "balance_loss_clip": 1.05825114,
      "balance_loss_mlp": 1.02415311,
      "epoch": 0.17735826369265917,
      "flos": 30222671437440.0,
      "grad_norm": 2.7149060726039753,
      "language_loss": 0.72806096,
      "learning_rate": 3.776642209797783e-06,
      "loss": 0.75071084,
      "num_input_tokens_seen": 31198370,
      "step": 1475,
      "time_per_iteration": 2.771339178085327
    },
    {
      "auxiliary_loss_clip": 0.01220094,
      "auxiliary_loss_mlp": 0.01031316,
      "balance_loss_clip": 1.05504954,
      "balance_loss_mlp": 1.0199734,
      "epoch": 0.17747850658329825,
      "flos": 21397588980480.0,
      "grad_norm": 1.9622178543525086,
      "language_loss": 0.78342915,
      "learning_rate": 3.7762843534789205e-06,
      "loss": 0.80594325,
      "num_input_tokens_seen": 31217120,
      "step": 1476,
      "time_per_iteration": 2.6650753021240234
    },
    {
      "auxiliary_loss_clip": 0.01231227,
      "auxiliary_loss_mlp": 0.01033972,
      "balance_loss_clip": 1.05527306,
      "balance_loss_mlp": 1.02342176,
      "epoch": 0.17759874947393736,
      "flos": 16983341856000.0,
      "grad_norm": 2.65302336531708,
      "language_loss": 0.88624769,
      "learning_rate": 3.7759262276991343e-06,
      "loss": 0.90889966,
      "num_input_tokens_seen": 31234730,
      "step": 1477,
      "time_per_iteration": 2.707862138748169
    },
    {
      "auxiliary_loss_clip": 0.0123406,
      "auxiliary_loss_mlp": 0.0102993,
      "balance_loss_clip": 1.05773091,
      "balance_loss_mlp": 1.01902282,
      "epoch": 0.17771899236457644,
      "flos": 11546107390080.0,
      "grad_norm": 2.813671475663223,
      "language_loss": 0.80546308,
      "learning_rate": 3.7755678325127506e-06,
      "loss": 0.82810307,
      "num_input_tokens_seen": 31252410,
      "step": 1478,
      "time_per_iteration": 2.6655478477478027
    },
    {
      "auxiliary_loss_clip": 0.01226011,
      "auxiliary_loss_mlp": 0.01031464,
      "balance_loss_clip": 1.05567515,
      "balance_loss_mlp": 1.02079439,
      "epoch": 0.17783923525521553,
      "flos": 18807747494400.0,
      "grad_norm": 3.3072116168287824,
      "language_loss": 0.75804502,
      "learning_rate": 3.7752091679741393e-06,
      "loss": 0.7806198,
      "num_input_tokens_seen": 31270200,
      "step": 1479,
      "time_per_iteration": 2.6601688861846924
    },
    {
      "auxiliary_loss_clip": 0.01224084,
      "auxiliary_loss_mlp": 0.0103182,
      "balance_loss_clip": 1.05741918,
      "balance_loss_mlp": 1.02091193,
      "epoch": 0.17795947814585464,
      "flos": 30408365773440.0,
      "grad_norm": 3.1101748353699987,
      "language_loss": 0.77833056,
      "learning_rate": 3.774850234137708e-06,
      "loss": 0.80088961,
      "num_input_tokens_seen": 31287495,
      "step": 1480,
      "time_per_iteration": 2.7273006439208984
    },
    {
      "auxiliary_loss_clip": 0.01221617,
      "auxiliary_loss_mlp": 0.01035368,
      "balance_loss_clip": 1.05601716,
      "balance_loss_mlp": 1.02479434,
      "epoch": 0.17807972103649372,
      "flos": 24389055411840.0,
      "grad_norm": 2.3560588066018933,
      "language_loss": 0.82939231,
      "learning_rate": 3.7744910310579076e-06,
      "loss": 0.85196221,
      "num_input_tokens_seen": 31306420,
      "step": 1481,
      "time_per_iteration": 2.6570003032684326
    },
    {
      "auxiliary_loss_clip": 0.01222211,
      "auxiliary_loss_mlp": 0.01028475,
      "balance_loss_clip": 1.06004691,
      "balance_loss_mlp": 1.01880682,
      "epoch": 0.1781999639271328,
      "flos": 20301559332480.0,
      "grad_norm": 2.3554068487672093,
      "language_loss": 0.85789812,
      "learning_rate": 3.774131558789229e-06,
      "loss": 0.88040501,
      "num_input_tokens_seen": 31325750,
      "step": 1482,
      "time_per_iteration": 2.7447943687438965
    },
    {
      "auxiliary_loss_clip": 0.01224615,
      "auxiliary_loss_mlp": 0.01209881,
      "balance_loss_clip": 1.06052637,
      "balance_loss_mlp": 1.00030112,
      "epoch": 0.1783202068177719,
      "flos": 15924479806080.0,
      "grad_norm": 2.516445362335944,
      "language_loss": 0.69803882,
      "learning_rate": 3.773771817386203e-06,
      "loss": 0.72238386,
      "num_input_tokens_seen": 31343080,
      "step": 1483,
      "time_per_iteration": 2.5641062259674072
    },
    {
      "auxiliary_loss_clip": 0.01224943,
      "auxiliary_loss_mlp": 0.0103163,
      "balance_loss_clip": 1.05516922,
      "balance_loss_mlp": 1.02154446,
      "epoch": 0.178440449708411,
      "flos": 20631758083200.0,
      "grad_norm": 1.7077644798516696,
      "language_loss": 0.79594892,
      "learning_rate": 3.773411806903403e-06,
      "loss": 0.81851465,
      "num_input_tokens_seen": 31362160,
      "step": 1484,
      "time_per_iteration": 2.68406343460083
    },
    {
      "auxiliary_loss_clip": 0.01234702,
      "auxiliary_loss_mlp": 0.01037361,
      "balance_loss_clip": 1.05026782,
      "balance_loss_mlp": 1.0259409,
      "epoch": 0.17856069259905008,
      "flos": 21686059105920.0,
      "grad_norm": 2.383621877282538,
      "language_loss": 0.94943368,
      "learning_rate": 3.7730515273954415e-06,
      "loss": 0.97215426,
      "num_input_tokens_seen": 31380770,
      "step": 1485,
      "time_per_iteration": 2.752227544784546
    },
    {
      "auxiliary_loss_clip": 0.0122407,
      "auxiliary_loss_mlp": 0.01033438,
      "balance_loss_clip": 1.06114542,
      "balance_loss_mlp": 1.02355015,
      "epoch": 0.17868093548968916,
      "flos": 26572962320640.0,
      "grad_norm": 1.7624413872220068,
      "language_loss": 0.85295868,
      "learning_rate": 3.772690978916973e-06,
      "loss": 0.8755337,
      "num_input_tokens_seen": 31400525,
      "step": 1486,
      "time_per_iteration": 2.7792584896087646
    },
    {
      "auxiliary_loss_clip": 0.01225723,
      "auxiliary_loss_mlp": 0.010357,
      "balance_loss_clip": 1.05923653,
      "balance_loss_mlp": 1.02482247,
      "epoch": 0.17880117838032827,
      "flos": 18581006891520.0,
      "grad_norm": 2.5519970680971698,
      "language_loss": 0.86788285,
      "learning_rate": 3.772330161522693e-06,
      "loss": 0.89049709,
      "num_input_tokens_seen": 31418435,
      "step": 1487,
      "time_per_iteration": 3.5345489978790283
    },
    {
      "auxiliary_loss_clip": 0.01225502,
      "auxiliary_loss_mlp": 0.01036589,
      "balance_loss_clip": 1.05975866,
      "balance_loss_mlp": 1.02592587,
      "epoch": 0.17892142127096736,
      "flos": 26541217676160.0,
      "grad_norm": 2.0964591552754803,
      "language_loss": 0.80043268,
      "learning_rate": 3.7719690752673365e-06,
      "loss": 0.8230536,
      "num_input_tokens_seen": 31439230,
      "step": 1488,
      "time_per_iteration": 2.7330360412597656
    },
    {
      "auxiliary_loss_clip": 0.01231696,
      "auxiliary_loss_mlp": 0.01033664,
      "balance_loss_clip": 1.05688369,
      "balance_loss_mlp": 1.02331102,
      "epoch": 0.17904166416160644,
      "flos": 23872623621120.0,
      "grad_norm": 1.965483439077758,
      "language_loss": 0.7861917,
      "learning_rate": 3.7716077202056796e-06,
      "loss": 0.80884528,
      "num_input_tokens_seen": 31457705,
      "step": 1489,
      "time_per_iteration": 3.661832094192505
    },
    {
      "auxiliary_loss_clip": 0.01212988,
      "auxiliary_loss_mlp": 0.01032136,
      "balance_loss_clip": 1.05434513,
      "balance_loss_mlp": 1.02215195,
      "epoch": 0.17916190705224552,
      "flos": 19134426712320.0,
      "grad_norm": 2.4218990187187166,
      "language_loss": 0.93684459,
      "learning_rate": 3.7712460963925404e-06,
      "loss": 0.95929587,
      "num_input_tokens_seen": 31473645,
      "step": 1490,
      "time_per_iteration": 2.6236703395843506
    },
    {
      "auxiliary_loss_clip": 0.01218626,
      "auxiliary_loss_mlp": 0.01034246,
      "balance_loss_clip": 1.05331242,
      "balance_loss_mlp": 1.02361846,
      "epoch": 0.17928214994288463,
      "flos": 25152120961920.0,
      "grad_norm": 1.9485746284180576,
      "language_loss": 0.75456476,
      "learning_rate": 3.7708842038827775e-06,
      "loss": 0.77709353,
      "num_input_tokens_seen": 31492605,
      "step": 1491,
      "time_per_iteration": 2.747481346130371
    },
    {
      "auxiliary_loss_clip": 0.01223624,
      "auxiliary_loss_mlp": 0.01036313,
      "balance_loss_clip": 1.05565763,
      "balance_loss_mlp": 1.02644205,
      "epoch": 0.17940239283352372,
      "flos": 22384629786240.0,
      "grad_norm": 1.9882316025593263,
      "language_loss": 0.86107457,
      "learning_rate": 3.770522042731288e-06,
      "loss": 0.88367391,
      "num_input_tokens_seen": 31514500,
      "step": 1492,
      "time_per_iteration": 3.5690009593963623
    },
    {
      "auxiliary_loss_clip": 0.01222273,
      "auxiliary_loss_mlp": 0.01039849,
      "balance_loss_clip": 1.05356407,
      "balance_loss_mlp": 1.02889359,
      "epoch": 0.1795226357241628,
      "flos": 23178685795200.0,
      "grad_norm": 2.279253208760056,
      "language_loss": 0.87694937,
      "learning_rate": 3.7701596129930122e-06,
      "loss": 0.89957058,
      "num_input_tokens_seen": 31533225,
      "step": 1493,
      "time_per_iteration": 3.657083511352539
    },
    {
      "auxiliary_loss_clip": 0.01236448,
      "auxiliary_loss_mlp": 0.01029298,
      "balance_loss_clip": 1.05609536,
      "balance_loss_mlp": 1.01781189,
      "epoch": 0.1796428786148019,
      "flos": 22090413484800.0,
      "grad_norm": 1.9850295128729867,
      "language_loss": 0.74017805,
      "learning_rate": 3.7697969147229315e-06,
      "loss": 0.7628355,
      "num_input_tokens_seen": 31551385,
      "step": 1494,
      "time_per_iteration": 2.707972288131714
    },
    {
      "auxiliary_loss_clip": 0.01222471,
      "auxiliary_loss_mlp": 0.01033993,
      "balance_loss_clip": 1.05680966,
      "balance_loss_mlp": 1.02355599,
      "epoch": 0.179763121505441,
      "flos": 21324618501120.0,
      "grad_norm": 1.9126294966334327,
      "language_loss": 0.85231334,
      "learning_rate": 3.7694339479760647e-06,
      "loss": 0.87487793,
      "num_input_tokens_seen": 31570415,
      "step": 1495,
      "time_per_iteration": 2.624258041381836
    },
    {
      "auxiliary_loss_clip": 0.01129297,
      "auxiliary_loss_mlp": 0.01012331,
      "balance_loss_clip": 1.0227015,
      "balance_loss_mlp": 1.0098033,
      "epoch": 0.17988336439608008,
      "flos": 68161864815360.0,
      "grad_norm": 0.8067632225606437,
      "language_loss": 0.57308513,
      "learning_rate": 3.769070712807476e-06,
      "loss": 0.59450144,
      "num_input_tokens_seen": 31632445,
      "step": 1496,
      "time_per_iteration": 3.24920654296875
    },
    {
      "auxiliary_loss_clip": 0.01219374,
      "auxiliary_loss_mlp": 0.01036581,
      "balance_loss_clip": 1.05101824,
      "balance_loss_mlp": 1.02576292,
      "epoch": 0.18000360728671919,
      "flos": 21945047143680.0,
      "grad_norm": 2.9920313497324242,
      "language_loss": 0.79173911,
      "learning_rate": 3.768707209272266e-06,
      "loss": 0.81429869,
      "num_input_tokens_seen": 31652575,
      "step": 1497,
      "time_per_iteration": 2.780398368835449
    },
    {
      "auxiliary_loss_clip": 0.01221767,
      "auxiliary_loss_mlp": 0.01034842,
      "balance_loss_clip": 1.053936,
      "balance_loss_mlp": 1.02408898,
      "epoch": 0.18012385017735827,
      "flos": 18986330937600.0,
      "grad_norm": 2.462548768311989,
      "language_loss": 0.76428962,
      "learning_rate": 3.768343437425579e-06,
      "loss": 0.7868557,
      "num_input_tokens_seen": 31671145,
      "step": 1498,
      "time_per_iteration": 2.659527540206909
    },
    {
      "auxiliary_loss_clip": 0.01227296,
      "auxiliary_loss_mlp": 0.01032914,
      "balance_loss_clip": 1.05008197,
      "balance_loss_mlp": 1.02216101,
      "epoch": 0.18024409306799735,
      "flos": 19748103598080.0,
      "grad_norm": 2.8814796960735314,
      "language_loss": 0.86514527,
      "learning_rate": 3.7679793973225987e-06,
      "loss": 0.88774729,
      "num_input_tokens_seen": 31686955,
      "step": 1499,
      "time_per_iteration": 2.752941131591797
    },
    {
      "auxiliary_loss_clip": 0.01129856,
      "auxiliary_loss_mlp": 0.01005573,
      "balance_loss_clip": 1.01911902,
      "balance_loss_mlp": 1.00304627,
      "epoch": 0.18036433595863643,
      "flos": 67227183060480.0,
      "grad_norm": 0.8548869524576149,
      "language_loss": 0.61589849,
      "learning_rate": 3.767615089018549e-06,
      "loss": 0.63725281,
      "num_input_tokens_seen": 31749300,
      "step": 1500,
      "time_per_iteration": 3.1678006649017334
    },
    {
      "auxiliary_loss_clip": 0.01224703,
      "auxiliary_loss_mlp": 0.01039612,
      "balance_loss_clip": 1.05762315,
      "balance_loss_mlp": 1.02873445,
      "epoch": 0.18048457884927555,
      "flos": 18181464935040.0,
      "grad_norm": 2.708136241380943,
      "language_loss": 0.86143649,
      "learning_rate": 3.7672505125686966e-06,
      "loss": 0.88407958,
      "num_input_tokens_seen": 31765665,
      "step": 1501,
      "time_per_iteration": 2.6790454387664795
    },
    {
      "auxiliary_loss_clip": 0.01228663,
      "auxiliary_loss_mlp": 0.0103463,
      "balance_loss_clip": 1.04946351,
      "balance_loss_mlp": 1.02447331,
      "epoch": 0.18060482173991463,
      "flos": 15813767111040.0,
      "grad_norm": 16.788902713689378,
      "language_loss": 0.85131049,
      "learning_rate": 3.7668856680283455e-06,
      "loss": 0.87394345,
      "num_input_tokens_seen": 31782690,
      "step": 1502,
      "time_per_iteration": 2.679142951965332
    },
    {
      "auxiliary_loss_clip": 0.01232373,
      "auxiliary_loss_mlp": 0.01036719,
      "balance_loss_clip": 1.05744696,
      "balance_loss_mlp": 1.02635384,
      "epoch": 0.1807250646305537,
      "flos": 18587399512320.0,
      "grad_norm": 2.630611602506268,
      "language_loss": 0.82426101,
      "learning_rate": 3.7665205554528437e-06,
      "loss": 0.84695196,
      "num_input_tokens_seen": 31802045,
      "step": 1503,
      "time_per_iteration": 2.671821355819702
    },
    {
      "auxiliary_loss_clip": 0.0122939,
      "auxiliary_loss_mlp": 0.01032448,
      "balance_loss_clip": 1.06002831,
      "balance_loss_mlp": 1.02247643,
      "epoch": 0.18084530752119282,
      "flos": 23149131880320.0,
      "grad_norm": 1.6382848244957007,
      "language_loss": 0.74157012,
      "learning_rate": 3.7661551748975782e-06,
      "loss": 0.76418853,
      "num_input_tokens_seen": 31820220,
      "step": 1504,
      "time_per_iteration": 2.7095582485198975
    },
    {
      "auxiliary_loss_clip": 0.01130445,
      "auxiliary_loss_mlp": 0.01003193,
      "balance_loss_clip": 1.02296019,
      "balance_loss_mlp": 1.00078487,
      "epoch": 0.1809655504118319,
      "flos": 59803153568640.0,
      "grad_norm": 0.8169653618443995,
      "language_loss": 0.6053949,
      "learning_rate": 3.7657895264179772e-06,
      "loss": 0.62673122,
      "num_input_tokens_seen": 31876195,
      "step": 1505,
      "time_per_iteration": 3.191005229949951
    },
    {
      "auxiliary_loss_clip": 0.01215886,
      "auxiliary_loss_mlp": 0.01033045,
      "balance_loss_clip": 1.05174959,
      "balance_loss_mlp": 1.02326989,
      "epoch": 0.181085793302471,
      "flos": 44201941188480.0,
      "grad_norm": 1.997008978582065,
      "language_loss": 0.74469024,
      "learning_rate": 3.765423610069509e-06,
      "loss": 0.76717955,
      "num_input_tokens_seen": 31901585,
      "step": 1506,
      "time_per_iteration": 2.8691248893737793
    },
    {
      "auxiliary_loss_clip": 0.0122621,
      "auxiliary_loss_mlp": 0.01037355,
      "balance_loss_clip": 1.05748153,
      "balance_loss_mlp": 1.02700782,
      "epoch": 0.18120603619311007,
      "flos": 34898384638080.0,
      "grad_norm": 2.397835256441917,
      "language_loss": 0.72486126,
      "learning_rate": 3.765057425907683e-06,
      "loss": 0.74749696,
      "num_input_tokens_seen": 31923045,
      "step": 1507,
      "time_per_iteration": 2.7677395343780518
    },
    {
      "auxiliary_loss_clip": 0.01225558,
      "auxiliary_loss_mlp": 0.01036713,
      "balance_loss_clip": 1.05582571,
      "balance_loss_mlp": 1.02598464,
      "epoch": 0.18132627908374918,
      "flos": 21506757390720.0,
      "grad_norm": 2.563053934066331,
      "language_loss": 0.78594315,
      "learning_rate": 3.764690973988048e-06,
      "loss": 0.80856586,
      "num_input_tokens_seen": 31943385,
      "step": 1508,
      "time_per_iteration": 2.674414873123169
    },
    {
      "auxiliary_loss_clip": 0.01229639,
      "auxiliary_loss_mlp": 0.01031465,
      "balance_loss_clip": 1.05469275,
      "balance_loss_mlp": 1.02149868,
      "epoch": 0.18144652197438826,
      "flos": 29057693633280.0,
      "grad_norm": 2.06427777395998,
      "language_loss": 0.73947555,
      "learning_rate": 3.7643242543661967e-06,
      "loss": 0.76208663,
      "num_input_tokens_seen": 31966045,
      "step": 1509,
      "time_per_iteration": 2.72343111038208
    },
    {
      "auxiliary_loss_clip": 0.01119915,
      "auxiliary_loss_mlp": 0.01001654,
      "balance_loss_clip": 1.02046692,
      "balance_loss_mlp": 0.99907935,
      "epoch": 0.18156676486502735,
      "flos": 68675064382080.0,
      "grad_norm": 0.8178067060631041,
      "language_loss": 0.60489368,
      "learning_rate": 3.7639572670977573e-06,
      "loss": 0.62610942,
      "num_input_tokens_seen": 32021540,
      "step": 1510,
      "time_per_iteration": 3.1515631675720215
    },
    {
      "auxiliary_loss_clip": 0.01227968,
      "auxiliary_loss_mlp": 0.01036496,
      "balance_loss_clip": 1.0522573,
      "balance_loss_mlp": 1.026649,
      "epoch": 0.18168700775566646,
      "flos": 26471515334400.0,
      "grad_norm": 1.6366213065160249,
      "language_loss": 0.76815724,
      "learning_rate": 3.7635900122384042e-06,
      "loss": 0.79080194,
      "num_input_tokens_seen": 32044535,
      "step": 1511,
      "time_per_iteration": 2.7506113052368164
    },
    {
      "auxiliary_loss_clip": 0.01225711,
      "auxiliary_loss_mlp": 0.01039204,
      "balance_loss_clip": 1.05304599,
      "balance_loss_mlp": 1.02792704,
      "epoch": 0.18180725064630554,
      "flos": 15005668884480.0,
      "grad_norm": 2.321446155745943,
      "language_loss": 0.87502825,
      "learning_rate": 3.7632224898438477e-06,
      "loss": 0.89767742,
      "num_input_tokens_seen": 32061010,
      "step": 1512,
      "time_per_iteration": 2.598966121673584
    },
    {
      "auxiliary_loss_clip": 0.01229027,
      "auxiliary_loss_mlp": 0.01035058,
      "balance_loss_clip": 1.0507195,
      "balance_loss_mlp": 1.02536595,
      "epoch": 0.18192749353694462,
      "flos": 19682387665920.0,
      "grad_norm": 1.6625875498311837,
      "language_loss": 0.79157066,
      "learning_rate": 3.762854699969842e-06,
      "loss": 0.81421149,
      "num_input_tokens_seen": 32081520,
      "step": 1513,
      "time_per_iteration": 2.7043604850769043
    },
    {
      "auxiliary_loss_clip": 0.01222228,
      "auxiliary_loss_mlp": 0.01041428,
      "balance_loss_clip": 1.05890846,
      "balance_loss_mlp": 1.03022194,
      "epoch": 0.1820477364275837,
      "flos": 20702717400960.0,
      "grad_norm": 2.042928578580492,
      "language_loss": 0.7342667,
      "learning_rate": 3.762486642672179e-06,
      "loss": 0.75690323,
      "num_input_tokens_seen": 32098460,
      "step": 1514,
      "time_per_iteration": 3.5577962398529053
    },
    {
      "auxiliary_loss_clip": 0.01222792,
      "auxiliary_loss_mlp": 0.01032146,
      "balance_loss_clip": 1.05379319,
      "balance_loss_mlp": 1.02178097,
      "epoch": 0.18216797931822282,
      "flos": 17128708197120.0,
      "grad_norm": 3.1110206271784957,
      "language_loss": 0.86958939,
      "learning_rate": 3.7621183180066946e-06,
      "loss": 0.89213878,
      "num_input_tokens_seen": 32116420,
      "step": 1515,
      "time_per_iteration": 2.623222827911377
    },
    {
      "auxiliary_loss_clip": 0.01224182,
      "auxiliary_loss_mlp": 0.01033889,
      "balance_loss_clip": 1.05364585,
      "balance_loss_mlp": 1.0237143,
      "epoch": 0.1822882222088619,
      "flos": 29242561956480.0,
      "grad_norm": 1.6350216503599275,
      "language_loss": 0.73894465,
      "learning_rate": 3.7617497260292625e-06,
      "loss": 0.76152539,
      "num_input_tokens_seen": 32138475,
      "step": 1516,
      "time_per_iteration": 3.634840488433838
    },
    {
      "auxiliary_loss_clip": 0.01218236,
      "auxiliary_loss_mlp": 0.01033678,
      "balance_loss_clip": 1.05530679,
      "balance_loss_mlp": 1.0228126,
      "epoch": 0.18240846509950098,
      "flos": 17702739446400.0,
      "grad_norm": 2.779911341470086,
      "language_loss": 0.79126775,
      "learning_rate": 3.7613808667957967e-06,
      "loss": 0.81378686,
      "num_input_tokens_seen": 32151165,
      "step": 1517,
      "time_per_iteration": 2.6311464309692383
    },
    {
      "auxiliary_loss_clip": 0.0122595,
      "auxiliary_loss_mlp": 0.01039319,
      "balance_loss_clip": 1.05453563,
      "balance_loss_mlp": 1.02912688,
      "epoch": 0.1825287079901401,
      "flos": 14790025584000.0,
      "grad_norm": 2.3743846892366602,
      "language_loss": 0.91416663,
      "learning_rate": 3.7610117403622547e-06,
      "loss": 0.93681931,
      "num_input_tokens_seen": 32167725,
      "step": 1518,
      "time_per_iteration": 2.638896942138672
    },
    {
      "auxiliary_loss_clip": 0.01218434,
      "auxiliary_loss_mlp": 0.01040988,
      "balance_loss_clip": 1.04945445,
      "balance_loss_mlp": 1.03002095,
      "epoch": 0.18264895088077918,
      "flos": 21946232292480.0,
      "grad_norm": 2.9752348797470654,
      "language_loss": 0.90237391,
      "learning_rate": 3.7606423467846313e-06,
      "loss": 0.92496812,
      "num_input_tokens_seen": 32187330,
      "step": 1519,
      "time_per_iteration": 3.631962299346924
    },
    {
      "auxiliary_loss_clip": 0.01233745,
      "auxiliary_loss_mlp": 0.01038794,
      "balance_loss_clip": 1.05682015,
      "balance_loss_mlp": 1.02857208,
      "epoch": 0.18276919377141826,
      "flos": 20886759711360.0,
      "grad_norm": 1.8285762289800862,
      "language_loss": 0.79710209,
      "learning_rate": 3.760272686118964e-06,
      "loss": 0.81982756,
      "num_input_tokens_seen": 32205550,
      "step": 1520,
      "time_per_iteration": 2.690690517425537
    },
    {
      "auxiliary_loss_clip": 0.0122632,
      "auxiliary_loss_mlp": 0.01036601,
      "balance_loss_clip": 1.05496895,
      "balance_loss_mlp": 1.02678382,
      "epoch": 0.18288943666205737,
      "flos": 21469877101440.0,
      "grad_norm": 2.3606503582183342,
      "language_loss": 0.9295966,
      "learning_rate": 3.7599027584213297e-06,
      "loss": 0.9522258,
      "num_input_tokens_seen": 32224430,
      "step": 1521,
      "time_per_iteration": 2.711979389190674
    },
    {
      "auxiliary_loss_clip": 0.01227134,
      "auxiliary_loss_mlp": 0.01033658,
      "balance_loss_clip": 1.05592036,
      "balance_loss_mlp": 1.02315521,
      "epoch": 0.18300967955269645,
      "flos": 21539363961600.0,
      "grad_norm": 2.191587246687912,
      "language_loss": 0.78365457,
      "learning_rate": 3.7595325637478465e-06,
      "loss": 0.80626249,
      "num_input_tokens_seen": 32242455,
      "step": 1522,
      "time_per_iteration": 2.674525737762451
    },
    {
      "auxiliary_loss_clip": 0.01214465,
      "auxiliary_loss_mlp": 0.01038803,
      "balance_loss_clip": 1.05431855,
      "balance_loss_mlp": 1.02769268,
      "epoch": 0.18312992244333554,
      "flos": 28876237102080.0,
      "grad_norm": 2.8084797399622596,
      "language_loss": 0.815687,
      "learning_rate": 3.7591621021546723e-06,
      "loss": 0.83821964,
      "num_input_tokens_seen": 32264450,
      "step": 1523,
      "time_per_iteration": 2.665463924407959
    },
    {
      "auxiliary_loss_clip": 0.01215069,
      "auxiliary_loss_mlp": 0.01031256,
      "balance_loss_clip": 1.05394411,
      "balance_loss_mlp": 1.01978803,
      "epoch": 0.18325016533397462,
      "flos": 20120102801280.0,
      "grad_norm": 2.071757178890669,
      "language_loss": 0.81750703,
      "learning_rate": 3.7587913736980062e-06,
      "loss": 0.83997023,
      "num_input_tokens_seen": 32284090,
      "step": 1524,
      "time_per_iteration": 2.6602883338928223
    },
    {
      "auxiliary_loss_clip": 0.01212614,
      "auxiliary_loss_mlp": 0.01031365,
      "balance_loss_clip": 1.04525506,
      "balance_loss_mlp": 1.02107108,
      "epoch": 0.18337040822461373,
      "flos": 23329187781120.0,
      "grad_norm": 1.8313554861043195,
      "language_loss": 0.84781915,
      "learning_rate": 3.7584203784340865e-06,
      "loss": 0.87025893,
      "num_input_tokens_seen": 32303260,
      "step": 1525,
      "time_per_iteration": 2.744483232498169
    },
    {
      "auxiliary_loss_clip": 0.01219826,
      "auxiliary_loss_mlp": 0.01038179,
      "balance_loss_clip": 1.05213332,
      "balance_loss_mlp": 1.02790904,
      "epoch": 0.1834906511152528,
      "flos": 25009555881600.0,
      "grad_norm": 1.9839796153924738,
      "language_loss": 0.85913277,
      "learning_rate": 3.7580491164191938e-06,
      "loss": 0.88171279,
      "num_input_tokens_seen": 32321570,
      "step": 1526,
      "time_per_iteration": 2.7070558071136475
    },
    {
      "auxiliary_loss_clip": 0.01123983,
      "auxiliary_loss_mlp": 0.0100488,
      "balance_loss_clip": 1.0215795,
      "balance_loss_mlp": 1.00268698,
      "epoch": 0.1836108940058919,
      "flos": 67251493589760.0,
      "grad_norm": 0.7474562213110654,
      "language_loss": 0.61259329,
      "learning_rate": 3.757677587709648e-06,
      "loss": 0.63388193,
      "num_input_tokens_seen": 32384835,
      "step": 1527,
      "time_per_iteration": 3.2754855155944824
    },
    {
      "auxiliary_loss_clip": 0.01223254,
      "auxiliary_loss_mlp": 0.01030855,
      "balance_loss_clip": 1.05566478,
      "balance_loss_mlp": 1.02068591,
      "epoch": 0.183731136896531,
      "flos": 25738721971200.0,
      "grad_norm": 2.145529114817314,
      "language_loss": 0.75959861,
      "learning_rate": 3.7573057923618095e-06,
      "loss": 0.78213978,
      "num_input_tokens_seen": 32404930,
      "step": 1528,
      "time_per_iteration": 2.658268451690674
    },
    {
      "auxiliary_loss_clip": 0.01226773,
      "auxiliary_loss_mlp": 0.01033698,
      "balance_loss_clip": 1.04900038,
      "balance_loss_mlp": 1.02281451,
      "epoch": 0.1838513797871701,
      "flos": 20449403712000.0,
      "grad_norm": 2.021290951940841,
      "language_loss": 0.7434718,
      "learning_rate": 3.7569337304320793e-06,
      "loss": 0.76607656,
      "num_input_tokens_seen": 32424515,
      "step": 1529,
      "time_per_iteration": 2.7442626953125
    },
    {
      "auxiliary_loss_clip": 0.01124467,
      "auxiliary_loss_mlp": 0.01003938,
      "balance_loss_clip": 1.01941824,
      "balance_loss_mlp": 1.00178003,
      "epoch": 0.18397162267780917,
      "flos": 68565141786240.0,
      "grad_norm": 0.8451078347469492,
      "language_loss": 0.64461625,
      "learning_rate": 3.756561401976899e-06,
      "loss": 0.66590023,
      "num_input_tokens_seen": 32484220,
      "step": 1530,
      "time_per_iteration": 3.0374674797058105
    },
    {
      "auxiliary_loss_clip": 0.01224376,
      "auxiliary_loss_mlp": 0.01032511,
      "balance_loss_clip": 1.05986309,
      "balance_loss_mlp": 1.02258098,
      "epoch": 0.18409186556844825,
      "flos": 31941104976000.0,
      "grad_norm": 2.1060587608878953,
      "language_loss": 0.82750034,
      "learning_rate": 3.7561888070527514e-06,
      "loss": 0.85006917,
      "num_input_tokens_seen": 32506260,
      "step": 1531,
      "time_per_iteration": 2.696082592010498
    },
    {
      "auxiliary_loss_clip": 0.01209661,
      "auxiliary_loss_mlp": 0.01208943,
      "balance_loss_clip": 1.0513432,
      "balance_loss_mlp": 1.00052309,
      "epoch": 0.18421210845908736,
      "flos": 20120533764480.0,
      "grad_norm": 2.264970234958236,
      "language_loss": 0.80325681,
      "learning_rate": 3.7558159457161577e-06,
      "loss": 0.82744288,
      "num_input_tokens_seen": 32524225,
      "step": 1532,
      "time_per_iteration": 2.683621406555176
    },
    {
      "auxiliary_loss_clip": 0.01227974,
      "auxiliary_loss_mlp": 0.01209475,
      "balance_loss_clip": 1.05796623,
      "balance_loss_mlp": 1.0005455,
      "epoch": 0.18433235134972645,
      "flos": 23110491824640.0,
      "grad_norm": 2.6621063172911574,
      "language_loss": 0.78376198,
      "learning_rate": 3.755442818023681e-06,
      "loss": 0.80813646,
      "num_input_tokens_seen": 32543850,
      "step": 1533,
      "time_per_iteration": 2.7519640922546387
    },
    {
      "auxiliary_loss_clip": 0.01227147,
      "auxiliary_loss_mlp": 0.01030475,
      "balance_loss_clip": 1.05367446,
      "balance_loss_mlp": 1.02108693,
      "epoch": 0.18445259424036553,
      "flos": 18291351617280.0,
      "grad_norm": 2.5155055396459938,
      "language_loss": 0.7599166,
      "learning_rate": 3.7550694240319246e-06,
      "loss": 0.78249282,
      "num_input_tokens_seen": 32561725,
      "step": 1534,
      "time_per_iteration": 2.678382158279419
    },
    {
      "auxiliary_loss_clip": 0.01225392,
      "auxiliary_loss_mlp": 0.01028411,
      "balance_loss_clip": 1.05563784,
      "balance_loss_mlp": 1.0186125,
      "epoch": 0.18457283713100464,
      "flos": 21324079797120.0,
      "grad_norm": 1.9874483404488021,
      "language_loss": 0.77060652,
      "learning_rate": 3.7546957637975326e-06,
      "loss": 0.79314458,
      "num_input_tokens_seen": 32579135,
      "step": 1535,
      "time_per_iteration": 2.623957633972168
    },
    {
      "auxiliary_loss_clip": 0.01217964,
      "auxiliary_loss_mlp": 0.01029263,
      "balance_loss_clip": 1.04396462,
      "balance_loss_mlp": 1.01942801,
      "epoch": 0.18469308002164372,
      "flos": 20375679047040.0,
      "grad_norm": 1.6525910830901065,
      "language_loss": 0.74118829,
      "learning_rate": 3.7543218373771873e-06,
      "loss": 0.76366055,
      "num_input_tokens_seen": 32598460,
      "step": 1536,
      "time_per_iteration": 2.7211897373199463
    },
    {
      "auxiliary_loss_clip": 0.0122165,
      "auxiliary_loss_mlp": 0.0120915,
      "balance_loss_clip": 1.04906106,
      "balance_loss_mlp": 1.00059938,
      "epoch": 0.1848133229122828,
      "flos": 26435892021120.0,
      "grad_norm": 1.3818259432546827,
      "language_loss": 0.78345191,
      "learning_rate": 3.753947644827615e-06,
      "loss": 0.80775994,
      "num_input_tokens_seen": 32621920,
      "step": 1537,
      "time_per_iteration": 2.74827241897583
    },
    {
      "auxiliary_loss_clip": 0.01125891,
      "auxiliary_loss_mlp": 0.01008207,
      "balance_loss_clip": 1.01866758,
      "balance_loss_mlp": 1.00601375,
      "epoch": 0.1849335658029219,
      "flos": 70547447612160.0,
      "grad_norm": 0.9239072590646119,
      "language_loss": 0.57152569,
      "learning_rate": 3.753573186205579e-06,
      "loss": 0.59286666,
      "num_input_tokens_seen": 32690040,
      "step": 1538,
      "time_per_iteration": 3.3095126152038574
    },
    {
      "auxiliary_loss_clip": 0.01214247,
      "auxiliary_loss_mlp": 0.0120934,
      "balance_loss_clip": 1.05256534,
      "balance_loss_mlp": 1.00064445,
      "epoch": 0.185053808693561,
      "flos": 17384140788480.0,
      "grad_norm": 2.5175737481943012,
      "language_loss": 0.78236032,
      "learning_rate": 3.753198461567885e-06,
      "loss": 0.80659622,
      "num_input_tokens_seen": 32707285,
      "step": 1539,
      "time_per_iteration": 2.6269023418426514
    },
    {
      "auxiliary_loss_clip": 0.01220961,
      "auxiliary_loss_mlp": 0.01033339,
      "balance_loss_clip": 1.05575526,
      "balance_loss_mlp": 1.02360582,
      "epoch": 0.18517405158420008,
      "flos": 28986159697920.0,
      "grad_norm": 2.1474942775998693,
      "language_loss": 0.91973293,
      "learning_rate": 3.7528234709713783e-06,
      "loss": 0.942276,
      "num_input_tokens_seen": 32730030,
      "step": 1540,
      "time_per_iteration": 3.666217803955078
    },
    {
      "auxiliary_loss_clip": 0.0122544,
      "auxiliary_loss_mlp": 0.01031859,
      "balance_loss_clip": 1.0588026,
      "balance_loss_mlp": 1.02210152,
      "epoch": 0.18529429447483917,
      "flos": 26794962328320.0,
      "grad_norm": 2.0201207852312484,
      "language_loss": 0.84642828,
      "learning_rate": 3.7524482144729447e-06,
      "loss": 0.86900127,
      "num_input_tokens_seen": 32749485,
      "step": 1541,
      "time_per_iteration": 2.695064067840576
    },
    {
      "auxiliary_loss_clip": 0.01220786,
      "auxiliary_loss_mlp": 0.01041445,
      "balance_loss_clip": 1.05000174,
      "balance_loss_mlp": 1.03087723,
      "epoch": 0.18541453736547828,
      "flos": 13581595301760.0,
      "grad_norm": 3.021414051033329,
      "language_loss": 0.83653712,
      "learning_rate": 3.7520726921295106e-06,
      "loss": 0.85915947,
      "num_input_tokens_seen": 32766205,
      "step": 1542,
      "time_per_iteration": 3.5622036457061768
    },
    {
      "auxiliary_loss_clip": 0.01217669,
      "auxiliary_loss_mlp": 0.01034011,
      "balance_loss_clip": 1.05274642,
      "balance_loss_mlp": 1.02423,
      "epoch": 0.18553478025611736,
      "flos": 24025424077440.0,
      "grad_norm": 2.3150034396046153,
      "language_loss": 0.72485262,
      "learning_rate": 3.751696903998042e-06,
      "loss": 0.74736947,
      "num_input_tokens_seen": 32784840,
      "step": 1543,
      "time_per_iteration": 2.701585531234741
    },
    {
      "auxiliary_loss_clip": 0.0121909,
      "auxiliary_loss_mlp": 0.01032543,
      "balance_loss_clip": 1.05642653,
      "balance_loss_mlp": 1.02248764,
      "epoch": 0.18565502314675644,
      "flos": 25885165720320.0,
      "grad_norm": 2.173188699393676,
      "language_loss": 0.69785845,
      "learning_rate": 3.7513208501355456e-06,
      "loss": 0.72037482,
      "num_input_tokens_seen": 32805945,
      "step": 1544,
      "time_per_iteration": 2.738481044769287
    },
    {
      "auxiliary_loss_clip": 0.01218562,
      "auxiliary_loss_mlp": 0.0103235,
      "balance_loss_clip": 1.0518713,
      "balance_loss_mlp": 1.02266979,
      "epoch": 0.18577526603739553,
      "flos": 19610063631360.0,
      "grad_norm": 2.021702007673384,
      "language_loss": 0.84032893,
      "learning_rate": 3.750944530599069e-06,
      "loss": 0.86283803,
      "num_input_tokens_seen": 32825515,
      "step": 1545,
      "time_per_iteration": 3.587829113006592
    },
    {
      "auxiliary_loss_clip": 0.01229186,
      "auxiliary_loss_mlp": 0.01034125,
      "balance_loss_clip": 1.05818152,
      "balance_loss_mlp": 1.02330673,
      "epoch": 0.18589550892803464,
      "flos": 18474891137280.0,
      "grad_norm": 2.2307025471262785,
      "language_loss": 0.81206512,
      "learning_rate": 3.7505679454456992e-06,
      "loss": 0.83469832,
      "num_input_tokens_seen": 32842125,
      "step": 1546,
      "time_per_iteration": 3.5550503730773926
    },
    {
      "auxiliary_loss_clip": 0.01222987,
      "auxiliary_loss_mlp": 0.01031627,
      "balance_loss_clip": 1.04596138,
      "balance_loss_mlp": 1.02094007,
      "epoch": 0.18601575181867372,
      "flos": 23549966726400.0,
      "grad_norm": 1.9979730346886548,
      "language_loss": 0.70426172,
      "learning_rate": 3.750191094732564e-06,
      "loss": 0.72680783,
      "num_input_tokens_seen": 32862990,
      "step": 1547,
      "time_per_iteration": 2.7620317935943604
    },
    {
      "auxiliary_loss_clip": 0.01224295,
      "auxiliary_loss_mlp": 0.01209514,
      "balance_loss_clip": 1.0460422,
      "balance_loss_mlp": 1.00066221,
      "epoch": 0.1861359947093128,
      "flos": 26360192108160.0,
      "grad_norm": 1.932551085343101,
      "language_loss": 0.75684798,
      "learning_rate": 3.7498139785168313e-06,
      "loss": 0.78118604,
      "num_input_tokens_seen": 32883595,
      "step": 1548,
      "time_per_iteration": 2.7899162769317627
    },
    {
      "auxiliary_loss_clip": 0.01221669,
      "auxiliary_loss_mlp": 0.0103778,
      "balance_loss_clip": 1.05891144,
      "balance_loss_mlp": 1.02699804,
      "epoch": 0.1862562375999519,
      "flos": 23331198942720.0,
      "grad_norm": 1.829103799860569,
      "language_loss": 0.7761364,
      "learning_rate": 3.749436596855709e-06,
      "loss": 0.79873091,
      "num_input_tokens_seen": 32902895,
      "step": 1549,
      "time_per_iteration": 2.653069019317627
    },
    {
      "auxiliary_loss_clip": 0.01217556,
      "auxiliary_loss_mlp": 0.01033159,
      "balance_loss_clip": 1.05378795,
      "balance_loss_mlp": 1.02254987,
      "epoch": 0.186376480490591,
      "flos": 16648222942080.0,
      "grad_norm": 2.3250457090713867,
      "language_loss": 0.90760636,
      "learning_rate": 3.749058949806446e-06,
      "loss": 0.93011355,
      "num_input_tokens_seen": 32919620,
      "step": 1550,
      "time_per_iteration": 2.6187820434570312
    },
    {
      "auxiliary_loss_clip": 0.01221812,
      "auxiliary_loss_mlp": 0.01029458,
      "balance_loss_clip": 1.0542717,
      "balance_loss_mlp": 1.01973104,
      "epoch": 0.18649672338123008,
      "flos": 21468656039040.0,
      "grad_norm": 1.86073090716882,
      "language_loss": 0.84180307,
      "learning_rate": 3.748681037426331e-06,
      "loss": 0.86431581,
      "num_input_tokens_seen": 32938830,
      "step": 1551,
      "time_per_iteration": 2.670980215072632
    },
    {
      "auxiliary_loss_clip": 0.0122327,
      "auxiliary_loss_mlp": 0.0103911,
      "balance_loss_clip": 1.05937028,
      "balance_loss_mlp": 1.02938211,
      "epoch": 0.1866169662718692,
      "flos": 12312728386560.0,
      "grad_norm": 2.358168803094463,
      "language_loss": 0.91771924,
      "learning_rate": 3.7483028597726936e-06,
      "loss": 0.94034308,
      "num_input_tokens_seen": 32955600,
      "step": 1552,
      "time_per_iteration": 2.566044330596924
    },
    {
      "auxiliary_loss_clip": 0.0122729,
      "auxiliary_loss_mlp": 0.01038877,
      "balance_loss_clip": 1.05506182,
      "balance_loss_mlp": 1.02797532,
      "epoch": 0.18673720916250827,
      "flos": 23581280407680.0,
      "grad_norm": 2.380060777102665,
      "language_loss": 0.62943053,
      "learning_rate": 3.7479244169029017e-06,
      "loss": 0.65209222,
      "num_input_tokens_seen": 32975390,
      "step": 1553,
      "time_per_iteration": 2.7365100383758545
    },
    {
      "auxiliary_loss_clip": 0.01225179,
      "auxiliary_loss_mlp": 0.01027973,
      "balance_loss_clip": 1.05328822,
      "balance_loss_mlp": 1.01786375,
      "epoch": 0.18685745205314735,
      "flos": 19718370115200.0,
      "grad_norm": 2.4170511803751644,
      "language_loss": 0.73597062,
      "learning_rate": 3.7475457088743658e-06,
      "loss": 0.75850219,
      "num_input_tokens_seen": 32992640,
      "step": 1554,
      "time_per_iteration": 2.6072304248809814
    },
    {
      "auxiliary_loss_clip": 0.01218348,
      "auxiliary_loss_mlp": 0.01039446,
      "balance_loss_clip": 1.05621028,
      "balance_loss_mlp": 1.02801943,
      "epoch": 0.18697769494378644,
      "flos": 34204123589760.0,
      "grad_norm": 2.0768404615573495,
      "language_loss": 0.74702084,
      "learning_rate": 3.7471667357445348e-06,
      "loss": 0.76959872,
      "num_input_tokens_seen": 33012470,
      "step": 1555,
      "time_per_iteration": 2.800868272781372
    },
    {
      "auxiliary_loss_clip": 0.01232832,
      "auxiliary_loss_mlp": 0.01029724,
      "balance_loss_clip": 1.05191314,
      "balance_loss_mlp": 1.02008045,
      "epoch": 0.18709793783442555,
      "flos": 34241327101440.0,
      "grad_norm": 2.9451607389114036,
      "language_loss": 0.72905898,
      "learning_rate": 3.7467874975709e-06,
      "loss": 0.75168455,
      "num_input_tokens_seen": 33033275,
      "step": 1556,
      "time_per_iteration": 2.8036012649536133
    },
    {
      "auxiliary_loss_clip": 0.01229196,
      "auxiliary_loss_mlp": 0.01042465,
      "balance_loss_clip": 1.05848169,
      "balance_loss_mlp": 1.03211141,
      "epoch": 0.18721818072506463,
      "flos": 40734550529280.0,
      "grad_norm": 2.0131570870390716,
      "language_loss": 0.7806778,
      "learning_rate": 3.7464079944109904e-06,
      "loss": 0.80339444,
      "num_input_tokens_seen": 33055135,
      "step": 1557,
      "time_per_iteration": 2.8124959468841553
    },
    {
      "auxiliary_loss_clip": 0.01229577,
      "auxiliary_loss_mlp": 0.01030969,
      "balance_loss_clip": 1.05381751,
      "balance_loss_mlp": 1.0211463,
      "epoch": 0.18733842361570371,
      "flos": 22157386392960.0,
      "grad_norm": 2.202158212373708,
      "language_loss": 0.77577233,
      "learning_rate": 3.746028226322376e-06,
      "loss": 0.79837775,
      "num_input_tokens_seen": 33071015,
      "step": 1558,
      "time_per_iteration": 2.6351332664489746
    },
    {
      "auxiliary_loss_clip": 0.01221398,
      "auxiliary_loss_mlp": 0.01031927,
      "balance_loss_clip": 1.05466044,
      "balance_loss_mlp": 1.02237844,
      "epoch": 0.18745866650634282,
      "flos": 18914940656640.0,
      "grad_norm": 1.970305387049277,
      "language_loss": 0.75899875,
      "learning_rate": 3.745648193362669e-06,
      "loss": 0.78153199,
      "num_input_tokens_seen": 33090370,
      "step": 1559,
      "time_per_iteration": 2.6922547817230225
    },
    {
      "auxiliary_loss_clip": 0.01225346,
      "auxiliary_loss_mlp": 0.01033907,
      "balance_loss_clip": 1.05487895,
      "balance_loss_mlp": 1.02487135,
      "epoch": 0.1875789093969819,
      "flos": 19314626267520.0,
      "grad_norm": 1.9795039225208415,
      "language_loss": 0.72948605,
      "learning_rate": 3.745267895589518e-06,
      "loss": 0.75207853,
      "num_input_tokens_seen": 33108910,
      "step": 1560,
      "time_per_iteration": 2.648768663406372
    },
    {
      "auxiliary_loss_clip": 0.01226917,
      "auxiliary_loss_mlp": 0.01033773,
      "balance_loss_clip": 1.05699348,
      "balance_loss_mlp": 1.02340817,
      "epoch": 0.187699152287621,
      "flos": 17018965169280.0,
      "grad_norm": 1.9302842758080094,
      "language_loss": 0.82288611,
      "learning_rate": 3.7448873330606154e-06,
      "loss": 0.84549296,
      "num_input_tokens_seen": 33126680,
      "step": 1561,
      "time_per_iteration": 2.6292097568511963
    },
    {
      "auxiliary_loss_clip": 0.01219174,
      "auxiliary_loss_mlp": 0.01034411,
      "balance_loss_clip": 1.05418456,
      "balance_loss_mlp": 1.02355111,
      "epoch": 0.18781939517826007,
      "flos": 22346384780160.0,
      "grad_norm": 2.030360926246517,
      "language_loss": 0.87414515,
      "learning_rate": 3.7445065058336914e-06,
      "loss": 0.89668101,
      "num_input_tokens_seen": 33145550,
      "step": 1562,
      "time_per_iteration": 2.6598777770996094
    },
    {
      "auxiliary_loss_clip": 0.01214114,
      "auxiliary_loss_mlp": 0.01030254,
      "balance_loss_clip": 1.04603791,
      "balance_loss_mlp": 1.02084243,
      "epoch": 0.18793963806889918,
      "flos": 14611478054400.0,
      "grad_norm": 1.9517871262965316,
      "language_loss": 0.86890662,
      "learning_rate": 3.7441254139665176e-06,
      "loss": 0.89135033,
      "num_input_tokens_seen": 33161735,
      "step": 1563,
      "time_per_iteration": 2.706392288208008
    },
    {
      "auxiliary_loss_clip": 0.01223354,
      "auxiliary_loss_mlp": 0.01034391,
      "balance_loss_clip": 1.06202805,
      "balance_loss_mlp": 1.02508712,
      "epoch": 0.18805988095953827,
      "flos": 17457075354240.0,
      "grad_norm": 1.6992275475721956,
      "language_loss": 0.83047569,
      "learning_rate": 3.743744057516905e-06,
      "loss": 0.85305315,
      "num_input_tokens_seen": 33179795,
      "step": 1564,
      "time_per_iteration": 2.60733962059021
    },
    {
      "auxiliary_loss_clip": 0.01229906,
      "auxiliary_loss_mlp": 0.01041528,
      "balance_loss_clip": 1.05258429,
      "balance_loss_mlp": 1.03045309,
      "epoch": 0.18818012385017735,
      "flos": 15043877976960.0,
      "grad_norm": 4.283408041285684,
      "language_loss": 0.87429094,
      "learning_rate": 3.743362436542706e-06,
      "loss": 0.89700526,
      "num_input_tokens_seen": 33194485,
      "step": 1565,
      "time_per_iteration": 2.663628578186035
    },
    {
      "auxiliary_loss_clip": 0.01218994,
      "auxiliary_loss_mlp": 0.01031251,
      "balance_loss_clip": 1.05751467,
      "balance_loss_mlp": 1.02162516,
      "epoch": 0.18830036674081646,
      "flos": 47551975136640.0,
      "grad_norm": 1.7339105590216717,
      "language_loss": 0.76748943,
      "learning_rate": 3.7429805511018115e-06,
      "loss": 0.78999192,
      "num_input_tokens_seen": 33216145,
      "step": 1566,
      "time_per_iteration": 2.8608407974243164
    },
    {
      "auxiliary_loss_clip": 0.01220662,
      "auxiliary_loss_mlp": 0.01209737,
      "balance_loss_clip": 1.05419445,
      "balance_loss_mlp": 1.00071585,
      "epoch": 0.18842060963145554,
      "flos": 30044626698240.0,
      "grad_norm": 1.7476766475482195,
      "language_loss": 0.78479517,
      "learning_rate": 3.7425984012521524e-06,
      "loss": 0.8090992,
      "num_input_tokens_seen": 33236345,
      "step": 1567,
      "time_per_iteration": 3.684521198272705
    },
    {
      "auxiliary_loss_clip": 0.01126329,
      "auxiliary_loss_mlp": 0.01200773,
      "balance_loss_clip": 1.02121365,
      "balance_loss_mlp": 1.00033247,
      "epoch": 0.18854085252209463,
      "flos": 70318372625280.0,
      "grad_norm": 0.7446693199958768,
      "language_loss": 0.60455483,
      "learning_rate": 3.7422159870517025e-06,
      "loss": 0.62782586,
      "num_input_tokens_seen": 33301600,
      "step": 1568,
      "time_per_iteration": 3.194420576095581
    },
    {
      "auxiliary_loss_clip": 0.01220423,
      "auxiliary_loss_mlp": 0.01030294,
      "balance_loss_clip": 1.05387449,
      "balance_loss_mlp": 1.02027416,
      "epoch": 0.1886610954127337,
      "flos": 21289318410240.0,
      "grad_norm": 1.6612150811403943,
      "language_loss": 0.79309893,
      "learning_rate": 3.7418333085584717e-06,
      "loss": 0.81560612,
      "num_input_tokens_seen": 33322785,
      "step": 1569,
      "time_per_iteration": 3.6194517612457275
    },
    {
      "auxiliary_loss_clip": 0.01229693,
      "auxiliary_loss_mlp": 0.01037156,
      "balance_loss_clip": 1.05705404,
      "balance_loss_mlp": 1.02660561,
      "epoch": 0.18878133830337282,
      "flos": 17266819991040.0,
      "grad_norm": 2.1365278538052204,
      "language_loss": 0.91084361,
      "learning_rate": 3.7414503658305128e-06,
      "loss": 0.93351209,
      "num_input_tokens_seen": 33340020,
      "step": 1570,
      "time_per_iteration": 2.635061025619507
    },
    {
      "auxiliary_loss_clip": 0.01232626,
      "auxiliary_loss_mlp": 0.01035172,
      "balance_loss_clip": 1.04889536,
      "balance_loss_mlp": 1.02516425,
      "epoch": 0.1889015811940119,
      "flos": 25775207210880.0,
      "grad_norm": 2.4117251149077314,
      "language_loss": 0.77924645,
      "learning_rate": 3.7410671589259185e-06,
      "loss": 0.80192441,
      "num_input_tokens_seen": 33358620,
      "step": 1571,
      "time_per_iteration": 2.765968084335327
    },
    {
      "auxiliary_loss_clip": 0.01222302,
      "auxiliary_loss_mlp": 0.01035218,
      "balance_loss_clip": 1.05966258,
      "balance_loss_mlp": 1.02510273,
      "epoch": 0.18902182408465099,
      "flos": 21032197879680.0,
      "grad_norm": 3.9992328222988323,
      "language_loss": 0.79514301,
      "learning_rate": 3.7406836879028205e-06,
      "loss": 0.81771821,
      "num_input_tokens_seen": 33378845,
      "step": 1572,
      "time_per_iteration": 4.566009759902954
    },
    {
      "auxiliary_loss_clip": 0.01221148,
      "auxiliary_loss_mlp": 0.0103046,
      "balance_loss_clip": 1.05808353,
      "balance_loss_mlp": 1.02054179,
      "epoch": 0.1891420669752901,
      "flos": 22272121411200.0,
      "grad_norm": 3.5571312481784285,
      "language_loss": 0.76545531,
      "learning_rate": 3.7402999528193907e-06,
      "loss": 0.78797138,
      "num_input_tokens_seen": 33398345,
      "step": 1573,
      "time_per_iteration": 2.7396645545959473
    },
    {
      "auxiliary_loss_clip": 0.0121103,
      "auxiliary_loss_mlp": 0.01209447,
      "balance_loss_clip": 1.05165744,
      "balance_loss_mlp": 1.00078464,
      "epoch": 0.18926230986592918,
      "flos": 22017802141440.0,
      "grad_norm": 5.757679602349063,
      "language_loss": 0.85990506,
      "learning_rate": 3.739915953733842e-06,
      "loss": 0.88410985,
      "num_input_tokens_seen": 33416390,
      "step": 1574,
      "time_per_iteration": 2.7201614379882812
    },
    {
      "auxiliary_loss_clip": 0.01218882,
      "auxiliary_loss_mlp": 0.0103307,
      "balance_loss_clip": 1.05775309,
      "balance_loss_mlp": 1.02332497,
      "epoch": 0.18938255275656826,
      "flos": 24462672336000.0,
      "grad_norm": 1.8128656753346566,
      "language_loss": 0.82215106,
      "learning_rate": 3.7395316907044264e-06,
      "loss": 0.84467053,
      "num_input_tokens_seen": 33437175,
      "step": 1575,
      "time_per_iteration": 2.6385791301727295
    },
    {
      "auxiliary_loss_clip": 0.01222685,
      "auxiliary_loss_mlp": 0.01038291,
      "balance_loss_clip": 1.05663204,
      "balance_loss_mlp": 1.02830124,
      "epoch": 0.18950279564720737,
      "flos": 24427049022720.0,
      "grad_norm": 3.937007559832635,
      "language_loss": 0.80027461,
      "learning_rate": 3.7391471637894364e-06,
      "loss": 0.82288444,
      "num_input_tokens_seen": 33459440,
      "step": 1576,
      "time_per_iteration": 2.747525930404663
    },
    {
      "auxiliary_loss_clip": 0.01228184,
      "auxiliary_loss_mlp": 0.01035369,
      "balance_loss_clip": 1.05208206,
      "balance_loss_mlp": 1.02589178,
      "epoch": 0.18962303853784646,
      "flos": 19756291898880.0,
      "grad_norm": 1.7762498252754892,
      "language_loss": 0.85162961,
      "learning_rate": 3.738762373047205e-06,
      "loss": 0.87426519,
      "num_input_tokens_seen": 33479360,
      "step": 1577,
      "time_per_iteration": 2.7257702350616455
    },
    {
      "auxiliary_loss_clip": 0.01225977,
      "auxiliary_loss_mlp": 0.01035589,
      "balance_loss_clip": 1.05553102,
      "balance_loss_mlp": 1.02602208,
      "epoch": 0.18974328142848554,
      "flos": 21032054225280.0,
      "grad_norm": 1.7509638891579018,
      "language_loss": 0.83744299,
      "learning_rate": 3.738377318536103e-06,
      "loss": 0.86005867,
      "num_input_tokens_seen": 33499245,
      "step": 1578,
      "time_per_iteration": 2.7273459434509277
    },
    {
      "auxiliary_loss_clip": 0.01216985,
      "auxiliary_loss_mlp": 0.0103557,
      "balance_loss_clip": 1.06002438,
      "balance_loss_mlp": 1.02679658,
      "epoch": 0.18986352431912462,
      "flos": 12966122736000.0,
      "grad_norm": 2.401300230675055,
      "language_loss": 0.7142182,
      "learning_rate": 3.7379920003145447e-06,
      "loss": 0.73674381,
      "num_input_tokens_seen": 33513520,
      "step": 1579,
      "time_per_iteration": 2.6177778244018555
    },
    {
      "auxiliary_loss_clip": 0.01216827,
      "auxiliary_loss_mlp": 0.01035198,
      "balance_loss_clip": 1.05713224,
      "balance_loss_mlp": 1.02476108,
      "epoch": 0.18998376720976373,
      "flos": 23767908497280.0,
      "grad_norm": 2.0411503815779253,
      "language_loss": 0.83970892,
      "learning_rate": 3.7376064184409817e-06,
      "loss": 0.86222923,
      "num_input_tokens_seen": 33533100,
      "step": 1580,
      "time_per_iteration": 2.6927289962768555
    },
    {
      "auxiliary_loss_clip": 0.01222981,
      "auxiliary_loss_mlp": 0.01031732,
      "balance_loss_clip": 1.05795312,
      "balance_loss_mlp": 1.02157545,
      "epoch": 0.19010401010040281,
      "flos": 22966023323520.0,
      "grad_norm": 1.4638001730816566,
      "language_loss": 0.87067854,
      "learning_rate": 3.7372205729739063e-06,
      "loss": 0.89322561,
      "num_input_tokens_seen": 33554915,
      "step": 1581,
      "time_per_iteration": 2.7113540172576904
    },
    {
      "auxiliary_loss_clip": 0.01225197,
      "auxiliary_loss_mlp": 0.01031828,
      "balance_loss_clip": 1.0576179,
      "balance_loss_mlp": 1.02128994,
      "epoch": 0.1902242529910419,
      "flos": 19135647774720.0,
      "grad_norm": 2.0606116719263,
      "language_loss": 0.72014391,
      "learning_rate": 3.7368344639718514e-06,
      "loss": 0.74271417,
      "num_input_tokens_seen": 33572850,
      "step": 1582,
      "time_per_iteration": 2.611332893371582
    },
    {
      "auxiliary_loss_clip": 0.01221607,
      "auxiliary_loss_mlp": 0.01038892,
      "balance_loss_clip": 1.05577016,
      "balance_loss_mlp": 1.03001118,
      "epoch": 0.190344495881681,
      "flos": 25483935824640.0,
      "grad_norm": 1.7135028485606691,
      "language_loss": 0.80705673,
      "learning_rate": 3.7364480914933895e-06,
      "loss": 0.82966173,
      "num_input_tokens_seen": 33593090,
      "step": 1583,
      "time_per_iteration": 2.695603370666504
    },
    {
      "auxiliary_loss_clip": 0.01222648,
      "auxiliary_loss_mlp": 0.01209632,
      "balance_loss_clip": 1.05266476,
      "balance_loss_mlp": 1.00073266,
      "epoch": 0.1904647387723201,
      "flos": 26792843425920.0,
      "grad_norm": 2.8257358440930003,
      "language_loss": 0.81091857,
      "learning_rate": 3.7360614555971325e-06,
      "loss": 0.83524138,
      "num_input_tokens_seen": 33612745,
      "step": 1584,
      "time_per_iteration": 2.744777202606201
    },
    {
      "auxiliary_loss_clip": 0.01220896,
      "auxiliary_loss_mlp": 0.01208957,
      "balance_loss_clip": 1.0581305,
      "balance_loss_mlp": 1.00059617,
      "epoch": 0.19058498166295917,
      "flos": 23987753688960.0,
      "grad_norm": 2.0312956395349437,
      "language_loss": 0.85535157,
      "learning_rate": 3.735674556341733e-06,
      "loss": 0.87965012,
      "num_input_tokens_seen": 33632360,
      "step": 1585,
      "time_per_iteration": 2.7227272987365723
    },
    {
      "auxiliary_loss_clip": 0.01222377,
      "auxiliary_loss_mlp": 0.01036361,
      "balance_loss_clip": 1.05863476,
      "balance_loss_mlp": 1.0266459,
      "epoch": 0.19070522455359826,
      "flos": 28293299280000.0,
      "grad_norm": 2.0895359782262584,
      "language_loss": 0.82831812,
      "learning_rate": 3.7352873937858835e-06,
      "loss": 0.85090548,
      "num_input_tokens_seen": 33653895,
      "step": 1586,
      "time_per_iteration": 2.694425106048584
    },
    {
      "auxiliary_loss_clip": 0.01218214,
      "auxiliary_loss_mlp": 0.01209455,
      "balance_loss_clip": 1.05388403,
      "balance_loss_mlp": 1.00056219,
      "epoch": 0.19082546744423737,
      "flos": 25660220797440.0,
      "grad_norm": 2.283813159094983,
      "language_loss": 0.7224586,
      "learning_rate": 3.734899967988316e-06,
      "loss": 0.74673533,
      "num_input_tokens_seen": 33672075,
      "step": 1587,
      "time_per_iteration": 2.8177008628845215
    },
    {
      "auxiliary_loss_clip": 0.01216191,
      "auxiliary_loss_mlp": 0.01028401,
      "balance_loss_clip": 1.05195928,
      "balance_loss_mlp": 1.01882243,
      "epoch": 0.19094571033487645,
      "flos": 19719483436800.0,
      "grad_norm": 1.8072894152447845,
      "language_loss": 0.8414017,
      "learning_rate": 3.7345122790078026e-06,
      "loss": 0.86384767,
      "num_input_tokens_seen": 33689640,
      "step": 1588,
      "time_per_iteration": 2.658514976501465
    },
    {
      "auxiliary_loss_clip": 0.01220347,
      "auxiliary_loss_mlp": 0.01032583,
      "balance_loss_clip": 1.05741096,
      "balance_loss_mlp": 1.02213407,
      "epoch": 0.19106595322551553,
      "flos": 21616320850560.0,
      "grad_norm": 2.98402857120886,
      "language_loss": 0.93074465,
      "learning_rate": 3.7341243269031556e-06,
      "loss": 0.95327389,
      "num_input_tokens_seen": 33708630,
      "step": 1589,
      "time_per_iteration": 2.7313225269317627
    },
    {
      "auxiliary_loss_clip": 0.01215092,
      "auxiliary_loss_mlp": 0.01031558,
      "balance_loss_clip": 1.05693388,
      "balance_loss_mlp": 1.02203941,
      "epoch": 0.19118619611615464,
      "flos": 29896890059520.0,
      "grad_norm": 1.8119352497889627,
      "language_loss": 0.7746076,
      "learning_rate": 3.7337361117332275e-06,
      "loss": 0.79707408,
      "num_input_tokens_seen": 33730370,
      "step": 1590,
      "time_per_iteration": 2.7226195335388184
    },
    {
      "auxiliary_loss_clip": 0.0122442,
      "auxiliary_loss_mlp": 0.01029447,
      "balance_loss_clip": 1.05241656,
      "balance_loss_mlp": 1.02039313,
      "epoch": 0.19130643900679373,
      "flos": 17273428093440.0,
      "grad_norm": 1.850292646044904,
      "language_loss": 0.77075684,
      "learning_rate": 3.7333476335569087e-06,
      "loss": 0.7932955,
      "num_input_tokens_seen": 33748370,
      "step": 1591,
      "time_per_iteration": 2.709378480911255
    },
    {
      "auxiliary_loss_clip": 0.01221339,
      "auxiliary_loss_mlp": 0.01032475,
      "balance_loss_clip": 1.05681682,
      "balance_loss_mlp": 1.02207971,
      "epoch": 0.1914266818974328,
      "flos": 24826339584000.0,
      "grad_norm": 2.2667442706890935,
      "language_loss": 0.66895932,
      "learning_rate": 3.7329588924331325e-06,
      "loss": 0.69149745,
      "num_input_tokens_seen": 33769575,
      "step": 1592,
      "time_per_iteration": 2.705207347869873
    },
    {
      "auxiliary_loss_clip": 0.01215875,
      "auxiliary_loss_mlp": 0.01033846,
      "balance_loss_clip": 1.05062342,
      "balance_loss_mlp": 1.02441049,
      "epoch": 0.1915469247880719,
      "flos": 18952467390720.0,
      "grad_norm": 1.8187288867562954,
      "language_loss": 0.82846856,
      "learning_rate": 3.732569888420871e-06,
      "loss": 0.8509658,
      "num_input_tokens_seen": 33789110,
      "step": 1593,
      "time_per_iteration": 3.6374707221984863
    },
    {
      "auxiliary_loss_clip": 0.0122038,
      "auxiliary_loss_mlp": 0.01032552,
      "balance_loss_clip": 1.05865228,
      "balance_loss_mlp": 1.02165627,
      "epoch": 0.191667167678711,
      "flos": 21032952065280.0,
      "grad_norm": 2.0684926301198816,
      "language_loss": 0.82317376,
      "learning_rate": 3.732180621579134e-06,
      "loss": 0.84570301,
      "num_input_tokens_seen": 33808325,
      "step": 1594,
      "time_per_iteration": 2.6307389736175537
    },
    {
      "auxiliary_loss_clip": 0.01235846,
      "auxiliary_loss_mlp": 0.01033146,
      "balance_loss_clip": 1.05838716,
      "balance_loss_mlp": 1.02311444,
      "epoch": 0.1917874105693501,
      "flos": 34237663914240.0,
      "grad_norm": 2.293659663991112,
      "language_loss": 0.81373829,
      "learning_rate": 3.7317910919669745e-06,
      "loss": 0.83642823,
      "num_input_tokens_seen": 33829520,
      "step": 1595,
      "time_per_iteration": 2.785019636154175
    },
    {
      "auxiliary_loss_clip": 0.01220779,
      "auxiliary_loss_mlp": 0.01036274,
      "balance_loss_clip": 1.0595299,
      "balance_loss_mlp": 1.02620125,
      "epoch": 0.19190765345998917,
      "flos": 23550613171200.0,
      "grad_norm": 4.579456009085113,
      "language_loss": 0.76035523,
      "learning_rate": 3.7314012996434826e-06,
      "loss": 0.78292572,
      "num_input_tokens_seen": 33848250,
      "step": 1596,
      "time_per_iteration": 3.528470277786255
    },
    {
      "auxiliary_loss_clip": 0.01224499,
      "auxiliary_loss_mlp": 0.01029487,
      "balance_loss_clip": 1.05833626,
      "balance_loss_mlp": 1.01978898,
      "epoch": 0.19202789635062828,
      "flos": 19861330245120.0,
      "grad_norm": 2.311247956120626,
      "language_loss": 0.81430811,
      "learning_rate": 3.7310112446677907e-06,
      "loss": 0.83684796,
      "num_input_tokens_seen": 33866160,
      "step": 1597,
      "time_per_iteration": 2.7123098373413086
    },
    {
      "auxiliary_loss_clip": 0.01222561,
      "auxiliary_loss_mlp": 0.01029273,
      "balance_loss_clip": 1.06142497,
      "balance_loss_mlp": 1.0196116,
      "epoch": 0.19214813924126736,
      "flos": 20922957642240.0,
      "grad_norm": 2.382181982935039,
      "language_loss": 0.69147396,
      "learning_rate": 3.7306209270990695e-06,
      "loss": 0.7139923,
      "num_input_tokens_seen": 33884165,
      "step": 1598,
      "time_per_iteration": 2.5957956314086914
    },
    {
      "auxiliary_loss_clip": 0.01224052,
      "auxiliary_loss_mlp": 0.01040934,
      "balance_loss_clip": 1.05831182,
      "balance_loss_mlp": 1.03141499,
      "epoch": 0.19226838213190645,
      "flos": 26359725231360.0,
      "grad_norm": 1.8357342243668142,
      "language_loss": 0.86693287,
      "learning_rate": 3.7302303469965292e-06,
      "loss": 0.88958275,
      "num_input_tokens_seen": 33903705,
      "step": 1599,
      "time_per_iteration": 4.542972087860107
    },
    {
      "auxiliary_loss_clip": 0.01220799,
      "auxiliary_loss_mlp": 0.01038263,
      "balance_loss_clip": 1.0585779,
      "balance_loss_mlp": 1.02838063,
      "epoch": 0.19238862502254553,
      "flos": 20850525866880.0,
      "grad_norm": 2.17353738123402,
      "language_loss": 0.70884824,
      "learning_rate": 3.7298395044194206e-06,
      "loss": 0.73143888,
      "num_input_tokens_seen": 33922515,
      "step": 1600,
      "time_per_iteration": 2.659531593322754
    },
    {
      "auxiliary_loss_clip": 0.01223921,
      "auxiliary_loss_mlp": 0.01030491,
      "balance_loss_clip": 1.06345367,
      "balance_loss_mlp": 1.02075791,
      "epoch": 0.19250886791318464,
      "flos": 21726063878400.0,
      "grad_norm": 1.993663569536001,
      "language_loss": 0.94596785,
      "learning_rate": 3.7294483994270356e-06,
      "loss": 0.96851194,
      "num_input_tokens_seen": 33940840,
      "step": 1601,
      "time_per_iteration": 2.6348876953125
    },
    {
      "auxiliary_loss_clip": 0.01214961,
      "auxiliary_loss_mlp": 0.01030737,
      "balance_loss_clip": 1.05150104,
      "balance_loss_mlp": 1.02225542,
      "epoch": 0.19262911080382372,
      "flos": 23367827836800.0,
      "grad_norm": 2.5491908401264602,
      "language_loss": 0.78384894,
      "learning_rate": 3.7290570320787033e-06,
      "loss": 0.80630589,
      "num_input_tokens_seen": 33960420,
      "step": 1602,
      "time_per_iteration": 2.6996512413024902
    },
    {
      "auxiliary_loss_clip": 0.01220789,
      "auxiliary_loss_mlp": 0.0103393,
      "balance_loss_clip": 1.06019485,
      "balance_loss_mlp": 1.02404773,
      "epoch": 0.1927493536944628,
      "flos": 21943502858880.0,
      "grad_norm": 2.419784206638762,
      "language_loss": 0.7126779,
      "learning_rate": 3.728665402433793e-06,
      "loss": 0.73522502,
      "num_input_tokens_seen": 33978990,
      "step": 1603,
      "time_per_iteration": 2.6583192348480225
    },
    {
      "auxiliary_loss_clip": 0.01223181,
      "auxiliary_loss_mlp": 0.010324,
      "balance_loss_clip": 1.05855787,
      "balance_loss_mlp": 1.02339923,
      "epoch": 0.19286959658510192,
      "flos": 16545590807040.0,
      "grad_norm": 2.5077110432021903,
      "language_loss": 0.86025852,
      "learning_rate": 3.7282735105517164e-06,
      "loss": 0.88281429,
      "num_input_tokens_seen": 33997115,
      "step": 1604,
      "time_per_iteration": 2.6411023139953613
    },
    {
      "auxiliary_loss_clip": 0.01231935,
      "auxiliary_loss_mlp": 0.01033602,
      "balance_loss_clip": 1.05108309,
      "balance_loss_mlp": 1.02430344,
      "epoch": 0.192989839475741,
      "flos": 21616967295360.0,
      "grad_norm": 2.4924083487463746,
      "language_loss": 0.67800164,
      "learning_rate": 3.727881356491922e-06,
      "loss": 0.70065701,
      "num_input_tokens_seen": 34015525,
      "step": 1605,
      "time_per_iteration": 2.6942145824432373
    },
    {
      "auxiliary_loss_clip": 0.01219397,
      "auxiliary_loss_mlp": 0.01032314,
      "balance_loss_clip": 1.06139779,
      "balance_loss_mlp": 1.02368915,
      "epoch": 0.19311008236638008,
      "flos": 19281516906240.0,
      "grad_norm": 2.1362754491682066,
      "language_loss": 0.7588768,
      "learning_rate": 3.7274889403139002e-06,
      "loss": 0.78139389,
      "num_input_tokens_seen": 34033150,
      "step": 1606,
      "time_per_iteration": 2.618109941482544
    },
    {
      "auxiliary_loss_clip": 0.0122323,
      "auxiliary_loss_mlp": 0.01029646,
      "balance_loss_clip": 1.05565941,
      "balance_loss_mlp": 1.02039528,
      "epoch": 0.1932303252570192,
      "flos": 28652369587200.0,
      "grad_norm": 2.6191253451811485,
      "language_loss": 0.78726172,
      "learning_rate": 3.727096262077179e-06,
      "loss": 0.80979049,
      "num_input_tokens_seen": 34052145,
      "step": 1607,
      "time_per_iteration": 2.723945379257202
    },
    {
      "auxiliary_loss_clip": 0.01219643,
      "auxiliary_loss_mlp": 0.01033249,
      "balance_loss_clip": 1.05815864,
      "balance_loss_mlp": 1.02380157,
      "epoch": 0.19335056814765827,
      "flos": 18368990864640.0,
      "grad_norm": 1.8936668961664809,
      "language_loss": 0.85547924,
      "learning_rate": 3.7267033218413285e-06,
      "loss": 0.87800813,
      "num_input_tokens_seen": 34069940,
      "step": 1608,
      "time_per_iteration": 2.67311692237854
    },
    {
      "auxiliary_loss_clip": 0.01227473,
      "auxiliary_loss_mlp": 0.01037368,
      "balance_loss_clip": 1.04738843,
      "balance_loss_mlp": 1.02663279,
      "epoch": 0.19347081103829736,
      "flos": 13260877741440.0,
      "grad_norm": 3.1651098223683163,
      "language_loss": 0.81521297,
      "learning_rate": 3.726310119665957e-06,
      "loss": 0.83786136,
      "num_input_tokens_seen": 34086275,
      "step": 1609,
      "time_per_iteration": 2.716780185699463
    },
    {
      "auxiliary_loss_clip": 0.01219629,
      "auxiliary_loss_mlp": 0.01030803,
      "balance_loss_clip": 1.05755699,
      "balance_loss_mlp": 1.02139759,
      "epoch": 0.19359105392893644,
      "flos": 20300122788480.0,
      "grad_norm": 1.7547429454889423,
      "language_loss": 0.85531831,
      "learning_rate": 3.725916655610713e-06,
      "loss": 0.87782264,
      "num_input_tokens_seen": 34105605,
      "step": 1610,
      "time_per_iteration": 2.6468889713287354
    },
    {
      "auxiliary_loss_clip": 0.01213977,
      "auxiliary_loss_mlp": 0.01031053,
      "balance_loss_clip": 1.05351663,
      "balance_loss_mlp": 1.02113473,
      "epoch": 0.19371129681957555,
      "flos": 20484596062080.0,
      "grad_norm": 3.082485139677261,
      "language_loss": 0.75553691,
      "learning_rate": 3.725522929735284e-06,
      "loss": 0.77798718,
      "num_input_tokens_seen": 34122540,
      "step": 1611,
      "time_per_iteration": 2.6909327507019043
    },
    {
      "auxiliary_loss_clip": 0.01227339,
      "auxiliary_loss_mlp": 0.01032655,
      "balance_loss_clip": 1.05432332,
      "balance_loss_mlp": 1.02311873,
      "epoch": 0.19383153971021463,
      "flos": 30445497457920.0,
      "grad_norm": 3.3063898430078806,
      "language_loss": 0.74405283,
      "learning_rate": 3.725128942099399e-06,
      "loss": 0.7666527,
      "num_input_tokens_seen": 34142940,
      "step": 1612,
      "time_per_iteration": 2.7520246505737305
    },
    {
      "auxiliary_loss_clip": 0.01212113,
      "auxiliary_loss_mlp": 0.0103446,
      "balance_loss_clip": 1.05340469,
      "balance_loss_mlp": 1.02489376,
      "epoch": 0.19395178260085372,
      "flos": 24569937325440.0,
      "grad_norm": 1.604950368349793,
      "language_loss": 0.8016392,
      "learning_rate": 3.7247346927628245e-06,
      "loss": 0.82410491,
      "num_input_tokens_seen": 34162875,
      "step": 1613,
      "time_per_iteration": 2.721773386001587
    },
    {
      "auxiliary_loss_clip": 0.01218418,
      "auxiliary_loss_mlp": 0.01209168,
      "balance_loss_clip": 1.05485964,
      "balance_loss_mlp": 1.00049901,
      "epoch": 0.19407202549149283,
      "flos": 28950608211840.0,
      "grad_norm": 1.9396810445538462,
      "language_loss": 0.79669297,
      "learning_rate": 3.7243401817853694e-06,
      "loss": 0.82096887,
      "num_input_tokens_seen": 34183565,
      "step": 1614,
      "time_per_iteration": 2.7079391479492188
    },
    {
      "auxiliary_loss_clip": 0.01214378,
      "auxiliary_loss_mlp": 0.01031767,
      "balance_loss_clip": 1.05703616,
      "balance_loss_mlp": 1.02252209,
      "epoch": 0.1941922683821319,
      "flos": 18004497603840.0,
      "grad_norm": 2.2585311948174396,
      "language_loss": 0.72066343,
      "learning_rate": 3.723945409226879e-06,
      "loss": 0.74312484,
      "num_input_tokens_seen": 34202055,
      "step": 1615,
      "time_per_iteration": 2.6250109672546387
    },
    {
      "auxiliary_loss_clip": 0.01219328,
      "auxiliary_loss_mlp": 0.01036905,
      "balance_loss_clip": 1.05777812,
      "balance_loss_mlp": 1.02709413,
      "epoch": 0.194312511272771,
      "flos": 9720337034880.0,
      "grad_norm": 2.2719007438916448,
      "language_loss": 0.79770827,
      "learning_rate": 3.723550375147241e-06,
      "loss": 0.82027066,
      "num_input_tokens_seen": 34216830,
      "step": 1616,
      "time_per_iteration": 2.592637538909912
    },
    {
      "auxiliary_loss_clip": 0.01208695,
      "auxiliary_loss_mlp": 0.01034199,
      "balance_loss_clip": 1.04828286,
      "balance_loss_mlp": 1.02429318,
      "epoch": 0.19443275416341008,
      "flos": 27016208150400.0,
      "grad_norm": 1.8288201433009106,
      "language_loss": 0.80336773,
      "learning_rate": 3.7231550796063816e-06,
      "loss": 0.8257966,
      "num_input_tokens_seen": 34236840,
      "step": 1617,
      "time_per_iteration": 2.747328042984009
    },
    {
      "auxiliary_loss_clip": 0.01227967,
      "auxiliary_loss_mlp": 0.01036906,
      "balance_loss_clip": 1.05831826,
      "balance_loss_mlp": 1.02648723,
      "epoch": 0.1945529970540492,
      "flos": 15846625077120.0,
      "grad_norm": 2.7349017279198384,
      "language_loss": 0.65108526,
      "learning_rate": 3.722759522664266e-06,
      "loss": 0.67373395,
      "num_input_tokens_seen": 34254140,
      "step": 1618,
      "time_per_iteration": 2.6098899841308594
    },
    {
      "auxiliary_loss_clip": 0.01225462,
      "auxiliary_loss_mlp": 0.01028919,
      "balance_loss_clip": 1.05226111,
      "balance_loss_mlp": 1.01925111,
      "epoch": 0.19467323994468827,
      "flos": 19314985403520.0,
      "grad_norm": 2.3910450769171097,
      "language_loss": 0.818519,
      "learning_rate": 3.7223637043809016e-06,
      "loss": 0.84106278,
      "num_input_tokens_seen": 34273120,
      "step": 1619,
      "time_per_iteration": 2.679191827774048
    },
    {
      "auxiliary_loss_clip": 0.01226104,
      "auxiliary_loss_mlp": 0.01037651,
      "balance_loss_clip": 1.05629921,
      "balance_loss_mlp": 1.02854347,
      "epoch": 0.19479348283532735,
      "flos": 24133227770880.0,
      "grad_norm": 1.7646799906554969,
      "language_loss": 0.86318147,
      "learning_rate": 3.7219676248163322e-06,
      "loss": 0.88581902,
      "num_input_tokens_seen": 34290285,
      "step": 1620,
      "time_per_iteration": 3.73335599899292
    },
    {
      "auxiliary_loss_clip": 0.01225119,
      "auxiliary_loss_mlp": 0.01033903,
      "balance_loss_clip": 1.05946565,
      "balance_loss_mlp": 1.02397299,
      "epoch": 0.19491372572596646,
      "flos": 25775638174080.0,
      "grad_norm": 1.7904524555417223,
      "language_loss": 0.93375492,
      "learning_rate": 3.721571284030643e-06,
      "loss": 0.9563452,
      "num_input_tokens_seen": 34310095,
      "step": 1621,
      "time_per_iteration": 2.6870124340057373
    },
    {
      "auxiliary_loss_clip": 0.01223359,
      "auxiliary_loss_mlp": 0.01026011,
      "balance_loss_clip": 1.05859721,
      "balance_loss_mlp": 1.01630151,
      "epoch": 0.19503396861660555,
      "flos": 19645220067840.0,
      "grad_norm": 2.3426658350474754,
      "language_loss": 0.79138762,
      "learning_rate": 3.7211746820839587e-06,
      "loss": 0.81388134,
      "num_input_tokens_seen": 34327190,
      "step": 1622,
      "time_per_iteration": 3.5037405490875244
    },
    {
      "auxiliary_loss_clip": 0.01211921,
      "auxiliary_loss_mlp": 0.01027043,
      "balance_loss_clip": 1.04736245,
      "balance_loss_mlp": 1.01720262,
      "epoch": 0.19515421150724463,
      "flos": 21033023892480.0,
      "grad_norm": 2.0419524348299984,
      "language_loss": 0.8131963,
      "learning_rate": 3.7207778190364437e-06,
      "loss": 0.83558595,
      "num_input_tokens_seen": 34345615,
      "step": 1623,
      "time_per_iteration": 2.7390458583831787
    },
    {
      "auxiliary_loss_clip": 0.01213842,
      "auxiliary_loss_mlp": 0.01031412,
      "balance_loss_clip": 1.0492717,
      "balance_loss_mlp": 1.0218153,
      "epoch": 0.1952744543978837,
      "flos": 32961255143040.0,
      "grad_norm": 1.6740205637910381,
      "language_loss": 0.74055856,
      "learning_rate": 3.720380694948302e-06,
      "loss": 0.7630111,
      "num_input_tokens_seen": 34368500,
      "step": 1624,
      "time_per_iteration": 2.794052839279175
    },
    {
      "auxiliary_loss_clip": 0.01146345,
      "auxiliary_loss_mlp": 0.01007746,
      "balance_loss_clip": 1.0286808,
      "balance_loss_mlp": 1.00585008,
      "epoch": 0.19539469728852282,
      "flos": 64044312030720.0,
      "grad_norm": 1.1663210399799755,
      "language_loss": 0.71182752,
      "learning_rate": 3.719983309879777e-06,
      "loss": 0.73336846,
      "num_input_tokens_seen": 34428280,
      "step": 1625,
      "time_per_iteration": 4.2033302783966064
    },
    {
      "auxiliary_loss_clip": 0.01214976,
      "auxiliary_loss_mlp": 0.01037599,
      "balance_loss_clip": 1.05283689,
      "balance_loss_mlp": 1.0285455,
      "epoch": 0.1955149401791619,
      "flos": 13370908078080.0,
      "grad_norm": 1.6540621325572944,
      "language_loss": 0.77572894,
      "learning_rate": 3.719585663891151e-06,
      "loss": 0.79825467,
      "num_input_tokens_seen": 34445815,
      "step": 1626,
      "time_per_iteration": 3.5540459156036377
    },
    {
      "auxiliary_loss_clip": 0.01219813,
      "auxiliary_loss_mlp": 0.01035652,
      "balance_loss_clip": 1.0550158,
      "balance_loss_mlp": 1.02548981,
      "epoch": 0.195635183069801,
      "flos": 18728887184640.0,
      "grad_norm": 2.259541411025333,
      "language_loss": 0.79112351,
      "learning_rate": 3.719187757042747e-06,
      "loss": 0.81367821,
      "num_input_tokens_seen": 34463635,
      "step": 1627,
      "time_per_iteration": 2.730978488922119
    },
    {
      "auxiliary_loss_clip": 0.01127986,
      "auxiliary_loss_mlp": 0.0100228,
      "balance_loss_clip": 1.02588582,
      "balance_loss_mlp": 1.00047958,
      "epoch": 0.1957554259604401,
      "flos": 69313952615040.0,
      "grad_norm": 0.7257108192686333,
      "language_loss": 0.54925841,
      "learning_rate": 3.7187895893949275e-06,
      "loss": 0.57056111,
      "num_input_tokens_seen": 34530105,
      "step": 1628,
      "time_per_iteration": 3.254513740539551
    },
    {
      "auxiliary_loss_clip": 0.01211138,
      "auxiliary_loss_mlp": 0.01023838,
      "balance_loss_clip": 1.04857206,
      "balance_loss_mlp": 1.01358569,
      "epoch": 0.19587566885107918,
      "flos": 21069257736960.0,
      "grad_norm": 2.5387710598482336,
      "language_loss": 0.76579833,
      "learning_rate": 3.7183911610080937e-06,
      "loss": 0.78814811,
      "num_input_tokens_seen": 34546970,
      "step": 1629,
      "time_per_iteration": 2.7136127948760986
    },
    {
      "auxiliary_loss_clip": 0.01226446,
      "auxiliary_loss_mlp": 0.01041154,
      "balance_loss_clip": 1.05526483,
      "balance_loss_mlp": 1.03091931,
      "epoch": 0.19599591174171827,
      "flos": 22194661731840.0,
      "grad_norm": 2.2524426354396225,
      "language_loss": 0.74866784,
      "learning_rate": 3.7179924719426872e-06,
      "loss": 0.77134389,
      "num_input_tokens_seen": 34564865,
      "step": 1630,
      "time_per_iteration": 2.661940097808838
    },
    {
      "auxiliary_loss_clip": 0.0122581,
      "auxiliary_loss_mlp": 0.01035222,
      "balance_loss_clip": 1.06107998,
      "balance_loss_mlp": 1.02555966,
      "epoch": 0.19611615463235738,
      "flos": 23768375374080.0,
      "grad_norm": 2.1278310013123503,
      "language_loss": 0.76137155,
      "learning_rate": 3.7175935222591885e-06,
      "loss": 0.78398192,
      "num_input_tokens_seen": 34584165,
      "step": 1631,
      "time_per_iteration": 2.5990326404571533
    },
    {
      "auxiliary_loss_clip": 0.01228012,
      "auxiliary_loss_mlp": 0.01032356,
      "balance_loss_clip": 1.06331527,
      "balance_loss_mlp": 1.0228076,
      "epoch": 0.19623639752299646,
      "flos": 28618218731520.0,
      "grad_norm": 1.9099149445702504,
      "language_loss": 0.74284601,
      "learning_rate": 3.717194312018118e-06,
      "loss": 0.76544964,
      "num_input_tokens_seen": 34603150,
      "step": 1632,
      "time_per_iteration": 2.6991024017333984
    },
    {
      "auxiliary_loss_clip": 0.01219414,
      "auxiliary_loss_mlp": 0.01033241,
      "balance_loss_clip": 1.05762076,
      "balance_loss_mlp": 1.02319717,
      "epoch": 0.19635664041363554,
      "flos": 21032700670080.0,
      "grad_norm": 4.554112487051468,
      "language_loss": 0.76458782,
      "learning_rate": 3.716794841280036e-06,
      "loss": 0.78711432,
      "num_input_tokens_seen": 34621855,
      "step": 1633,
      "time_per_iteration": 2.6237704753875732
    },
    {
      "auxiliary_loss_clip": 0.01226149,
      "auxiliary_loss_mlp": 0.01034544,
      "balance_loss_clip": 1.05854774,
      "balance_loss_mlp": 1.02460766,
      "epoch": 0.19647688330427462,
      "flos": 18879748306560.0,
      "grad_norm": 2.17106892421271,
      "language_loss": 0.77223831,
      "learning_rate": 3.7163951101055407e-06,
      "loss": 0.79484528,
      "num_input_tokens_seen": 34639915,
      "step": 1634,
      "time_per_iteration": 2.5846877098083496
    },
    {
      "auxiliary_loss_clip": 0.01219556,
      "auxiliary_loss_mlp": 0.01033131,
      "balance_loss_clip": 1.05772853,
      "balance_loss_mlp": 1.02324855,
      "epoch": 0.19659712619491373,
      "flos": 24242503921920.0,
      "grad_norm": 3.4608873906031308,
      "language_loss": 0.79263532,
      "learning_rate": 3.715995118555273e-06,
      "loss": 0.81516218,
      "num_input_tokens_seen": 34659890,
      "step": 1635,
      "time_per_iteration": 2.6375386714935303
    },
    {
      "auxiliary_loss_clip": 0.01223728,
      "auxiliary_loss_mlp": 0.01039099,
      "balance_loss_clip": 1.05389559,
      "balance_loss_mlp": 1.02842975,
      "epoch": 0.19671736908555282,
      "flos": 24717422568960.0,
      "grad_norm": 2.1994315756433886,
      "language_loss": 0.86233509,
      "learning_rate": 3.71559486668991e-06,
      "loss": 0.88496339,
      "num_input_tokens_seen": 34678750,
      "step": 1636,
      "time_per_iteration": 2.6868252754211426
    },
    {
      "auxiliary_loss_clip": 0.01229395,
      "auxiliary_loss_mlp": 0.01208822,
      "balance_loss_clip": 1.06337619,
      "balance_loss_mlp": 1.00033903,
      "epoch": 0.1968376119761919,
      "flos": 23842279607040.0,
      "grad_norm": 1.989739894228941,
      "language_loss": 0.77776241,
      "learning_rate": 3.715194354570169e-06,
      "loss": 0.80214453,
      "num_input_tokens_seen": 34698755,
      "step": 1637,
      "time_per_iteration": 2.632681131362915
    },
    {
      "auxiliary_loss_clip": 0.01223287,
      "auxiliary_loss_mlp": 0.01036368,
      "balance_loss_clip": 1.06286407,
      "balance_loss_mlp": 1.02698028,
      "epoch": 0.196957854866831,
      "flos": 18113917409280.0,
      "grad_norm": 1.9811716332082663,
      "language_loss": 0.83829916,
      "learning_rate": 3.714793582256809e-06,
      "loss": 0.86089575,
      "num_input_tokens_seen": 34715820,
      "step": 1638,
      "time_per_iteration": 2.557523250579834
    },
    {
      "auxiliary_loss_clip": 0.01218556,
      "auxiliary_loss_mlp": 0.01032799,
      "balance_loss_clip": 1.06065357,
      "balance_loss_mlp": 1.02285135,
      "epoch": 0.1970780977574701,
      "flos": 21653129312640.0,
      "grad_norm": 3.7215343403126333,
      "language_loss": 0.85142791,
      "learning_rate": 3.7143925498106253e-06,
      "loss": 0.87394142,
      "num_input_tokens_seen": 34734360,
      "step": 1639,
      "time_per_iteration": 2.6080424785614014
    },
    {
      "auxiliary_loss_clip": 0.01220433,
      "auxiliary_loss_mlp": 0.0103312,
      "balance_loss_clip": 1.05214524,
      "balance_loss_mlp": 1.02243924,
      "epoch": 0.19719834064810918,
      "flos": 20811813984000.0,
      "grad_norm": 1.7957163098189857,
      "language_loss": 0.78833163,
      "learning_rate": 3.7139912572924558e-06,
      "loss": 0.81086719,
      "num_input_tokens_seen": 34753390,
      "step": 1640,
      "time_per_iteration": 2.601335048675537
    },
    {
      "auxiliary_loss_clip": 0.01218555,
      "auxiliary_loss_mlp": 0.01035057,
      "balance_loss_clip": 1.05621457,
      "balance_loss_mlp": 1.02562714,
      "epoch": 0.19731858353874826,
      "flos": 23434800744960.0,
      "grad_norm": 4.706902163417018,
      "language_loss": 0.80743396,
      "learning_rate": 3.7135897047631744e-06,
      "loss": 0.82997006,
      "num_input_tokens_seen": 34771275,
      "step": 1641,
      "time_per_iteration": 2.720517873764038
    },
    {
      "auxiliary_loss_clip": 0.0122593,
      "auxiliary_loss_mlp": 0.01033793,
      "balance_loss_clip": 1.05908656,
      "balance_loss_mlp": 1.02349353,
      "epoch": 0.19743882642938737,
      "flos": 23988184652160.0,
      "grad_norm": 1.9188519202924204,
      "language_loss": 0.76297516,
      "learning_rate": 3.713187892283698e-06,
      "loss": 0.78557235,
      "num_input_tokens_seen": 34790885,
      "step": 1642,
      "time_per_iteration": 2.7066457271575928
    },
    {
      "auxiliary_loss_clip": 0.01224589,
      "auxiliary_loss_mlp": 0.01034524,
      "balance_loss_clip": 1.05038261,
      "balance_loss_mlp": 1.02455163,
      "epoch": 0.19755906932002645,
      "flos": 15004340081280.0,
      "grad_norm": 2.1004493567294293,
      "language_loss": 0.87519264,
      "learning_rate": 3.71278581991498e-06,
      "loss": 0.89778364,
      "num_input_tokens_seen": 34806745,
      "step": 1643,
      "time_per_iteration": 2.7332570552825928
    },
    {
      "auxiliary_loss_clip": 0.01229401,
      "auxiliary_loss_mlp": 0.01209889,
      "balance_loss_clip": 1.06119967,
      "balance_loss_mlp": 1.00035834,
      "epoch": 0.19767931221066554,
      "flos": 19494466686720.0,
      "grad_norm": 1.7168113862374852,
      "language_loss": 0.79439896,
      "learning_rate": 3.712383487718015e-06,
      "loss": 0.81879187,
      "num_input_tokens_seen": 34824985,
      "step": 1644,
      "time_per_iteration": 2.6780529022216797
    },
    {
      "auxiliary_loss_clip": 0.01222774,
      "auxiliary_loss_mlp": 0.01030636,
      "balance_loss_clip": 1.05133271,
      "balance_loss_mlp": 1.02152252,
      "epoch": 0.19779955510130465,
      "flos": 25737895958400.0,
      "grad_norm": 2.0824516380646676,
      "language_loss": 0.87125772,
      "learning_rate": 3.7119808957538365e-06,
      "loss": 0.89379179,
      "num_input_tokens_seen": 34843980,
      "step": 1645,
      "time_per_iteration": 2.7914464473724365
    },
    {
      "auxiliary_loss_clip": 0.01216363,
      "auxiliary_loss_mlp": 0.01031762,
      "balance_loss_clip": 1.05274773,
      "balance_loss_mlp": 1.02132535,
      "epoch": 0.19791979799194373,
      "flos": 20777699041920.0,
      "grad_norm": 2.1334667348124023,
      "language_loss": 0.80659801,
      "learning_rate": 3.711578044083517e-06,
      "loss": 0.82907927,
      "num_input_tokens_seen": 34860780,
      "step": 1646,
      "time_per_iteration": 3.536680221557617
    },
    {
      "auxiliary_loss_clip": 0.01225507,
      "auxiliary_loss_mlp": 0.01037546,
      "balance_loss_clip": 1.05766153,
      "balance_loss_mlp": 1.02742457,
      "epoch": 0.1980400408825828,
      "flos": 25589010084480.0,
      "grad_norm": 1.8664892606782368,
      "language_loss": 0.74728382,
      "learning_rate": 3.7111749327681698e-06,
      "loss": 0.76991433,
      "num_input_tokens_seen": 34880815,
      "step": 1647,
      "time_per_iteration": 2.7159554958343506
    },
    {
      "auxiliary_loss_clip": 0.01228163,
      "auxiliary_loss_mlp": 0.01029042,
      "balance_loss_clip": 1.06356359,
      "balance_loss_mlp": 1.02005327,
      "epoch": 0.1981602837732219,
      "flos": 23513840622720.0,
      "grad_norm": 1.8826787227990114,
      "language_loss": 0.86687589,
      "learning_rate": 3.7107715618689455e-06,
      "loss": 0.88944787,
      "num_input_tokens_seen": 34899790,
      "step": 1648,
      "time_per_iteration": 2.6442999839782715
    },
    {
      "auxiliary_loss_clip": 0.01217038,
      "auxiliary_loss_mlp": 0.01034771,
      "balance_loss_clip": 1.05800557,
      "balance_loss_mlp": 1.02463186,
      "epoch": 0.198280526663861,
      "flos": 23185365724800.0,
      "grad_norm": 1.4687679369561466,
      "language_loss": 0.83604449,
      "learning_rate": 3.710367931447035e-06,
      "loss": 0.85856259,
      "num_input_tokens_seen": 34921570,
      "step": 1649,
      "time_per_iteration": 3.5344796180725098
    },
    {
      "auxiliary_loss_clip": 0.0122877,
      "auxiliary_loss_mlp": 0.01035171,
      "balance_loss_clip": 1.05947483,
      "balance_loss_mlp": 1.02494287,
      "epoch": 0.1984007695545001,
      "flos": 21689470897920.0,
      "grad_norm": 2.0760735982132923,
      "language_loss": 0.87203306,
      "learning_rate": 3.70996404156367e-06,
      "loss": 0.89467245,
      "num_input_tokens_seen": 34941205,
      "step": 1650,
      "time_per_iteration": 2.66867995262146
    },
    {
      "auxiliary_loss_clip": 0.01214232,
      "auxiliary_loss_mlp": 0.01037396,
      "balance_loss_clip": 1.04932332,
      "balance_loss_mlp": 1.02806807,
      "epoch": 0.19852101244513917,
      "flos": 36064008887040.0,
      "grad_norm": 1.6230644813108062,
      "language_loss": 0.73003745,
      "learning_rate": 3.7095598922801187e-06,
      "loss": 0.75255376,
      "num_input_tokens_seen": 34963280,
      "step": 1651,
      "time_per_iteration": 2.7870051860809326
    },
    {
      "auxiliary_loss_clip": 0.01220772,
      "auxiliary_loss_mlp": 0.01032422,
      "balance_loss_clip": 1.06217349,
      "balance_loss_mlp": 1.02238417,
      "epoch": 0.19864125533577828,
      "flos": 23105894883840.0,
      "grad_norm": 2.132459915867954,
      "language_loss": 0.76271605,
      "learning_rate": 3.7091554836576914e-06,
      "loss": 0.78524804,
      "num_input_tokens_seen": 34979955,
      "step": 1652,
      "time_per_iteration": 3.5327329635620117
    },
    {
      "auxiliary_loss_clip": 0.01218512,
      "auxiliary_loss_mlp": 0.01208885,
      "balance_loss_clip": 1.05899096,
      "balance_loss_mlp": 1.00039566,
      "epoch": 0.19876149822641737,
      "flos": 24608505553920.0,
      "grad_norm": 1.809783668625359,
      "language_loss": 0.82784927,
      "learning_rate": 3.708750815757736e-06,
      "loss": 0.8521232,
      "num_input_tokens_seen": 35000725,
      "step": 1653,
      "time_per_iteration": 2.6448709964752197
    },
    {
      "auxiliary_loss_clip": 0.01222157,
      "auxiliary_loss_mlp": 0.01039341,
      "balance_loss_clip": 1.0593133,
      "balance_loss_mlp": 1.02904761,
      "epoch": 0.19888174111705645,
      "flos": 32196645308160.0,
      "grad_norm": 2.1461174291157663,
      "language_loss": 0.73432958,
      "learning_rate": 3.7083458886416407e-06,
      "loss": 0.75694454,
      "num_input_tokens_seen": 35019920,
      "step": 1654,
      "time_per_iteration": 2.7387309074401855
    },
    {
      "auxiliary_loss_clip": 0.01231102,
      "auxiliary_loss_mlp": 0.01035164,
      "balance_loss_clip": 1.05355406,
      "balance_loss_mlp": 1.02529395,
      "epoch": 0.19900198400769553,
      "flos": 24608469640320.0,
      "grad_norm": 2.066674099861803,
      "language_loss": 0.8807354,
      "learning_rate": 3.707940702370832e-06,
      "loss": 0.90339804,
      "num_input_tokens_seen": 35040765,
      "step": 1655,
      "time_per_iteration": 2.68766188621521
    },
    {
      "auxiliary_loss_clip": 0.0112113,
      "auxiliary_loss_mlp": 0.01003878,
      "balance_loss_clip": 1.02653849,
      "balance_loss_mlp": 1.00162458,
      "epoch": 0.19912222689833464,
      "flos": 67915805673600.0,
      "grad_norm": 0.7624286503205981,
      "language_loss": 0.58302402,
      "learning_rate": 3.707535257006777e-06,
      "loss": 0.60427415,
      "num_input_tokens_seen": 35106390,
      "step": 1656,
      "time_per_iteration": 3.2605719566345215
    },
    {
      "auxiliary_loss_clip": 0.01225426,
      "auxiliary_loss_mlp": 0.01036254,
      "balance_loss_clip": 1.05726767,
      "balance_loss_mlp": 1.02587104,
      "epoch": 0.19924246978897373,
      "flos": 15742340916480.0,
      "grad_norm": 2.5295174397151308,
      "language_loss": 0.8904832,
      "learning_rate": 3.707129552610981e-06,
      "loss": 0.9131,
      "num_input_tokens_seen": 35125040,
      "step": 1657,
      "time_per_iteration": 2.630767583847046
    },
    {
      "auxiliary_loss_clip": 0.01217514,
      "auxiliary_loss_mlp": 0.01032211,
      "balance_loss_clip": 1.05845714,
      "balance_loss_mlp": 1.02241778,
      "epoch": 0.1993627126796128,
      "flos": 17566566986880.0,
      "grad_norm": 1.7843530228319706,
      "language_loss": 0.73672557,
      "learning_rate": 3.70672358924499e-06,
      "loss": 0.75922275,
      "num_input_tokens_seen": 35144280,
      "step": 1658,
      "time_per_iteration": 2.6841390132904053
    },
    {
      "auxiliary_loss_clip": 0.01223857,
      "auxiliary_loss_mlp": 0.01034344,
      "balance_loss_clip": 1.05976415,
      "balance_loss_mlp": 1.02435374,
      "epoch": 0.19948295557025192,
      "flos": 40843826680320.0,
      "grad_norm": 1.948792956235618,
      "language_loss": 0.78892088,
      "learning_rate": 3.706317366970386e-06,
      "loss": 0.81150287,
      "num_input_tokens_seen": 35165280,
      "step": 1659,
      "time_per_iteration": 2.980703115463257
    },
    {
      "auxiliary_loss_clip": 0.01221381,
      "auxiliary_loss_mlp": 0.01209733,
      "balance_loss_clip": 1.05906272,
      "balance_loss_mlp": 1.00040996,
      "epoch": 0.199603198460891,
      "flos": 25082418620160.0,
      "grad_norm": 2.1239450306017877,
      "language_loss": 0.84136927,
      "learning_rate": 3.705910885848795e-06,
      "loss": 0.8656804,
      "num_input_tokens_seen": 35183655,
      "step": 1660,
      "time_per_iteration": 2.6904284954071045
    },
    {
      "auxiliary_loss_clip": 0.01219865,
      "auxiliary_loss_mlp": 0.01029477,
      "balance_loss_clip": 1.05827296,
      "balance_loss_mlp": 1.01969635,
      "epoch": 0.19972344135153008,
      "flos": 20084120352000.0,
      "grad_norm": 2.9764284745247687,
      "language_loss": 0.84900236,
      "learning_rate": 3.705504145941879e-06,
      "loss": 0.87149584,
      "num_input_tokens_seen": 35201825,
      "step": 1661,
      "time_per_iteration": 2.6514923572540283
    },
    {
      "auxiliary_loss_clip": 0.01216534,
      "auxiliary_loss_mlp": 0.01030727,
      "balance_loss_clip": 1.05838692,
      "balance_loss_mlp": 1.02085662,
      "epoch": 0.1998436842421692,
      "flos": 23727472761600.0,
      "grad_norm": 1.9808715523590454,
      "language_loss": 0.78972423,
      "learning_rate": 3.7050971473113403e-06,
      "loss": 0.81219685,
      "num_input_tokens_seen": 35221600,
      "step": 1662,
      "time_per_iteration": 2.6060550212860107
    },
    {
      "auxiliary_loss_clip": 0.01216082,
      "auxiliary_loss_mlp": 0.01208997,
      "balance_loss_clip": 1.05650568,
      "balance_loss_mlp": 1.00038671,
      "epoch": 0.19996392713280828,
      "flos": 36102361633920.0,
      "grad_norm": 1.7395479782289327,
      "language_loss": 0.80251104,
      "learning_rate": 3.7046898900189196e-06,
      "loss": 0.82676184,
      "num_input_tokens_seen": 35245935,
      "step": 1663,
      "time_per_iteration": 2.806856870651245
    },
    {
      "auxiliary_loss_clip": 0.01229424,
      "auxiliary_loss_mlp": 0.01039322,
      "balance_loss_clip": 1.05804777,
      "balance_loss_mlp": 1.02899814,
      "epoch": 0.20008417002344736,
      "flos": 23657662679040.0,
      "grad_norm": 1.7612582377285646,
      "language_loss": 0.83274186,
      "learning_rate": 3.704282374126398e-06,
      "loss": 0.85542929,
      "num_input_tokens_seen": 35265615,
      "step": 1664,
      "time_per_iteration": 2.728848457336426
    },
    {
      "auxiliary_loss_clip": 0.01222168,
      "auxiliary_loss_mlp": 0.01032302,
      "balance_loss_clip": 1.05388725,
      "balance_loss_mlp": 1.02234769,
      "epoch": 0.20020441291408644,
      "flos": 21872076664320.0,
      "grad_norm": 15.293834746253607,
      "language_loss": 0.87668598,
      "learning_rate": 3.7038745996955954e-06,
      "loss": 0.89923072,
      "num_input_tokens_seen": 35284960,
      "step": 1665,
      "time_per_iteration": 3.0092761516571045
    },
    {
      "auxiliary_loss_clip": 0.01230587,
      "auxiliary_loss_mlp": 0.01035839,
      "balance_loss_clip": 1.05552053,
      "balance_loss_mlp": 1.02615941,
      "epoch": 0.20032465580472555,
      "flos": 23179691376000.0,
      "grad_norm": 2.368225984471653,
      "language_loss": 0.72278017,
      "learning_rate": 3.703466566788371e-06,
      "loss": 0.74544442,
      "num_input_tokens_seen": 35304090,
      "step": 1666,
      "time_per_iteration": 2.6859917640686035
    },
    {
      "auxiliary_loss_clip": 0.01216699,
      "auxiliary_loss_mlp": 0.01033899,
      "balance_loss_clip": 1.05829167,
      "balance_loss_mlp": 1.02334952,
      "epoch": 0.20044489869536464,
      "flos": 23873521461120.0,
      "grad_norm": 1.9092982077504852,
      "language_loss": 0.75063658,
      "learning_rate": 3.703058275466622e-06,
      "loss": 0.77314258,
      "num_input_tokens_seen": 35323325,
      "step": 1667,
      "time_per_iteration": 2.7687833309173584
    },
    {
      "auxiliary_loss_clip": 0.0122058,
      "auxiliary_loss_mlp": 0.01035857,
      "balance_loss_clip": 1.05595851,
      "balance_loss_mlp": 1.02618933,
      "epoch": 0.20056514158600372,
      "flos": 21945226711680.0,
      "grad_norm": 1.8407636788743533,
      "language_loss": 0.78387433,
      "learning_rate": 3.7026497257922877e-06,
      "loss": 0.80643868,
      "num_input_tokens_seen": 35343635,
      "step": 1668,
      "time_per_iteration": 2.689640522003174
    },
    {
      "auxiliary_loss_clip": 0.01219944,
      "auxiliary_loss_mlp": 0.01043454,
      "balance_loss_clip": 1.05009508,
      "balance_loss_mlp": 1.03326774,
      "epoch": 0.20068538447664283,
      "flos": 23879159896320.0,
      "grad_norm": 1.66486683646679,
      "language_loss": 0.85448003,
      "learning_rate": 3.7022409178273436e-06,
      "loss": 0.877114,
      "num_input_tokens_seen": 35364615,
      "step": 1669,
      "time_per_iteration": 2.8048744201660156
    },
    {
      "auxiliary_loss_clip": 0.01216175,
      "auxiliary_loss_mlp": 0.01027235,
      "balance_loss_clip": 1.05777645,
      "balance_loss_mlp": 1.01783538,
      "epoch": 0.2008056273672819,
      "flos": 18442823270400.0,
      "grad_norm": 1.9192007104248232,
      "language_loss": 0.78965712,
      "learning_rate": 3.7018318516338054e-06,
      "loss": 0.81209123,
      "num_input_tokens_seen": 35383775,
      "step": 1670,
      "time_per_iteration": 2.720865249633789
    },
    {
      "auxiliary_loss_clip": 0.0122396,
      "auxiliary_loss_mlp": 0.01027658,
      "balance_loss_clip": 1.05883718,
      "balance_loss_mlp": 1.01823425,
      "epoch": 0.200925870257921,
      "flos": 23659530186240.0,
      "grad_norm": 3.290545151734653,
      "language_loss": 0.82124478,
      "learning_rate": 3.7014225272737284e-06,
      "loss": 0.84376097,
      "num_input_tokens_seen": 35403000,
      "step": 1671,
      "time_per_iteration": 2.629793643951416
    },
    {
      "auxiliary_loss_clip": 0.01213352,
      "auxiliary_loss_mlp": 0.01031607,
      "balance_loss_clip": 1.05603099,
      "balance_loss_mlp": 1.02158141,
      "epoch": 0.20104611314856008,
      "flos": 16217115909120.0,
      "grad_norm": 2.320344831804497,
      "language_loss": 0.74327761,
      "learning_rate": 3.701012944809207e-06,
      "loss": 0.76572716,
      "num_input_tokens_seen": 35420115,
      "step": 1672,
      "time_per_iteration": 3.599170446395874
    },
    {
      "auxiliary_loss_clip": 0.01223976,
      "auxiliary_loss_mlp": 0.01208929,
      "balance_loss_clip": 1.05946267,
      "balance_loss_mlp": 1.00034988,
      "epoch": 0.2011663560391992,
      "flos": 21397373498880.0,
      "grad_norm": 1.9305471606882951,
      "language_loss": 0.79404271,
      "learning_rate": 3.700603104302374e-06,
      "loss": 0.81837177,
      "num_input_tokens_seen": 35439925,
      "step": 1673,
      "time_per_iteration": 2.660748243331909
    },
    {
      "auxiliary_loss_clip": 0.01115188,
      "auxiliary_loss_mlp": 0.01002742,
      "balance_loss_clip": 1.02152121,
      "balance_loss_mlp": 1.00081062,
      "epoch": 0.20128659892983827,
      "flos": 62229459409920.0,
      "grad_norm": 0.892243609064012,
      "language_loss": 0.55931687,
      "learning_rate": 3.7001930058154027e-06,
      "loss": 0.58049619,
      "num_input_tokens_seen": 35504885,
      "step": 1674,
      "time_per_iteration": 3.2587521076202393
    },
    {
      "auxiliary_loss_clip": 0.01223199,
      "auxiliary_loss_mlp": 0.01037528,
      "balance_loss_clip": 1.05503869,
      "balance_loss_mlp": 1.02660251,
      "epoch": 0.20140684182047736,
      "flos": 28438737448320.0,
      "grad_norm": 3.033413307292015,
      "language_loss": 0.79995704,
      "learning_rate": 3.6997826494105037e-06,
      "loss": 0.8225643,
      "num_input_tokens_seen": 35525330,
      "step": 1675,
      "time_per_iteration": 3.684298515319824
    },
    {
      "auxiliary_loss_clip": 0.01223149,
      "auxiliary_loss_mlp": 0.01030832,
      "balance_loss_clip": 1.05799437,
      "balance_loss_mlp": 1.02102101,
      "epoch": 0.20152708471111647,
      "flos": 28074064619520.0,
      "grad_norm": 2.1527762259922216,
      "language_loss": 0.70450884,
      "learning_rate": 3.6993720351499286e-06,
      "loss": 0.72704864,
      "num_input_tokens_seen": 35546455,
      "step": 1676,
      "time_per_iteration": 2.656777858734131
    },
    {
      "auxiliary_loss_clip": 0.01216579,
      "auxiliary_loss_mlp": 0.01031668,
      "balance_loss_clip": 1.0600636,
      "balance_loss_mlp": 1.02218509,
      "epoch": 0.20164732760175555,
      "flos": 23549751244800.0,
      "grad_norm": 2.1505878740771553,
      "language_loss": 0.77039111,
      "learning_rate": 3.6989611630959666e-06,
      "loss": 0.79287356,
      "num_input_tokens_seen": 35565010,
      "step": 1677,
      "time_per_iteration": 2.7806379795074463
    },
    {
      "auxiliary_loss_clip": 0.01122613,
      "auxiliary_loss_mlp": 0.01001322,
      "balance_loss_clip": 1.02556479,
      "balance_loss_mlp": 0.99915236,
      "epoch": 0.20176757049239463,
      "flos": 71100616037760.0,
      "grad_norm": 0.6807214415662175,
      "language_loss": 0.58304954,
      "learning_rate": 3.6985500333109474e-06,
      "loss": 0.60428882,
      "num_input_tokens_seen": 35633340,
      "step": 1678,
      "time_per_iteration": 4.12067985534668
    },
    {
      "auxiliary_loss_clip": 0.01216595,
      "auxiliary_loss_mlp": 0.01036153,
      "balance_loss_clip": 1.05238569,
      "balance_loss_mlp": 1.02691436,
      "epoch": 0.20188781338303372,
      "flos": 21430159637760.0,
      "grad_norm": 2.644101342924557,
      "language_loss": 0.76806957,
      "learning_rate": 3.6981386458572385e-06,
      "loss": 0.79059702,
      "num_input_tokens_seen": 35651315,
      "step": 1679,
      "time_per_iteration": 3.6140055656433105
    },
    {
      "auxiliary_loss_clip": 0.01218318,
      "auxiliary_loss_mlp": 0.01036662,
      "balance_loss_clip": 1.05320132,
      "balance_loss_mlp": 1.02617741,
      "epoch": 0.20200805627367283,
      "flos": 11546215130880.0,
      "grad_norm": 2.76352720329977,
      "language_loss": 0.76335001,
      "learning_rate": 3.6977270007972468e-06,
      "loss": 0.78589982,
      "num_input_tokens_seen": 35668850,
      "step": 1680,
      "time_per_iteration": 2.671489953994751
    },
    {
      "auxiliary_loss_clip": 0.01223917,
      "auxiliary_loss_mlp": 0.01034643,
      "balance_loss_clip": 1.05799365,
      "balance_loss_mlp": 1.02487373,
      "epoch": 0.2021282991643119,
      "flos": 28545391906560.0,
      "grad_norm": 2.443716426164837,
      "language_loss": 0.72641039,
      "learning_rate": 3.6973150981934196e-06,
      "loss": 0.74899602,
      "num_input_tokens_seen": 35690080,
      "step": 1681,
      "time_per_iteration": 2.6660385131835938
    },
    {
      "auxiliary_loss_clip": 0.01222161,
      "auxiliary_loss_mlp": 0.01036537,
      "balance_loss_clip": 1.06084275,
      "balance_loss_mlp": 1.02590919,
      "epoch": 0.202248542054951,
      "flos": 17923446564480.0,
      "grad_norm": 2.438852951946514,
      "language_loss": 0.83848679,
      "learning_rate": 3.6969029381082415e-06,
      "loss": 0.86107373,
      "num_input_tokens_seen": 35706075,
      "step": 1682,
      "time_per_iteration": 2.6048734188079834
    },
    {
      "auxiliary_loss_clip": 0.01219739,
      "auxiliary_loss_mlp": 0.01031431,
      "balance_loss_clip": 1.05803084,
      "balance_loss_mlp": 1.02153683,
      "epoch": 0.2023687849455901,
      "flos": 19864634296320.0,
      "grad_norm": 1.863680117662153,
      "language_loss": 0.7944479,
      "learning_rate": 3.696490520604237e-06,
      "loss": 0.81695956,
      "num_input_tokens_seen": 35724765,
      "step": 1683,
      "time_per_iteration": 2.674832582473755
    },
    {
      "auxiliary_loss_clip": 0.01217965,
      "auxiliary_loss_mlp": 0.01025256,
      "balance_loss_clip": 1.06084919,
      "balance_loss_mlp": 1.01598775,
      "epoch": 0.20248902783622919,
      "flos": 22564721600640.0,
      "grad_norm": 2.4751918062812317,
      "language_loss": 0.80833745,
      "learning_rate": 3.696077845743968e-06,
      "loss": 0.83076966,
      "num_input_tokens_seen": 35744355,
      "step": 1684,
      "time_per_iteration": 2.6920177936553955
    },
    {
      "auxiliary_loss_clip": 0.0122275,
      "auxiliary_loss_mlp": 0.01036511,
      "balance_loss_clip": 1.06133711,
      "balance_loss_mlp": 1.02597857,
      "epoch": 0.20260927072686827,
      "flos": 22709728805760.0,
      "grad_norm": 5.33377141204076,
      "language_loss": 0.73536819,
      "learning_rate": 3.69566491359004e-06,
      "loss": 0.7579608,
      "num_input_tokens_seen": 35761000,
      "step": 1685,
      "time_per_iteration": 2.67116379737854
    },
    {
      "auxiliary_loss_clip": 0.01218915,
      "auxiliary_loss_mlp": 0.01035646,
      "balance_loss_clip": 1.05522776,
      "balance_loss_mlp": 1.02535272,
      "epoch": 0.20272951361750738,
      "flos": 51023998650240.0,
      "grad_norm": 1.97045211897469,
      "language_loss": 0.69574547,
      "learning_rate": 3.695251724205092e-06,
      "loss": 0.71829104,
      "num_input_tokens_seen": 35785360,
      "step": 1686,
      "time_per_iteration": 2.8475899696350098
    },
    {
      "auxiliary_loss_clip": 0.01219211,
      "auxiliary_loss_mlp": 0.01032364,
      "balance_loss_clip": 1.06124783,
      "balance_loss_mlp": 1.02210641,
      "epoch": 0.20284975650814646,
      "flos": 26578133879040.0,
      "grad_norm": 1.5948654558656756,
      "language_loss": 0.86519295,
      "learning_rate": 3.6948382776518054e-06,
      "loss": 0.88770872,
      "num_input_tokens_seen": 35806065,
      "step": 1687,
      "time_per_iteration": 2.6910693645477295
    },
    {
      "auxiliary_loss_clip": 0.01232875,
      "auxiliary_loss_mlp": 0.0104246,
      "balance_loss_clip": 1.05545712,
      "balance_loss_mlp": 1.03210104,
      "epoch": 0.20296999939878554,
      "flos": 16034222833920.0,
      "grad_norm": 2.2322306199230115,
      "language_loss": 0.79879826,
      "learning_rate": 3.6944245739929e-06,
      "loss": 0.82155156,
      "num_input_tokens_seen": 35822225,
      "step": 1688,
      "time_per_iteration": 2.6629788875579834
    },
    {
      "auxiliary_loss_clip": 0.01221666,
      "auxiliary_loss_mlp": 0.01039247,
      "balance_loss_clip": 1.0599395,
      "balance_loss_mlp": 1.02861381,
      "epoch": 0.20309024228942463,
      "flos": 19203374868480.0,
      "grad_norm": 2.7981184010246385,
      "language_loss": 0.71721935,
      "learning_rate": 3.6940106132911332e-06,
      "loss": 0.73982847,
      "num_input_tokens_seen": 35839410,
      "step": 1689,
      "time_per_iteration": 2.8195366859436035
    },
    {
      "auxiliary_loss_clip": 0.01225491,
      "auxiliary_loss_mlp": 0.01032555,
      "balance_loss_clip": 1.0616076,
      "balance_loss_mlp": 1.02274394,
      "epoch": 0.20321048518006374,
      "flos": 22821087945600.0,
      "grad_norm": 2.1678833690331456,
      "language_loss": 0.89027905,
      "learning_rate": 3.6935963956093037e-06,
      "loss": 0.91285956,
      "num_input_tokens_seen": 35859495,
      "step": 1690,
      "time_per_iteration": 2.629817247390747
    },
    {
      "auxiliary_loss_clip": 0.01211601,
      "auxiliary_loss_mlp": 0.01033672,
      "balance_loss_clip": 1.05599236,
      "balance_loss_mlp": 1.02411175,
      "epoch": 0.20333072807070282,
      "flos": 19096397187840.0,
      "grad_norm": 1.982039922740344,
      "language_loss": 0.69104004,
      "learning_rate": 3.6931819210102474e-06,
      "loss": 0.71349275,
      "num_input_tokens_seen": 35878890,
      "step": 1691,
      "time_per_iteration": 2.780613422393799
    },
    {
      "auxiliary_loss_clip": 0.01222832,
      "auxiliary_loss_mlp": 0.01031809,
      "balance_loss_clip": 1.06183887,
      "balance_loss_mlp": 1.02118707,
      "epoch": 0.2034509709613419,
      "flos": 18180962144640.0,
      "grad_norm": 1.8155477701193372,
      "language_loss": 0.84357709,
      "learning_rate": 3.6927671895568402e-06,
      "loss": 0.86612356,
      "num_input_tokens_seen": 35897950,
      "step": 1692,
      "time_per_iteration": 2.6639456748962402
    },
    {
      "auxiliary_loss_clip": 0.01221569,
      "auxiliary_loss_mlp": 0.01034803,
      "balance_loss_clip": 1.06319261,
      "balance_loss_mlp": 1.02475357,
      "epoch": 0.20357121385198101,
      "flos": 22923899648640.0,
      "grad_norm": 1.9548343206769088,
      "language_loss": 0.86931622,
      "learning_rate": 3.692352201311996e-06,
      "loss": 0.89187998,
      "num_input_tokens_seen": 35916800,
      "step": 1693,
      "time_per_iteration": 2.6533050537109375
    },
    {
      "auxiliary_loss_clip": 0.01223085,
      "auxiliary_loss_mlp": 0.01031139,
      "balance_loss_clip": 1.05535936,
      "balance_loss_mlp": 1.02097011,
      "epoch": 0.2036914567426201,
      "flos": 20922131629440.0,
      "grad_norm": 1.8549533548584018,
      "language_loss": 0.76649529,
      "learning_rate": 3.6919369563386687e-06,
      "loss": 0.78903753,
      "num_input_tokens_seen": 35936600,
      "step": 1694,
      "time_per_iteration": 2.714221239089966
    },
    {
      "auxiliary_loss_clip": 0.01220873,
      "auxiliary_loss_mlp": 0.01034538,
      "balance_loss_clip": 1.05907059,
      "balance_loss_mlp": 1.02460241,
      "epoch": 0.20381169963325918,
      "flos": 15519155760000.0,
      "grad_norm": 1.8760662267034491,
      "language_loss": 0.78921592,
      "learning_rate": 3.69152145469985e-06,
      "loss": 0.81177008,
      "num_input_tokens_seen": 35953645,
      "step": 1695,
      "time_per_iteration": 2.5788509845733643
    },
    {
      "auxiliary_loss_clip": 0.01229985,
      "auxiliary_loss_mlp": 0.01043689,
      "balance_loss_clip": 1.05347145,
      "balance_loss_mlp": 1.03203034,
      "epoch": 0.20393194252389826,
      "flos": 28833143760000.0,
      "grad_norm": 2.125784808207375,
      "language_loss": 0.82232594,
      "learning_rate": 3.691105696458572e-06,
      "loss": 0.84506267,
      "num_input_tokens_seen": 35970940,
      "step": 1696,
      "time_per_iteration": 2.783858299255371
    },
    {
      "auxiliary_loss_clip": 0.01221647,
      "auxiliary_loss_mlp": 0.01030762,
      "balance_loss_clip": 1.06404829,
      "balance_loss_mlp": 1.02083766,
      "epoch": 0.20405218541453737,
      "flos": 22488554810880.0,
      "grad_norm": 2.6016656576880695,
      "language_loss": 0.67357802,
      "learning_rate": 3.690689681677904e-06,
      "loss": 0.69610214,
      "num_input_tokens_seen": 35989410,
      "step": 1697,
      "time_per_iteration": 2.5892724990844727
    },
    {
      "auxiliary_loss_clip": 0.01222769,
      "auxiliary_loss_mlp": 0.01028251,
      "balance_loss_clip": 1.05638051,
      "balance_loss_mlp": 1.01896429,
      "epoch": 0.20417242830517646,
      "flos": 25374408278400.0,
      "grad_norm": 1.8486610333694888,
      "language_loss": 0.88675582,
      "learning_rate": 3.690273410420956e-06,
      "loss": 0.90926605,
      "num_input_tokens_seen": 36009175,
      "step": 1698,
      "time_per_iteration": 2.7399191856384277
    },
    {
      "auxiliary_loss_clip": 0.01219066,
      "auxiliary_loss_mlp": 0.01036728,
      "balance_loss_clip": 1.0575614,
      "balance_loss_mlp": 1.0268153,
      "epoch": 0.20429267119581554,
      "flos": 14793078240000.0,
      "grad_norm": 3.5828709216072023,
      "language_loss": 0.76798403,
      "learning_rate": 3.689856882750875e-06,
      "loss": 0.79054201,
      "num_input_tokens_seen": 36024375,
      "step": 1699,
      "time_per_iteration": 3.413400173187256
    },
    {
      "auxiliary_loss_clip": 0.0121797,
      "auxiliary_loss_mlp": 0.01036011,
      "balance_loss_clip": 1.05922294,
      "balance_loss_mlp": 1.02674246,
      "epoch": 0.20441291408645465,
      "flos": 17781851151360.0,
      "grad_norm": 1.8891907533684722,
      "language_loss": 0.79064226,
      "learning_rate": 3.6894400987308486e-06,
      "loss": 0.81318212,
      "num_input_tokens_seen": 36041895,
      "step": 1700,
      "time_per_iteration": 2.7213683128356934
    },
    {
      "auxiliary_loss_clip": 0.01224858,
      "auxiliary_loss_mlp": 0.01035317,
      "balance_loss_clip": 1.059986,
      "balance_loss_mlp": 1.0249157,
      "epoch": 0.20453315697709373,
      "flos": 16435668211200.0,
      "grad_norm": 2.0699039406570545,
      "language_loss": 0.85266399,
      "learning_rate": 3.6890230584241024e-06,
      "loss": 0.87526572,
      "num_input_tokens_seen": 36058825,
      "step": 1701,
      "time_per_iteration": 3.595273733139038
    },
    {
      "auxiliary_loss_clip": 0.01114243,
      "auxiliary_loss_mlp": 0.01009026,
      "balance_loss_clip": 1.0229243,
      "balance_loss_mlp": 1.0070231,
      "epoch": 0.20465339986773282,
      "flos": 66713085653760.0,
      "grad_norm": 1.0827429111912805,
      "language_loss": 0.66442454,
      "learning_rate": 3.6886057618939016e-06,
      "loss": 0.68565714,
      "num_input_tokens_seen": 36121645,
      "step": 1702,
      "time_per_iteration": 3.263765573501587
    },
    {
      "auxiliary_loss_clip": 0.01216619,
      "auxiliary_loss_mlp": 0.010375,
      "balance_loss_clip": 1.05370569,
      "balance_loss_mlp": 1.02698553,
      "epoch": 0.2047736427583719,
      "flos": 41974114924800.0,
      "grad_norm": 2.3685567044928013,
      "language_loss": 0.69655967,
      "learning_rate": 3.6881882092035492e-06,
      "loss": 0.71910083,
      "num_input_tokens_seen": 36143030,
      "step": 1703,
      "time_per_iteration": 2.9270479679107666
    },
    {
      "auxiliary_loss_clip": 0.01137564,
      "auxiliary_loss_mlp": 0.01200489,
      "balance_loss_clip": 1.02548242,
      "balance_loss_mlp": 0.99961042,
      "epoch": 0.204893885649011,
      "flos": 69940878641280.0,
      "grad_norm": 0.930784350603954,
      "language_loss": 0.61304158,
      "learning_rate": 3.6877704004163873e-06,
      "loss": 0.6364221,
      "num_input_tokens_seen": 36203435,
      "step": 1704,
      "time_per_iteration": 4.224229097366333
    },
    {
      "auxiliary_loss_clip": 0.0122226,
      "auxiliary_loss_mlp": 0.01033495,
      "balance_loss_clip": 1.06252992,
      "balance_loss_mlp": 1.02271891,
      "epoch": 0.2050141285396501,
      "flos": 22200012858240.0,
      "grad_norm": 1.6996486711577015,
      "language_loss": 0.77932978,
      "learning_rate": 3.6873523355957984e-06,
      "loss": 0.80188733,
      "num_input_tokens_seen": 36222435,
      "step": 1705,
      "time_per_iteration": 3.5721559524536133
    },
    {
      "auxiliary_loss_clip": 0.01113102,
      "auxiliary_loss_mlp": 0.01001802,
      "balance_loss_clip": 1.02191663,
      "balance_loss_mlp": 0.99967974,
      "epoch": 0.20513437143028918,
      "flos": 46283721730560.0,
      "grad_norm": 0.9801988363272645,
      "language_loss": 0.64130372,
      "learning_rate": 3.686934014805201e-06,
      "loss": 0.66245276,
      "num_input_tokens_seen": 36273065,
      "step": 1706,
      "time_per_iteration": 2.9745306968688965
    },
    {
      "auxiliary_loss_clip": 0.01220252,
      "auxiliary_loss_mlp": 0.01037546,
      "balance_loss_clip": 1.06087041,
      "balance_loss_mlp": 1.02750242,
      "epoch": 0.20525461432092829,
      "flos": 21904324099200.0,
      "grad_norm": 1.779651122130598,
      "language_loss": 0.81158519,
      "learning_rate": 3.6865154381080552e-06,
      "loss": 0.83416319,
      "num_input_tokens_seen": 36293750,
      "step": 1707,
      "time_per_iteration": 2.710193395614624
    },
    {
      "auxiliary_loss_clip": 0.01228315,
      "auxiliary_loss_mlp": 0.01029254,
      "balance_loss_clip": 1.05029976,
      "balance_loss_mlp": 1.01984286,
      "epoch": 0.20537485721156737,
      "flos": 21214264942080.0,
      "grad_norm": 1.9759274031785377,
      "language_loss": 0.82343709,
      "learning_rate": 3.6860966055678585e-06,
      "loss": 0.84601283,
      "num_input_tokens_seen": 36310105,
      "step": 1708,
      "time_per_iteration": 2.714219570159912
    },
    {
      "auxiliary_loss_clip": 0.01223648,
      "auxiliary_loss_mlp": 0.01038395,
      "balance_loss_clip": 1.06214285,
      "balance_loss_mlp": 1.02775526,
      "epoch": 0.20549510010220645,
      "flos": 20191205773440.0,
      "grad_norm": 1.7858341807832523,
      "language_loss": 0.86479449,
      "learning_rate": 3.685677517248147e-06,
      "loss": 0.88741487,
      "num_input_tokens_seen": 36328995,
      "step": 1709,
      "time_per_iteration": 2.6470963954925537
    },
    {
      "auxiliary_loss_clip": 0.01222842,
      "auxiliary_loss_mlp": 0.01208753,
      "balance_loss_clip": 1.06259012,
      "balance_loss_mlp": 1.00006509,
      "epoch": 0.20561534299284553,
      "flos": 17016702612480.0,
      "grad_norm": 2.145767747612713,
      "language_loss": 0.80295408,
      "learning_rate": 3.6852581732124967e-06,
      "loss": 0.82727003,
      "num_input_tokens_seen": 36346340,
      "step": 1710,
      "time_per_iteration": 2.651798725128174
    },
    {
      "auxiliary_loss_clip": 0.01223069,
      "auxiliary_loss_mlp": 0.01036363,
      "balance_loss_clip": 1.06114268,
      "balance_loss_mlp": 1.02632523,
      "epoch": 0.20573558588348465,
      "flos": 22890467064960.0,
      "grad_norm": 2.0111488181256476,
      "language_loss": 0.76499557,
      "learning_rate": 3.6848385735245213e-06,
      "loss": 0.78758991,
      "num_input_tokens_seen": 36365430,
      "step": 1711,
      "time_per_iteration": 2.697817087173462
    },
    {
      "auxiliary_loss_clip": 0.01204536,
      "auxiliary_loss_mlp": 0.01031997,
      "balance_loss_clip": 1.0524137,
      "balance_loss_mlp": 1.02239442,
      "epoch": 0.20585582877412373,
      "flos": 24643123286400.0,
      "grad_norm": 1.864791583100162,
      "language_loss": 0.86616337,
      "learning_rate": 3.6844187182478734e-06,
      "loss": 0.8885287,
      "num_input_tokens_seen": 36386285,
      "step": 1712,
      "time_per_iteration": 2.7477614879608154
    },
    {
      "auxiliary_loss_clip": 0.01213143,
      "auxiliary_loss_mlp": 0.010292,
      "balance_loss_clip": 1.05396795,
      "balance_loss_mlp": 1.01947296,
      "epoch": 0.2059760716647628,
      "flos": 24206952435840.0,
      "grad_norm": 1.9778437236080308,
      "language_loss": 0.75328892,
      "learning_rate": 3.683998607446246e-06,
      "loss": 0.77571237,
      "num_input_tokens_seen": 36404935,
      "step": 1713,
      "time_per_iteration": 2.7126200199127197
    },
    {
      "auxiliary_loss_clip": 0.01223337,
      "auxiliary_loss_mlp": 0.01043051,
      "balance_loss_clip": 1.06220698,
      "balance_loss_mlp": 1.0339849,
      "epoch": 0.20609631455540192,
      "flos": 20229522606720.0,
      "grad_norm": 5.459825627239838,
      "language_loss": 0.75739896,
      "learning_rate": 3.6835782411833686e-06,
      "loss": 0.78006285,
      "num_input_tokens_seen": 36424455,
      "step": 1714,
      "time_per_iteration": 2.608490228652954
    },
    {
      "auxiliary_loss_clip": 0.01215463,
      "auxiliary_loss_mlp": 0.01032452,
      "balance_loss_clip": 1.05475235,
      "balance_loss_mlp": 1.02226567,
      "epoch": 0.206216557446041,
      "flos": 19864957518720.0,
      "grad_norm": 1.749393963244761,
      "language_loss": 0.74188006,
      "learning_rate": 3.68315761952301e-06,
      "loss": 0.76435924,
      "num_input_tokens_seen": 36441685,
      "step": 1715,
      "time_per_iteration": 2.7117435932159424
    },
    {
      "auxiliary_loss_clip": 0.0122089,
      "auxiliary_loss_mlp": 0.01033091,
      "balance_loss_clip": 1.0617975,
      "balance_loss_mlp": 1.02319098,
      "epoch": 0.2063368003366801,
      "flos": 24096311568000.0,
      "grad_norm": 2.454407836950627,
      "language_loss": 0.83155227,
      "learning_rate": 3.6827367425289797e-06,
      "loss": 0.85409212,
      "num_input_tokens_seen": 36461460,
      "step": 1716,
      "time_per_iteration": 2.6157257556915283
    },
    {
      "auxiliary_loss_clip": 0.01224655,
      "auxiliary_loss_mlp": 0.01033533,
      "balance_loss_clip": 1.05882764,
      "balance_loss_mlp": 1.02233887,
      "epoch": 0.2064570432273192,
      "flos": 20340163474560.0,
      "grad_norm": 2.222538857260491,
      "language_loss": 0.72508276,
      "learning_rate": 3.6823156102651225e-06,
      "loss": 0.74766457,
      "num_input_tokens_seen": 36479615,
      "step": 1717,
      "time_per_iteration": 2.6843035221099854
    },
    {
      "auxiliary_loss_clip": 0.01217055,
      "auxiliary_loss_mlp": 0.01035138,
      "balance_loss_clip": 1.05464077,
      "balance_loss_mlp": 1.02480829,
      "epoch": 0.20657728611795828,
      "flos": 20520363029760.0,
      "grad_norm": 1.739540948019074,
      "language_loss": 0.71286273,
      "learning_rate": 3.6818942227953257e-06,
      "loss": 0.7353847,
      "num_input_tokens_seen": 36500160,
      "step": 1718,
      "time_per_iteration": 2.701277732849121
    },
    {
      "auxiliary_loss_clip": 0.01226765,
      "auxiliary_loss_mlp": 0.01032751,
      "balance_loss_clip": 1.05872035,
      "balance_loss_mlp": 1.02255225,
      "epoch": 0.20669752900859736,
      "flos": 21799285752960.0,
      "grad_norm": 2.2406154421774596,
      "language_loss": 0.68874276,
      "learning_rate": 3.681472580183512e-06,
      "loss": 0.71133792,
      "num_input_tokens_seen": 36518810,
      "step": 1719,
      "time_per_iteration": 2.7568280696868896
    },
    {
      "auxiliary_loss_clip": 0.0121986,
      "auxiliary_loss_mlp": 0.0103269,
      "balance_loss_clip": 1.06137216,
      "balance_loss_mlp": 1.02348661,
      "epoch": 0.20681777189923645,
      "flos": 15122020014720.0,
      "grad_norm": 2.0193841645773,
      "language_loss": 0.86725467,
      "learning_rate": 3.6810506824936455e-06,
      "loss": 0.88978016,
      "num_input_tokens_seen": 36536890,
      "step": 1720,
      "time_per_iteration": 2.653534412384033
    },
    {
      "auxiliary_loss_clip": 0.01118543,
      "auxiliary_loss_mlp": 0.0100623,
      "balance_loss_clip": 1.01826787,
      "balance_loss_mlp": 1.00407219,
      "epoch": 0.20693801478987556,
      "flos": 56481021509760.0,
      "grad_norm": 1.0534178372367784,
      "language_loss": 0.62521625,
      "learning_rate": 3.680628529789726e-06,
      "loss": 0.64646399,
      "num_input_tokens_seen": 36589300,
      "step": 1721,
      "time_per_iteration": 3.0844991207122803
    },
    {
      "auxiliary_loss_clip": 0.01226609,
      "auxiliary_loss_mlp": 0.01039308,
      "balance_loss_clip": 1.06297755,
      "balance_loss_mlp": 1.0283469,
      "epoch": 0.20705825768051464,
      "flos": 21614201948160.0,
      "grad_norm": 1.740626012103153,
      "language_loss": 0.86319768,
      "learning_rate": 3.680206122135796e-06,
      "loss": 0.88585687,
      "num_input_tokens_seen": 36609905,
      "step": 1722,
      "time_per_iteration": 2.6721837520599365
    },
    {
      "auxiliary_loss_clip": 0.01234763,
      "auxiliary_loss_mlp": 0.01041544,
      "balance_loss_clip": 1.06130815,
      "balance_loss_mlp": 1.0322156,
      "epoch": 0.20717850057115372,
      "flos": 25848895962240.0,
      "grad_norm": 1.9668865706765075,
      "language_loss": 0.78379667,
      "learning_rate": 3.6797834595959323e-06,
      "loss": 0.80655968,
      "num_input_tokens_seen": 36629805,
      "step": 1723,
      "time_per_iteration": 2.7551510334014893
    },
    {
      "auxiliary_loss_clip": 0.01216336,
      "auxiliary_loss_mlp": 0.01040897,
      "balance_loss_clip": 1.05130839,
      "balance_loss_mlp": 1.02991176,
      "epoch": 0.20729874346179283,
      "flos": 29130807767040.0,
      "grad_norm": 2.4077749773802117,
      "language_loss": 0.7788443,
      "learning_rate": 3.679360542234254e-06,
      "loss": 0.80141664,
      "num_input_tokens_seen": 36649150,
      "step": 1724,
      "time_per_iteration": 2.7589774131774902
    },
    {
      "auxiliary_loss_clip": 0.01216909,
      "auxiliary_loss_mlp": 0.01209306,
      "balance_loss_clip": 1.05406892,
      "balance_loss_mlp": 1.00008059,
      "epoch": 0.20741898635243192,
      "flos": 29023363209600.0,
      "grad_norm": 1.6658893410680027,
      "language_loss": 0.72407234,
      "learning_rate": 3.678937370114916e-06,
      "loss": 0.74833453,
      "num_input_tokens_seen": 36668955,
      "step": 1725,
      "time_per_iteration": 3.7002131938934326
    },
    {
      "auxiliary_loss_clip": 0.01217034,
      "auxiliary_loss_mlp": 0.01027521,
      "balance_loss_clip": 1.05882239,
      "balance_loss_mlp": 1.01872349,
      "epoch": 0.207539229243071,
      "flos": 15559447841280.0,
      "grad_norm": 2.173021126501653,
      "language_loss": 0.78822923,
      "learning_rate": 3.678513943302114e-06,
      "loss": 0.81067479,
      "num_input_tokens_seen": 36685730,
      "step": 1726,
      "time_per_iteration": 2.6152429580688477
    },
    {
      "auxiliary_loss_clip": 0.01220181,
      "auxiliary_loss_mlp": 0.01033448,
      "balance_loss_clip": 1.06210017,
      "balance_loss_mlp": 1.0239588,
      "epoch": 0.20765947213371008,
      "flos": 20521081301760.0,
      "grad_norm": 2.076096417917156,
      "language_loss": 0.85266137,
      "learning_rate": 3.678090261860082e-06,
      "loss": 0.87519765,
      "num_input_tokens_seen": 36705460,
      "step": 1727,
      "time_per_iteration": 2.7597761154174805
    },
    {
      "auxiliary_loss_clip": 0.01223416,
      "auxiliary_loss_mlp": 0.01036988,
      "balance_loss_clip": 1.05168605,
      "balance_loss_mlp": 1.02756476,
      "epoch": 0.2077797150243492,
      "flos": 19354415558400.0,
      "grad_norm": 1.9241453034240121,
      "language_loss": 0.77594179,
      "learning_rate": 3.6776663258530906e-06,
      "loss": 0.79854584,
      "num_input_tokens_seen": 36724110,
      "step": 1728,
      "time_per_iteration": 3.6161036491394043
    },
    {
      "auxiliary_loss_clip": 0.01226048,
      "auxiliary_loss_mlp": 0.01032116,
      "balance_loss_clip": 1.06136501,
      "balance_loss_mlp": 1.02297878,
      "epoch": 0.20789995791498828,
      "flos": 21829952989440.0,
      "grad_norm": 1.840293985271432,
      "language_loss": 0.71454108,
      "learning_rate": 3.6772421353454516e-06,
      "loss": 0.73712271,
      "num_input_tokens_seen": 36742705,
      "step": 1729,
      "time_per_iteration": 2.7175076007843018
    },
    {
      "auxiliary_loss_clip": 0.01219921,
      "auxiliary_loss_mlp": 0.01030077,
      "balance_loss_clip": 1.06176889,
      "balance_loss_mlp": 1.02039683,
      "epoch": 0.20802020080562736,
      "flos": 23148844571520.0,
      "grad_norm": 1.870254159617275,
      "language_loss": 0.88606358,
      "learning_rate": 3.6768176904015153e-06,
      "loss": 0.90856355,
      "num_input_tokens_seen": 36762510,
      "step": 1730,
      "time_per_iteration": 2.618060350418091
    },
    {
      "auxiliary_loss_clip": 0.01221133,
      "auxiliary_loss_mlp": 0.01037049,
      "balance_loss_clip": 1.05854821,
      "balance_loss_mlp": 1.02711332,
      "epoch": 0.20814044369626647,
      "flos": 23072677781760.0,
      "grad_norm": 2.259147595116659,
      "language_loss": 0.6022836,
      "learning_rate": 3.6763929910856674e-06,
      "loss": 0.62486541,
      "num_input_tokens_seen": 36780960,
      "step": 1731,
      "time_per_iteration": 4.41132378578186
    },
    {
      "auxiliary_loss_clip": 0.01221702,
      "auxiliary_loss_mlp": 0.0103936,
      "balance_loss_clip": 1.06123173,
      "balance_loss_mlp": 1.02985263,
      "epoch": 0.20826068658690555,
      "flos": 19608016556160.0,
      "grad_norm": 2.5113438875799976,
      "language_loss": 0.77877367,
      "learning_rate": 3.6759680374623365e-06,
      "loss": 0.80138421,
      "num_input_tokens_seen": 36798875,
      "step": 1732,
      "time_per_iteration": 2.6325643062591553
    },
    {
      "auxiliary_loss_clip": 0.01220375,
      "auxiliary_loss_mlp": 0.01028009,
      "balance_loss_clip": 1.06390762,
      "balance_loss_mlp": 1.01834714,
      "epoch": 0.20838092947754464,
      "flos": 25374049142400.0,
      "grad_norm": 2.3393626398947514,
      "language_loss": 0.75485194,
      "learning_rate": 3.675542829595986e-06,
      "loss": 0.77733582,
      "num_input_tokens_seen": 36818540,
      "step": 1733,
      "time_per_iteration": 2.6656742095947266
    },
    {
      "auxiliary_loss_clip": 0.01223598,
      "auxiliary_loss_mlp": 0.01031242,
      "balance_loss_clip": 1.05909991,
      "balance_loss_mlp": 1.0214076,
      "epoch": 0.20850117236818372,
      "flos": 24061729749120.0,
      "grad_norm": 1.337921006286958,
      "language_loss": 0.79475129,
      "learning_rate": 3.6751173675511213e-06,
      "loss": 0.81729972,
      "num_input_tokens_seen": 36840585,
      "step": 1734,
      "time_per_iteration": 2.6954543590545654
    },
    {
      "auxiliary_loss_clip": 0.01218768,
      "auxiliary_loss_mlp": 0.01037224,
      "balance_loss_clip": 1.05315864,
      "balance_loss_mlp": 1.02795506,
      "epoch": 0.20862141525882283,
      "flos": 20077799558400.0,
      "grad_norm": 1.9646592114367405,
      "language_loss": 0.87643272,
      "learning_rate": 3.674691651392283e-06,
      "loss": 0.89899266,
      "num_input_tokens_seen": 36858255,
      "step": 1735,
      "time_per_iteration": 2.604410171508789
    },
    {
      "auxiliary_loss_clip": 0.01227544,
      "auxiliary_loss_mlp": 0.01040223,
      "balance_loss_clip": 1.06073546,
      "balance_loss_mlp": 1.03040028,
      "epoch": 0.2087416581494619,
      "flos": 39015183237120.0,
      "grad_norm": 2.003668601460239,
      "language_loss": 0.75806844,
      "learning_rate": 3.674265681184053e-06,
      "loss": 0.78074616,
      "num_input_tokens_seen": 36881515,
      "step": 1736,
      "time_per_iteration": 2.7767550945281982
    },
    {
      "auxiliary_loss_clip": 0.01224671,
      "auxiliary_loss_mlp": 0.01030757,
      "balance_loss_clip": 1.05734873,
      "balance_loss_mlp": 1.02153587,
      "epoch": 0.208861901040101,
      "flos": 26101994169600.0,
      "grad_norm": 1.680403404084415,
      "language_loss": 0.86718333,
      "learning_rate": 3.6738394569910504e-06,
      "loss": 0.88973761,
      "num_input_tokens_seen": 36902055,
      "step": 1737,
      "time_per_iteration": 2.7027840614318848
    },
    {
      "auxiliary_loss_clip": 0.01223487,
      "auxiliary_loss_mlp": 0.01027787,
      "balance_loss_clip": 1.06377339,
      "balance_loss_mlp": 1.01882279,
      "epoch": 0.2089821439307401,
      "flos": 28398732675840.0,
      "grad_norm": 2.6650064627146945,
      "language_loss": 0.82701385,
      "learning_rate": 3.6734129788779333e-06,
      "loss": 0.84952658,
      "num_input_tokens_seen": 36921230,
      "step": 1738,
      "time_per_iteration": 2.7582662105560303
    },
    {
      "auxiliary_loss_clip": 0.01223439,
      "auxiliary_loss_mlp": 0.01029148,
      "balance_loss_clip": 1.06045151,
      "balance_loss_mlp": 1.01961708,
      "epoch": 0.2091023868213792,
      "flos": 21069616872960.0,
      "grad_norm": 1.920659332549611,
      "language_loss": 0.90598035,
      "learning_rate": 3.6729862469093976e-06,
      "loss": 0.92850626,
      "num_input_tokens_seen": 36940325,
      "step": 1739,
      "time_per_iteration": 2.6974925994873047
    },
    {
      "auxiliary_loss_clip": 0.0120819,
      "auxiliary_loss_mlp": 0.01038354,
      "balance_loss_clip": 1.05423474,
      "balance_loss_mlp": 1.0283885,
      "epoch": 0.20922262971201827,
      "flos": 22455481363200.0,
      "grad_norm": 2.0605280676160174,
      "language_loss": 0.82844222,
      "learning_rate": 3.6725592611501782e-06,
      "loss": 0.85090768,
      "num_input_tokens_seen": 36959000,
      "step": 1740,
      "time_per_iteration": 2.7401540279388428
    },
    {
      "auxiliary_loss_clip": 0.01220637,
      "auxiliary_loss_mlp": 0.01034478,
      "balance_loss_clip": 1.05959976,
      "balance_loss_mlp": 1.02455413,
      "epoch": 0.20934287260265738,
      "flos": 27852244179840.0,
      "grad_norm": 2.51153614532572,
      "language_loss": 0.76407063,
      "learning_rate": 3.6721320216650496e-06,
      "loss": 0.78662181,
      "num_input_tokens_seen": 36979615,
      "step": 1741,
      "time_per_iteration": 2.7078335285186768
    },
    {
      "auxiliary_loss_clip": 0.01223695,
      "auxiliary_loss_mlp": 0.01038984,
      "balance_loss_clip": 1.0595696,
      "balance_loss_mlp": 1.02893472,
      "epoch": 0.20946311549329646,
      "flos": 16435309075200.0,
      "grad_norm": 1.7659617216419747,
      "language_loss": 0.83986098,
      "learning_rate": 3.6717045285188215e-06,
      "loss": 0.86248779,
      "num_input_tokens_seen": 36997310,
      "step": 1742,
      "time_per_iteration": 2.5615665912628174
    },
    {
      "auxiliary_loss_clip": 0.01211057,
      "auxiliary_loss_mlp": 0.010353,
      "balance_loss_clip": 1.05129302,
      "balance_loss_mlp": 1.02554297,
      "epoch": 0.20958335838393555,
      "flos": 22492720788480.0,
      "grad_norm": 2.110655655357508,
      "language_loss": 0.87083101,
      "learning_rate": 3.671276781776346e-06,
      "loss": 0.89329463,
      "num_input_tokens_seen": 37015965,
      "step": 1743,
      "time_per_iteration": 2.722954511642456
    },
    {
      "auxiliary_loss_clip": 0.01230854,
      "auxiliary_loss_mlp": 0.01030678,
      "balance_loss_clip": 1.05370247,
      "balance_loss_mlp": 1.02130771,
      "epoch": 0.20970360127457463,
      "flos": 25224768218880.0,
      "grad_norm": 2.010149119439798,
      "language_loss": 0.67557025,
      "learning_rate": 3.6708487815025128e-06,
      "loss": 0.69818556,
      "num_input_tokens_seen": 37036545,
      "step": 1744,
      "time_per_iteration": 2.7023634910583496
    },
    {
      "auxiliary_loss_clip": 0.01224981,
      "auxiliary_loss_mlp": 0.01029974,
      "balance_loss_clip": 1.05552244,
      "balance_loss_mlp": 1.02003169,
      "epoch": 0.20982384416521374,
      "flos": 18479164855680.0,
      "grad_norm": 2.0386554731107713,
      "language_loss": 0.74404174,
      "learning_rate": 3.6704205277622463e-06,
      "loss": 0.76659131,
      "num_input_tokens_seen": 37054985,
      "step": 1745,
      "time_per_iteration": 2.7325916290283203
    },
    {
      "auxiliary_loss_clip": 0.01225679,
      "auxiliary_loss_mlp": 0.01030992,
      "balance_loss_clip": 1.05681229,
      "balance_loss_mlp": 1.02118731,
      "epoch": 0.20994408705585282,
      "flos": 25373546352000.0,
      "grad_norm": 1.6837033707783668,
      "language_loss": 0.80463111,
      "learning_rate": 3.6699920206205146e-06,
      "loss": 0.82719779,
      "num_input_tokens_seen": 37075725,
      "step": 1746,
      "time_per_iteration": 2.744893789291382
    },
    {
      "auxiliary_loss_clip": 0.01223578,
      "auxiliary_loss_mlp": 0.01033264,
      "balance_loss_clip": 1.06051672,
      "balance_loss_mlp": 1.02380514,
      "epoch": 0.2100643299464919,
      "flos": 21320955313920.0,
      "grad_norm": 1.7240664749065155,
      "language_loss": 0.82060051,
      "learning_rate": 3.669563260142321e-06,
      "loss": 0.84316897,
      "num_input_tokens_seen": 37094615,
      "step": 1747,
      "time_per_iteration": 2.716142177581787
    },
    {
      "auxiliary_loss_clip": 0.01221064,
      "auxiliary_loss_mlp": 0.01035482,
      "balance_loss_clip": 1.06169426,
      "balance_loss_mlp": 1.0258857,
      "epoch": 0.21018457283713102,
      "flos": 19354379644800.0,
      "grad_norm": 3.0118361995678646,
      "language_loss": 0.84600377,
      "learning_rate": 3.6691342463927083e-06,
      "loss": 0.86856925,
      "num_input_tokens_seen": 37113610,
      "step": 1748,
      "time_per_iteration": 2.641356945037842
    },
    {
      "auxiliary_loss_clip": 0.012275,
      "auxiliary_loss_mlp": 0.01040688,
      "balance_loss_clip": 1.0566833,
      "balance_loss_mlp": 1.03056097,
      "epoch": 0.2103048157277701,
      "flos": 28330035914880.0,
      "grad_norm": 1.593405969832724,
      "language_loss": 0.82121789,
      "learning_rate": 3.668704979436758e-06,
      "loss": 0.84389979,
      "num_input_tokens_seen": 37133705,
      "step": 1749,
      "time_per_iteration": 2.7811458110809326
    },
    {
      "auxiliary_loss_clip": 0.01216422,
      "auxiliary_loss_mlp": 0.01034877,
      "balance_loss_clip": 1.05592155,
      "balance_loss_mlp": 1.02510202,
      "epoch": 0.21042505861840918,
      "flos": 17457290835840.0,
      "grad_norm": 2.0650219375025434,
      "language_loss": 0.7868157,
      "learning_rate": 3.668275459339588e-06,
      "loss": 0.80932868,
      "num_input_tokens_seen": 37152185,
      "step": 1750,
      "time_per_iteration": 2.6681365966796875
    },
    {
      "auxiliary_loss_clip": 0.01222816,
      "auxiliary_loss_mlp": 0.01036362,
      "balance_loss_clip": 1.06481004,
      "balance_loss_mlp": 1.02600229,
      "epoch": 0.21054530150904827,
      "flos": 14209817195520.0,
      "grad_norm": 1.7175095497184119,
      "language_loss": 0.80159611,
      "learning_rate": 3.667845686166358e-06,
      "loss": 0.82418787,
      "num_input_tokens_seen": 37169110,
      "step": 1751,
      "time_per_iteration": 2.624216079711914
    },
    {
      "auxiliary_loss_clip": 0.01208885,
      "auxiliary_loss_mlp": 0.01033615,
      "balance_loss_clip": 1.0531559,
      "balance_loss_mlp": 1.02369714,
      "epoch": 0.21066554439968738,
      "flos": 18618210403200.0,
      "grad_norm": 1.8484781490577642,
      "language_loss": 0.85998082,
      "learning_rate": 3.6674156599822634e-06,
      "loss": 0.88240588,
      "num_input_tokens_seen": 37184905,
      "step": 1752,
      "time_per_iteration": 3.6565210819244385
    },
    {
      "auxiliary_loss_clip": 0.01228754,
      "auxiliary_loss_mlp": 0.01037457,
      "balance_loss_clip": 1.05217028,
      "balance_loss_mlp": 1.02725852,
      "epoch": 0.21078578729032646,
      "flos": 23658883741440.0,
      "grad_norm": 1.9265673908527183,
      "language_loss": 0.82222843,
      "learning_rate": 3.666985380852539e-06,
      "loss": 0.84489048,
      "num_input_tokens_seen": 37203910,
      "step": 1753,
      "time_per_iteration": 3.69864559173584
    },
    {
      "auxiliary_loss_clip": 0.01222667,
      "auxiliary_loss_mlp": 0.01034257,
      "balance_loss_clip": 1.05934775,
      "balance_loss_mlp": 1.02437484,
      "epoch": 0.21090603018096554,
      "flos": 29346379240320.0,
      "grad_norm": 2.7890657360956586,
      "language_loss": 0.74932969,
      "learning_rate": 3.6665548488424576e-06,
      "loss": 0.77189898,
      "num_input_tokens_seen": 37222670,
      "step": 1754,
      "time_per_iteration": 2.697594404220581
    },
    {
      "auxiliary_loss_clip": 0.0122248,
      "auxiliary_loss_mlp": 0.01034364,
      "balance_loss_clip": 1.06301832,
      "balance_loss_mlp": 1.02395129,
      "epoch": 0.21102627307160465,
      "flos": 23261245205760.0,
      "grad_norm": 2.112871979397645,
      "language_loss": 0.88027489,
      "learning_rate": 3.6661240640173307e-06,
      "loss": 0.90284336,
      "num_input_tokens_seen": 37244140,
      "step": 1755,
      "time_per_iteration": 2.662663221359253
    },
    {
      "auxiliary_loss_clip": 0.01138352,
      "auxiliary_loss_mlp": 0.01002776,
      "balance_loss_clip": 1.03024673,
      "balance_loss_mlp": 1.00059462,
      "epoch": 0.21114651596224374,
      "flos": 54633454577280.0,
      "grad_norm": 0.8741453757742509,
      "language_loss": 0.57882959,
      "learning_rate": 3.6656930264425085e-06,
      "loss": 0.60024089,
      "num_input_tokens_seen": 37308185,
      "step": 1756,
      "time_per_iteration": 3.3157548904418945
    },
    {
      "auxiliary_loss_clip": 0.0122017,
      "auxiliary_loss_mlp": 0.01037857,
      "balance_loss_clip": 1.06231427,
      "balance_loss_mlp": 1.02800989,
      "epoch": 0.21126675885288282,
      "flos": 21543314457600.0,
      "grad_norm": 1.8844160685724702,
      "language_loss": 0.75626171,
      "learning_rate": 3.665261736183378e-06,
      "loss": 0.77884197,
      "num_input_tokens_seen": 37328220,
      "step": 1757,
      "time_per_iteration": 3.6360015869140625
    },
    {
      "auxiliary_loss_clip": 0.01225245,
      "auxiliary_loss_mlp": 0.01032435,
      "balance_loss_clip": 1.05785728,
      "balance_loss_mlp": 1.02219462,
      "epoch": 0.2113870017435219,
      "flos": 10961876678400.0,
      "grad_norm": 2.6037428684983475,
      "language_loss": 0.89330232,
      "learning_rate": 3.664830193305366e-06,
      "loss": 0.91587913,
      "num_input_tokens_seen": 37345995,
      "step": 1758,
      "time_per_iteration": 3.5512540340423584
    },
    {
      "auxiliary_loss_clip": 0.01219697,
      "auxiliary_loss_mlp": 0.01034396,
      "balance_loss_clip": 1.05268919,
      "balance_loss_mlp": 1.02442431,
      "epoch": 0.211507244634161,
      "flos": 16653825463680.0,
      "grad_norm": 2.2302227091962257,
      "language_loss": 0.77032948,
      "learning_rate": 3.6643983978739373e-06,
      "loss": 0.7928704,
      "num_input_tokens_seen": 37362610,
      "step": 1759,
      "time_per_iteration": 2.6227238178253174
    },
    {
      "auxiliary_loss_clip": 0.0121906,
      "auxiliary_loss_mlp": 0.01034179,
      "balance_loss_clip": 1.06114745,
      "balance_loss_mlp": 1.02423704,
      "epoch": 0.2116274875248001,
      "flos": 20954091755520.0,
      "grad_norm": 1.8099614043618015,
      "language_loss": 0.82660425,
      "learning_rate": 3.663966349954596e-06,
      "loss": 0.84913671,
      "num_input_tokens_seen": 37382790,
      "step": 1760,
      "time_per_iteration": 2.7383193969726562
    },
    {
      "auxiliary_loss_clip": 0.01124645,
      "auxiliary_loss_mlp": 0.01004328,
      "balance_loss_clip": 1.02663755,
      "balance_loss_mlp": 1.00211072,
      "epoch": 0.21174773041543918,
      "flos": 68196949424640.0,
      "grad_norm": 0.7879969508717017,
      "language_loss": 0.59696585,
      "learning_rate": 3.6635340496128816e-06,
      "loss": 0.61825556,
      "num_input_tokens_seen": 37439720,
      "step": 1761,
      "time_per_iteration": 3.06518816947937
    },
    {
      "auxiliary_loss_clip": 0.01223858,
      "auxiliary_loss_mlp": 0.01033631,
      "balance_loss_clip": 1.05456865,
      "balance_loss_mlp": 1.02441633,
      "epoch": 0.2118679733060783,
      "flos": 20668315150080.0,
      "grad_norm": 1.676128385796638,
      "language_loss": 0.93187201,
      "learning_rate": 3.6631014969143747e-06,
      "loss": 0.95444685,
      "num_input_tokens_seen": 37459410,
      "step": 1762,
      "time_per_iteration": 2.7191858291625977
    },
    {
      "auxiliary_loss_clip": 0.01224899,
      "auxiliary_loss_mlp": 0.01040289,
      "balance_loss_clip": 1.06432843,
      "balance_loss_mlp": 1.03094339,
      "epoch": 0.21198821619671737,
      "flos": 23223431162880.0,
      "grad_norm": 1.8847874196176235,
      "language_loss": 0.88782543,
      "learning_rate": 3.662668691924693e-06,
      "loss": 0.91047728,
      "num_input_tokens_seen": 37480460,
      "step": 1763,
      "time_per_iteration": 2.7232325077056885
    },
    {
      "auxiliary_loss_clip": 0.01224972,
      "auxiliary_loss_mlp": 0.01041123,
      "balance_loss_clip": 1.05686474,
      "balance_loss_mlp": 1.03037059,
      "epoch": 0.21210845908735645,
      "flos": 24498547044480.0,
      "grad_norm": 2.457058503024352,
      "language_loss": 0.71264005,
      "learning_rate": 3.6622356347094927e-06,
      "loss": 0.73530096,
      "num_input_tokens_seen": 37502025,
      "step": 1764,
      "time_per_iteration": 2.735189199447632
    },
    {
      "auxiliary_loss_clip": 0.01225297,
      "auxiliary_loss_mlp": 0.01035855,
      "balance_loss_clip": 1.05341542,
      "balance_loss_mlp": 1.02480435,
      "epoch": 0.21222870197799554,
      "flos": 27089789160960.0,
      "grad_norm": 2.2434770363500602,
      "language_loss": 0.79002035,
      "learning_rate": 3.6618023253344684e-06,
      "loss": 0.81263185,
      "num_input_tokens_seen": 37520885,
      "step": 1765,
      "time_per_iteration": 2.7939138412475586
    },
    {
      "auxiliary_loss_clip": 0.01221685,
      "auxiliary_loss_mlp": 0.01035376,
      "balance_loss_clip": 1.0597738,
      "balance_loss_mlp": 1.02511168,
      "epoch": 0.21234894486863465,
      "flos": 16873850223360.0,
      "grad_norm": 1.628458471292607,
      "language_loss": 0.83293343,
      "learning_rate": 3.6613687638653527e-06,
      "loss": 0.85550404,
      "num_input_tokens_seen": 37539055,
      "step": 1766,
      "time_per_iteration": 2.5917437076568604
    },
    {
      "auxiliary_loss_clip": 0.01219693,
      "auxiliary_loss_mlp": 0.01033779,
      "balance_loss_clip": 1.05872059,
      "balance_loss_mlp": 1.02344394,
      "epoch": 0.21246918775927373,
      "flos": 23474949171840.0,
      "grad_norm": 1.8986514447302516,
      "language_loss": 0.7802791,
      "learning_rate": 3.660934950367916e-06,
      "loss": 0.80281383,
      "num_input_tokens_seen": 37558300,
      "step": 1767,
      "time_per_iteration": 2.7290830612182617
    },
    {
      "auxiliary_loss_clip": 0.01225118,
      "auxiliary_loss_mlp": 0.01034225,
      "balance_loss_clip": 1.06226659,
      "balance_loss_mlp": 1.02416933,
      "epoch": 0.21258943064991281,
      "flos": 22382295402240.0,
      "grad_norm": 1.9943633436102766,
      "language_loss": 0.83583915,
      "learning_rate": 3.660500884907968e-06,
      "loss": 0.85843259,
      "num_input_tokens_seen": 37579040,
      "step": 1768,
      "time_per_iteration": 2.704437494277954
    },
    {
      "auxiliary_loss_clip": 0.01122908,
      "auxiliary_loss_mlp": 0.01003391,
      "balance_loss_clip": 1.0268563,
      "balance_loss_mlp": 1.00116146,
      "epoch": 0.21270967354055192,
      "flos": 59440168679040.0,
      "grad_norm": 0.8276566744973054,
      "language_loss": 0.60031754,
      "learning_rate": 3.660066567551356e-06,
      "loss": 0.62158054,
      "num_input_tokens_seen": 37639185,
      "step": 1769,
      "time_per_iteration": 3.1293692588806152
    },
    {
      "auxiliary_loss_clip": 0.01220947,
      "auxiliary_loss_mlp": 0.01209449,
      "balance_loss_clip": 1.05993152,
      "balance_loss_mlp": 1.00007176,
      "epoch": 0.212829916431191,
      "flos": 21544032729600.0,
      "grad_norm": 2.327130530228154,
      "language_loss": 0.84603608,
      "learning_rate": 3.6596319983639657e-06,
      "loss": 0.87033999,
      "num_input_tokens_seen": 37657765,
      "step": 1770,
      "time_per_iteration": 2.626560926437378
    },
    {
      "auxiliary_loss_clip": 0.01226258,
      "auxiliary_loss_mlp": 0.01209631,
      "balance_loss_clip": 1.0583396,
      "balance_loss_mlp": 1.00002146,
      "epoch": 0.2129501593218301,
      "flos": 28987739896320.0,
      "grad_norm": 1.5274753450374645,
      "language_loss": 0.86607832,
      "learning_rate": 3.6591971774117214e-06,
      "loss": 0.89043719,
      "num_input_tokens_seen": 37680740,
      "step": 1771,
      "time_per_iteration": 2.732079029083252
    },
    {
      "auxiliary_loss_clip": 0.01226335,
      "auxiliary_loss_mlp": 0.0104284,
      "balance_loss_clip": 1.06148481,
      "balance_loss_mlp": 1.03240299,
      "epoch": 0.2130704022124692,
      "flos": 18806993308800.0,
      "grad_norm": 2.1002367015983125,
      "language_loss": 0.80343884,
      "learning_rate": 3.6587621047605833e-06,
      "loss": 0.82613063,
      "num_input_tokens_seen": 37697910,
      "step": 1772,
      "time_per_iteration": 2.602562189102173
    },
    {
      "auxiliary_loss_clip": 0.01220588,
      "auxiliary_loss_mlp": 0.01033401,
      "balance_loss_clip": 1.05984068,
      "balance_loss_mlp": 1.02406657,
      "epoch": 0.21319064510310828,
      "flos": 13918150759680.0,
      "grad_norm": 1.911112517252714,
      "language_loss": 0.86641181,
      "learning_rate": 3.6583267804765542e-06,
      "loss": 0.8889516,
      "num_input_tokens_seen": 37712245,
      "step": 1773,
      "time_per_iteration": 2.705233573913574
    },
    {
      "auxiliary_loss_clip": 0.01217147,
      "auxiliary_loss_mlp": 0.01039119,
      "balance_loss_clip": 1.05669403,
      "balance_loss_mlp": 1.0280323,
      "epoch": 0.21331088799374737,
      "flos": 20959694277120.0,
      "grad_norm": 1.7381221841836227,
      "language_loss": 0.85876298,
      "learning_rate": 3.6578912046256702e-06,
      "loss": 0.8813256,
      "num_input_tokens_seen": 37730765,
      "step": 1774,
      "time_per_iteration": 2.593783140182495
    },
    {
      "auxiliary_loss_clip": 0.01218741,
      "auxiliary_loss_mlp": 0.01032226,
      "balance_loss_clip": 1.05141604,
      "balance_loss_mlp": 1.0211637,
      "epoch": 0.21343113088438645,
      "flos": 18624638937600.0,
      "grad_norm": 2.209797857041454,
      "language_loss": 0.7610817,
      "learning_rate": 3.6574553772740083e-06,
      "loss": 0.78359133,
      "num_input_tokens_seen": 37748695,
      "step": 1775,
      "time_per_iteration": 2.7364935874938965
    },
    {
      "auxiliary_loss_clip": 0.01116894,
      "auxiliary_loss_mlp": 0.01007177,
      "balance_loss_clip": 1.0266242,
      "balance_loss_mlp": 1.00487614,
      "epoch": 0.21355137377502556,
      "flos": 67413128791680.0,
      "grad_norm": 0.8602912827350035,
      "language_loss": 0.61905921,
      "learning_rate": 3.657019298487684e-06,
      "loss": 0.64029992,
      "num_input_tokens_seen": 37813705,
      "step": 1776,
      "time_per_iteration": 3.204710006713867
    },
    {
      "auxiliary_loss_clip": 0.01228911,
      "auxiliary_loss_mlp": 0.01210134,
      "balance_loss_clip": 1.05986869,
      "balance_loss_mlp": 1.00012386,
      "epoch": 0.21367161666566464,
      "flos": 34532095697280.0,
      "grad_norm": 1.6931906314696885,
      "language_loss": 0.83566564,
      "learning_rate": 3.6565829683328495e-06,
      "loss": 0.86005604,
      "num_input_tokens_seen": 37836330,
      "step": 1777,
      "time_per_iteration": 2.7314536571502686
    },
    {
      "auxiliary_loss_clip": 0.01215305,
      "auxiliary_loss_mlp": 0.01034016,
      "balance_loss_clip": 1.05744815,
      "balance_loss_mlp": 1.02324557,
      "epoch": 0.21379185955630373,
      "flos": 18989347680000.0,
      "grad_norm": 1.9029294751150239,
      "language_loss": 0.86095691,
      "learning_rate": 3.6561463868756965e-06,
      "loss": 0.88345015,
      "num_input_tokens_seen": 37855030,
      "step": 1778,
      "time_per_iteration": 3.5723888874053955
    },
    {
      "auxiliary_loss_clip": 0.01223444,
      "auxiliary_loss_mlp": 0.01036416,
      "balance_loss_clip": 1.06246638,
      "balance_loss_mlp": 1.02581811,
      "epoch": 0.21391210244694284,
      "flos": 28218497207040.0,
      "grad_norm": 1.7174922597461457,
      "language_loss": 0.78435105,
      "learning_rate": 3.655709554182452e-06,
      "loss": 0.80694968,
      "num_input_tokens_seen": 37875370,
      "step": 1779,
      "time_per_iteration": 2.7498652935028076
    },
    {
      "auxiliary_loss_clip": 0.01224058,
      "auxiliary_loss_mlp": 0.0103325,
      "balance_loss_clip": 1.05778086,
      "balance_loss_mlp": 1.02314675,
      "epoch": 0.21403234533758192,
      "flos": 17455064192640.0,
      "grad_norm": 1.6696188306692192,
      "language_loss": 0.84820789,
      "learning_rate": 3.6552724703193855e-06,
      "loss": 0.870781,
      "num_input_tokens_seen": 37892560,
      "step": 1780,
      "time_per_iteration": 3.5443034172058105
    },
    {
      "auxiliary_loss_clip": 0.01133395,
      "auxiliary_loss_mlp": 0.01010218,
      "balance_loss_clip": 1.02132809,
      "balance_loss_mlp": 1.00794065,
      "epoch": 0.214152588228221,
      "flos": 51637606686720.0,
      "grad_norm": 0.7929416132739361,
      "language_loss": 0.55937934,
      "learning_rate": 3.654835135352801e-06,
      "loss": 0.58081543,
      "num_input_tokens_seen": 37947370,
      "step": 1781,
      "time_per_iteration": 3.1288259029388428
    },
    {
      "auxiliary_loss_clip": 0.01225593,
      "auxiliary_loss_mlp": 0.01031507,
      "balance_loss_clip": 1.05075765,
      "balance_loss_mlp": 1.02114773,
      "epoch": 0.21427283111886009,
      "flos": 19496154625920.0,
      "grad_norm": 2.1668383347807096,
      "language_loss": 0.87774026,
      "learning_rate": 3.654397549349043e-06,
      "loss": 0.90031123,
      "num_input_tokens_seen": 37964745,
      "step": 1782,
      "time_per_iteration": 2.672586441040039
    },
    {
      "auxiliary_loss_clip": 0.01220771,
      "auxiliary_loss_mlp": 0.01037763,
      "balance_loss_clip": 1.05936527,
      "balance_loss_mlp": 1.02712989,
      "epoch": 0.2143930740094992,
      "flos": 20084802710400.0,
      "grad_norm": 2.222074931272223,
      "language_loss": 0.75451499,
      "learning_rate": 3.653959712374491e-06,
      "loss": 0.77710032,
      "num_input_tokens_seen": 37982850,
      "step": 1783,
      "time_per_iteration": 2.6054067611694336
    },
    {
      "auxiliary_loss_clip": 0.01218019,
      "auxiliary_loss_mlp": 0.01026,
      "balance_loss_clip": 1.05641389,
      "balance_loss_mlp": 1.01599884,
      "epoch": 0.21451331690013828,
      "flos": 21798603394560.0,
      "grad_norm": 1.6591263368953135,
      "language_loss": 0.83151376,
      "learning_rate": 3.6535216244955663e-06,
      "loss": 0.85395396,
      "num_input_tokens_seen": 38002745,
      "step": 1784,
      "time_per_iteration": 3.6129395961761475
    },
    {
      "auxiliary_loss_clip": 0.01219579,
      "auxiliary_loss_mlp": 0.01035554,
      "balance_loss_clip": 1.05759263,
      "balance_loss_mlp": 1.02557647,
      "epoch": 0.21463355979077736,
      "flos": 32853882412800.0,
      "grad_norm": 2.281615758986861,
      "language_loss": 0.71123433,
      "learning_rate": 3.653083285778726e-06,
      "loss": 0.73378563,
      "num_input_tokens_seen": 38024115,
      "step": 1785,
      "time_per_iteration": 3.664445638656616
    },
    {
      "auxiliary_loss_clip": 0.01225075,
      "auxiliary_loss_mlp": 0.01035801,
      "balance_loss_clip": 1.0598501,
      "balance_loss_mlp": 1.02472627,
      "epoch": 0.21475380268141647,
      "flos": 21543817248000.0,
      "grad_norm": 2.973300315902244,
      "language_loss": 0.81471598,
      "learning_rate": 3.6526446962904653e-06,
      "loss": 0.83732474,
      "num_input_tokens_seen": 38042830,
      "step": 1786,
      "time_per_iteration": 2.701495409011841
    },
    {
      "auxiliary_loss_clip": 0.01215478,
      "auxiliary_loss_mlp": 0.01045689,
      "balance_loss_clip": 1.0590198,
      "balance_loss_mlp": 1.03560376,
      "epoch": 0.21487404557205556,
      "flos": 32159082660480.0,
      "grad_norm": 1.5179165950175202,
      "language_loss": 0.7428242,
      "learning_rate": 3.652205856097318e-06,
      "loss": 0.76543587,
      "num_input_tokens_seen": 38066015,
      "step": 1787,
      "time_per_iteration": 2.7093701362609863
    },
    {
      "auxiliary_loss_clip": 0.01233385,
      "auxiliary_loss_mlp": 0.01209012,
      "balance_loss_clip": 1.05681348,
      "balance_loss_mlp": 1.00010312,
      "epoch": 0.21499428846269464,
      "flos": 12673091583360.0,
      "grad_norm": 1.7546003094251403,
      "language_loss": 0.79557955,
      "learning_rate": 3.651766765265856e-06,
      "loss": 0.82000357,
      "num_input_tokens_seen": 38083025,
      "step": 1788,
      "time_per_iteration": 2.7502052783966064
    },
    {
      "auxiliary_loss_clip": 0.01216013,
      "auxiliary_loss_mlp": 0.01025291,
      "balance_loss_clip": 1.05504346,
      "balance_loss_mlp": 1.01555157,
      "epoch": 0.21511453135333372,
      "flos": 23471573293440.0,
      "grad_norm": 2.304157684078906,
      "language_loss": 0.81460208,
      "learning_rate": 3.65132742386269e-06,
      "loss": 0.83701515,
      "num_input_tokens_seen": 38098245,
      "step": 1789,
      "time_per_iteration": 2.6836097240448
    },
    {
      "auxiliary_loss_clip": 0.01217968,
      "auxiliary_loss_mlp": 0.01033502,
      "balance_loss_clip": 1.05916023,
      "balance_loss_mlp": 1.02296352,
      "epoch": 0.21523477424397283,
      "flos": 26943560893440.0,
      "grad_norm": 1.8570715649818246,
      "language_loss": 0.84889352,
      "learning_rate": 3.6508878319544656e-06,
      "loss": 0.87140816,
      "num_input_tokens_seen": 38118460,
      "step": 1790,
      "time_per_iteration": 2.833345651626587
    },
    {
      "auxiliary_loss_clip": 0.01210584,
      "auxiliary_loss_mlp": 0.0103809,
      "balance_loss_clip": 1.05662084,
      "balance_loss_mlp": 1.0284698,
      "epoch": 0.21535501713461191,
      "flos": 18916161719040.0,
      "grad_norm": 2.4612626262771395,
      "language_loss": 0.81915331,
      "learning_rate": 3.65044798960787e-06,
      "loss": 0.84164006,
      "num_input_tokens_seen": 38136800,
      "step": 1791,
      "time_per_iteration": 2.8601300716400146
    },
    {
      "auxiliary_loss_clip": 0.01214944,
      "auxiliary_loss_mlp": 0.01031911,
      "balance_loss_clip": 1.05277276,
      "balance_loss_mlp": 1.02209365,
      "epoch": 0.215475260025251,
      "flos": 17895113712000.0,
      "grad_norm": 1.8206802925140266,
      "language_loss": 0.7864719,
      "learning_rate": 3.650007896889627e-06,
      "loss": 0.80894047,
      "num_input_tokens_seen": 38155380,
      "step": 1792,
      "time_per_iteration": 2.6532864570617676
    },
    {
      "auxiliary_loss_clip": 0.0121924,
      "auxiliary_loss_mlp": 0.01034764,
      "balance_loss_clip": 1.06359744,
      "balance_loss_mlp": 1.0253644,
      "epoch": 0.2155955029158901,
      "flos": 16654292340480.0,
      "grad_norm": 1.695031527457591,
      "language_loss": 0.80695826,
      "learning_rate": 3.6495675538664974e-06,
      "loss": 0.82949829,
      "num_input_tokens_seen": 38174395,
      "step": 1793,
      "time_per_iteration": 2.57817006111145
    },
    {
      "auxiliary_loss_clip": 0.01221252,
      "auxiliary_loss_mlp": 0.01030207,
      "balance_loss_clip": 1.05486238,
      "balance_loss_mlp": 1.02022314,
      "epoch": 0.2157157458065292,
      "flos": 23621213352960.0,
      "grad_norm": 1.7640181355291962,
      "language_loss": 0.82612723,
      "learning_rate": 3.649126960605282e-06,
      "loss": 0.84864187,
      "num_input_tokens_seen": 38195380,
      "step": 1794,
      "time_per_iteration": 2.6367428302764893
    },
    {
      "auxiliary_loss_clip": 0.01216683,
      "auxiliary_loss_mlp": 0.01030699,
      "balance_loss_clip": 1.05607581,
      "balance_loss_mlp": 1.02066171,
      "epoch": 0.21583598869716827,
      "flos": 22127078292480.0,
      "grad_norm": 2.4926286773316635,
      "language_loss": 0.83882225,
      "learning_rate": 3.6486861171728174e-06,
      "loss": 0.86129612,
      "num_input_tokens_seen": 38213775,
      "step": 1795,
      "time_per_iteration": 2.6412909030914307
    },
    {
      "auxiliary_loss_clip": 0.01222143,
      "auxiliary_loss_mlp": 0.01032226,
      "balance_loss_clip": 1.05204773,
      "balance_loss_mlp": 1.02212322,
      "epoch": 0.21595623158780738,
      "flos": 23441229279360.0,
      "grad_norm": 1.5817998287781503,
      "language_loss": 0.78849757,
      "learning_rate": 3.6482450236359803e-06,
      "loss": 0.81104136,
      "num_input_tokens_seen": 38235630,
      "step": 1796,
      "time_per_iteration": 2.6796040534973145
    },
    {
      "auxiliary_loss_clip": 0.01219318,
      "auxiliary_loss_mlp": 0.0103832,
      "balance_loss_clip": 1.06110084,
      "balance_loss_mlp": 1.02920079,
      "epoch": 0.21607647447844647,
      "flos": 26906501036160.0,
      "grad_norm": 2.31158057246688,
      "language_loss": 0.77808237,
      "learning_rate": 3.647803680061683e-06,
      "loss": 0.80065876,
      "num_input_tokens_seen": 38256045,
      "step": 1797,
      "time_per_iteration": 2.684816837310791
    },
    {
      "auxiliary_loss_clip": 0.01222035,
      "auxiliary_loss_mlp": 0.01037781,
      "balance_loss_clip": 1.05831051,
      "balance_loss_mlp": 1.02674222,
      "epoch": 0.21619671736908555,
      "flos": 14495378319360.0,
      "grad_norm": 3.5052623516354218,
      "language_loss": 0.74944508,
      "learning_rate": 3.6473620865168776e-06,
      "loss": 0.77204323,
      "num_input_tokens_seen": 38272915,
      "step": 1798,
      "time_per_iteration": 2.6069419384002686
    },
    {
      "auxiliary_loss_clip": 0.01224594,
      "auxiliary_loss_mlp": 0.01033742,
      "balance_loss_clip": 1.06189489,
      "balance_loss_mlp": 1.0241878,
      "epoch": 0.21631696025972463,
      "flos": 17931096161280.0,
      "grad_norm": 1.8380657014048667,
      "language_loss": 0.81694329,
      "learning_rate": 3.646920243068554e-06,
      "loss": 0.83952665,
      "num_input_tokens_seen": 38290810,
      "step": 1799,
      "time_per_iteration": 2.6864867210388184
    },
    {
      "auxiliary_loss_clip": 0.01205764,
      "auxiliary_loss_mlp": 0.01035699,
      "balance_loss_clip": 1.05446482,
      "balance_loss_mlp": 1.02638257,
      "epoch": 0.21643720315036374,
      "flos": 24462385027200.0,
      "grad_norm": 1.7523080216204487,
      "language_loss": 0.7470364,
      "learning_rate": 3.6464781497837384e-06,
      "loss": 0.76945102,
      "num_input_tokens_seen": 38312785,
      "step": 1800,
      "time_per_iteration": 2.664900064468384
    },
    {
      "auxiliary_loss_clip": 0.01224669,
      "auxiliary_loss_mlp": 0.01041783,
      "balance_loss_clip": 1.05638373,
      "balance_loss_mlp": 1.03190005,
      "epoch": 0.21655744604100283,
      "flos": 28474432588800.0,
      "grad_norm": 2.329056852900094,
      "language_loss": 0.72742575,
      "learning_rate": 3.6460358067294965e-06,
      "loss": 0.75009024,
      "num_input_tokens_seen": 38334015,
      "step": 1801,
      "time_per_iteration": 2.7467293739318848
    },
    {
      "auxiliary_loss_clip": 0.01220682,
      "auxiliary_loss_mlp": 0.01030832,
      "balance_loss_clip": 1.05933118,
      "balance_loss_mlp": 1.02055633,
      "epoch": 0.2166776889316419,
      "flos": 20152960767360.0,
      "grad_norm": 2.346431810160201,
      "language_loss": 0.77761734,
      "learning_rate": 3.645593213972932e-06,
      "loss": 0.80013239,
      "num_input_tokens_seen": 38352920,
      "step": 1802,
      "time_per_iteration": 2.695174217224121
    },
    {
      "auxiliary_loss_clip": 0.01216478,
      "auxiliary_loss_mlp": 0.01031288,
      "balance_loss_clip": 1.05870152,
      "balance_loss_mlp": 1.02086926,
      "epoch": 0.21679793182228102,
      "flos": 15193482122880.0,
      "grad_norm": 2.16218054269024,
      "language_loss": 0.80167484,
      "learning_rate": 3.6451503715811852e-06,
      "loss": 0.82415247,
      "num_input_tokens_seen": 38371230,
      "step": 1803,
      "time_per_iteration": 2.6388683319091797
    },
    {
      "auxiliary_loss_clip": 0.01221503,
      "auxiliary_loss_mlp": 0.01032825,
      "balance_loss_clip": 1.06088567,
      "balance_loss_mlp": 1.02452207,
      "epoch": 0.2169181747129201,
      "flos": 17384464010880.0,
      "grad_norm": 2.2845298537749676,
      "language_loss": 0.80565286,
      "learning_rate": 3.6447072796214345e-06,
      "loss": 0.82819605,
      "num_input_tokens_seen": 38389795,
      "step": 1804,
      "time_per_iteration": 2.6357779502868652
    },
    {
      "auxiliary_loss_clip": 0.01138115,
      "auxiliary_loss_mlp": 0.0102,
      "balance_loss_clip": 1.02745366,
      "balance_loss_mlp": 1.01730633,
      "epoch": 0.21703841760355919,
      "flos": 58760955429120.0,
      "grad_norm": 0.9641067396149615,
      "language_loss": 0.63210148,
      "learning_rate": 3.644263938160898e-06,
      "loss": 0.65368265,
      "num_input_tokens_seen": 38445760,
      "step": 1805,
      "time_per_iteration": 4.075406074523926
    },
    {
      "auxiliary_loss_clip": 0.01221496,
      "auxiliary_loss_mlp": 0.01032722,
      "balance_loss_clip": 1.05660331,
      "balance_loss_mlp": 1.02218366,
      "epoch": 0.21715866049419827,
      "flos": 22418457419520.0,
      "grad_norm": 53.53532446738045,
      "language_loss": 0.71999419,
      "learning_rate": 3.6438203472668293e-06,
      "loss": 0.74253643,
      "num_input_tokens_seen": 38465405,
      "step": 1806,
      "time_per_iteration": 3.6420342922210693
    },
    {
      "auxiliary_loss_clip": 0.0122463,
      "auxiliary_loss_mlp": 0.01033428,
      "balance_loss_clip": 1.05797744,
      "balance_loss_mlp": 1.02367115,
      "epoch": 0.21727890338483738,
      "flos": 17237732952960.0,
      "grad_norm": 1.842850814185135,
      "language_loss": 0.81963038,
      "learning_rate": 3.6433765070065206e-06,
      "loss": 0.84221095,
      "num_input_tokens_seen": 38483195,
      "step": 1807,
      "time_per_iteration": 2.6567842960357666
    },
    {
      "auxiliary_loss_clip": 0.01219054,
      "auxiliary_loss_mlp": 0.01032606,
      "balance_loss_clip": 1.06040692,
      "balance_loss_mlp": 1.02231836,
      "epoch": 0.21739914627547646,
      "flos": 13434792416640.0,
      "grad_norm": 2.6136872394202615,
      "language_loss": 0.8817994,
      "learning_rate": 3.6429324174473025e-06,
      "loss": 0.90431601,
      "num_input_tokens_seen": 38496735,
      "step": 1808,
      "time_per_iteration": 2.550487995147705
    },
    {
      "auxiliary_loss_clip": 0.01220513,
      "auxiliary_loss_mlp": 0.01036269,
      "balance_loss_clip": 1.05673623,
      "balance_loss_mlp": 1.0263027,
      "epoch": 0.21751938916611555,
      "flos": 20959514709120.0,
      "grad_norm": 2.306432924761828,
      "language_loss": 0.84828508,
      "learning_rate": 3.6424880786565425e-06,
      "loss": 0.87085295,
      "num_input_tokens_seen": 38512880,
      "step": 1809,
      "time_per_iteration": 2.6647984981536865
    },
    {
      "auxiliary_loss_clip": 0.01221325,
      "auxiliary_loss_mlp": 0.01038153,
      "balance_loss_clip": 1.05578232,
      "balance_loss_mlp": 1.02734661,
      "epoch": 0.21763963205675466,
      "flos": 27599936071680.0,
      "grad_norm": 2.6696072794070176,
      "language_loss": 0.79688668,
      "learning_rate": 3.6420434907016482e-06,
      "loss": 0.81948149,
      "num_input_tokens_seen": 38532570,
      "step": 1810,
      "time_per_iteration": 3.618844985961914
    },
    {
      "auxiliary_loss_clip": 0.01221878,
      "auxiliary_loss_mlp": 0.01035199,
      "balance_loss_clip": 1.06318688,
      "balance_loss_mlp": 1.025442,
      "epoch": 0.21775987494739374,
      "flos": 21430411032960.0,
      "grad_norm": 1.6538082221481727,
      "language_loss": 0.81650555,
      "learning_rate": 3.6415986536500606e-06,
      "loss": 0.83907634,
      "num_input_tokens_seen": 38550900,
      "step": 1811,
      "time_per_iteration": 3.540147066116333
    },
    {
      "auxiliary_loss_clip": 0.0121872,
      "auxiliary_loss_mlp": 0.01040091,
      "balance_loss_clip": 1.05787599,
      "balance_loss_mlp": 1.03028631,
      "epoch": 0.21788011783803282,
      "flos": 18332972501760.0,
      "grad_norm": 1.6775264332190978,
      "language_loss": 0.80831981,
      "learning_rate": 3.641153567569263e-06,
      "loss": 0.83090788,
      "num_input_tokens_seen": 38569215,
      "step": 1812,
      "time_per_iteration": 2.6787898540496826
    },
    {
      "auxiliary_loss_clip": 0.01214071,
      "auxiliary_loss_mlp": 0.01028653,
      "balance_loss_clip": 1.05797076,
      "balance_loss_mlp": 1.01909232,
      "epoch": 0.2180003607286719,
      "flos": 30262748037120.0,
      "grad_norm": 2.2849550609399176,
      "language_loss": 0.95534492,
      "learning_rate": 3.640708232526774e-06,
      "loss": 0.97777212,
      "num_input_tokens_seen": 38587870,
      "step": 1813,
      "time_per_iteration": 2.655306339263916
    },
    {
      "auxiliary_loss_clip": 0.01218854,
      "auxiliary_loss_mlp": 0.0103382,
      "balance_loss_clip": 1.04528117,
      "balance_loss_mlp": 1.02338934,
      "epoch": 0.21812060361931102,
      "flos": 25480272637440.0,
      "grad_norm": 1.8154444002345247,
      "language_loss": 0.78544801,
      "learning_rate": 3.6402626485901504e-06,
      "loss": 0.80797476,
      "num_input_tokens_seen": 38606965,
      "step": 1814,
      "time_per_iteration": 2.8198821544647217
    },
    {
      "auxiliary_loss_clip": 0.01214405,
      "auxiliary_loss_mlp": 0.01031961,
      "balance_loss_clip": 1.06064022,
      "balance_loss_mlp": 1.02240622,
      "epoch": 0.2182408465099501,
      "flos": 21908166854400.0,
      "grad_norm": 1.9527311139141614,
      "language_loss": 0.78391778,
      "learning_rate": 3.639816815826988e-06,
      "loss": 0.80638146,
      "num_input_tokens_seen": 38626290,
      "step": 1815,
      "time_per_iteration": 2.5861921310424805
    },
    {
      "auxiliary_loss_clip": 0.01217495,
      "auxiliary_loss_mlp": 0.01027755,
      "balance_loss_clip": 1.05751228,
      "balance_loss_mlp": 1.01843262,
      "epoch": 0.21836108940058918,
      "flos": 23657339456640.0,
      "grad_norm": 1.7831031229015308,
      "language_loss": 0.78083706,
      "learning_rate": 3.6393707343049176e-06,
      "loss": 0.80328959,
      "num_input_tokens_seen": 38646620,
      "step": 1816,
      "time_per_iteration": 2.6953284740448
    },
    {
      "auxiliary_loss_clip": 0.01221106,
      "auxiliary_loss_mlp": 0.01029878,
      "balance_loss_clip": 1.05797338,
      "balance_loss_mlp": 1.02028728,
      "epoch": 0.2184813322912283,
      "flos": 24681009156480.0,
      "grad_norm": 2.313632498368903,
      "language_loss": 0.73680878,
      "learning_rate": 3.6389244040916104e-06,
      "loss": 0.75931859,
      "num_input_tokens_seen": 38665695,
      "step": 1817,
      "time_per_iteration": 2.639530658721924
    },
    {
      "auxiliary_loss_clip": 0.01211305,
      "auxiliary_loss_mlp": 0.01210086,
      "balance_loss_clip": 1.05514264,
      "balance_loss_mlp": 1.0000205,
      "epoch": 0.21860157518186737,
      "flos": 26574650259840.0,
      "grad_norm": 2.007304040363273,
      "language_loss": 0.79021591,
      "learning_rate": 3.6384778252547747e-06,
      "loss": 0.81442982,
      "num_input_tokens_seen": 38681575,
      "step": 1818,
      "time_per_iteration": 2.7013185024261475
    },
    {
      "auxiliary_loss_clip": 0.01216956,
      "auxiliary_loss_mlp": 0.01209096,
      "balance_loss_clip": 1.05981016,
      "balance_loss_mlp": 1.00007427,
      "epoch": 0.21872181807250646,
      "flos": 20886292834560.0,
      "grad_norm": 3.2044616294573234,
      "language_loss": 0.78591132,
      "learning_rate": 3.638030997862155e-06,
      "loss": 0.81017184,
      "num_input_tokens_seen": 38700510,
      "step": 1819,
      "time_per_iteration": 2.622744083404541
    },
    {
      "auxiliary_loss_clip": 0.01127732,
      "auxiliary_loss_mlp": 0.01002874,
      "balance_loss_clip": 1.03034282,
      "balance_loss_mlp": 1.0006566,
      "epoch": 0.21884206096314554,
      "flos": 61209452897280.0,
      "grad_norm": 0.7741263618578745,
      "language_loss": 0.59390259,
      "learning_rate": 3.6375839219815356e-06,
      "loss": 0.61520863,
      "num_input_tokens_seen": 38758310,
      "step": 1820,
      "time_per_iteration": 3.0998456478118896
    },
    {
      "auxiliary_loss_clip": 0.01217535,
      "auxiliary_loss_mlp": 0.01034257,
      "balance_loss_clip": 1.06148076,
      "balance_loss_mlp": 1.02385557,
      "epoch": 0.21896230385378465,
      "flos": 23473835850240.0,
      "grad_norm": 1.956437004828811,
      "language_loss": 0.83043027,
      "learning_rate": 3.6371365976807375e-06,
      "loss": 0.85294819,
      "num_input_tokens_seen": 38778705,
      "step": 1821,
      "time_per_iteration": 2.733929395675659
    },
    {
      "auxiliary_loss_clip": 0.01216362,
      "auxiliary_loss_mlp": 0.01029561,
      "balance_loss_clip": 1.05483675,
      "balance_loss_mlp": 1.02007222,
      "epoch": 0.21908254674442373,
      "flos": 25081915829760.0,
      "grad_norm": 1.8664472220754555,
      "language_loss": 0.83620918,
      "learning_rate": 3.6366890250276185e-06,
      "loss": 0.85866845,
      "num_input_tokens_seen": 38799660,
      "step": 1822,
      "time_per_iteration": 2.7159833908081055
    },
    {
      "auxiliary_loss_clip": 0.01217016,
      "auxiliary_loss_mlp": 0.01032384,
      "balance_loss_clip": 1.06094611,
      "balance_loss_mlp": 1.02247214,
      "epoch": 0.21920278963506282,
      "flos": 23513768795520.0,
      "grad_norm": 1.9109935571009977,
      "language_loss": 0.90239561,
      "learning_rate": 3.6362412040900764e-06,
      "loss": 0.92488962,
      "num_input_tokens_seen": 38819450,
      "step": 1823,
      "time_per_iteration": 2.651852607727051
    },
    {
      "auxiliary_loss_clip": 0.01220844,
      "auxiliary_loss_mlp": 0.01033483,
      "balance_loss_clip": 1.05816317,
      "balance_loss_mlp": 1.02372622,
      "epoch": 0.21932303252570193,
      "flos": 29242238734080.0,
      "grad_norm": 2.043252926827376,
      "language_loss": 0.80566204,
      "learning_rate": 3.635793134936044e-06,
      "loss": 0.82820529,
      "num_input_tokens_seen": 38840460,
      "step": 1824,
      "time_per_iteration": 2.688109874725342
    },
    {
      "auxiliary_loss_clip": 0.01216037,
      "auxiliary_loss_mlp": 0.01033677,
      "balance_loss_clip": 1.05887568,
      "balance_loss_mlp": 1.02390182,
      "epoch": 0.219443275416341,
      "flos": 20806857907200.0,
      "grad_norm": 1.8951287360611948,
      "language_loss": 0.735147,
      "learning_rate": 3.635344817633494e-06,
      "loss": 0.75764418,
      "num_input_tokens_seen": 38859775,
      "step": 1825,
      "time_per_iteration": 2.662970542907715
    },
    {
      "auxiliary_loss_clip": 0.01212674,
      "auxiliary_loss_mlp": 0.0103363,
      "balance_loss_clip": 1.05712438,
      "balance_loss_mlp": 1.02409887,
      "epoch": 0.2195635183069801,
      "flos": 14501555458560.0,
      "grad_norm": 2.5930086756066273,
      "language_loss": 0.76107657,
      "learning_rate": 3.634896252250436e-06,
      "loss": 0.78353959,
      "num_input_tokens_seen": 38876540,
      "step": 1826,
      "time_per_iteration": 2.553138494491577
    },
    {
      "auxiliary_loss_clip": 0.01218853,
      "auxiliary_loss_mlp": 0.01039406,
      "balance_loss_clip": 1.06105518,
      "balance_loss_mlp": 1.03017378,
      "epoch": 0.2196837611976192,
      "flos": 24243473589120.0,
      "grad_norm": 2.0570476568000204,
      "language_loss": 0.8230288,
      "learning_rate": 3.6344474388549157e-06,
      "loss": 0.84561133,
      "num_input_tokens_seen": 38896195,
      "step": 1827,
      "time_per_iteration": 2.695380210876465
    },
    {
      "auxiliary_loss_clip": 0.0122294,
      "auxiliary_loss_mlp": 0.01039432,
      "balance_loss_clip": 1.06273699,
      "balance_loss_mlp": 1.02893591,
      "epoch": 0.2198040040882583,
      "flos": 18074523168000.0,
      "grad_norm": 2.588890334035827,
      "language_loss": 0.80157137,
      "learning_rate": 3.6339983775150183e-06,
      "loss": 0.82419503,
      "num_input_tokens_seen": 38912755,
      "step": 1828,
      "time_per_iteration": 2.562777280807495
    },
    {
      "auxiliary_loss_clip": 0.01216314,
      "auxiliary_loss_mlp": 0.01029678,
      "balance_loss_clip": 1.05983353,
      "balance_loss_mlp": 1.01968241,
      "epoch": 0.21992424697889737,
      "flos": 17784185535360.0,
      "grad_norm": 2.529288591147785,
      "language_loss": 0.84099114,
      "learning_rate": 3.6335490682988664e-06,
      "loss": 0.86345112,
      "num_input_tokens_seen": 38928365,
      "step": 1829,
      "time_per_iteration": 2.650697708129883
    },
    {
      "auxiliary_loss_clip": 0.01216409,
      "auxiliary_loss_mlp": 0.01027547,
      "balance_loss_clip": 1.04931521,
      "balance_loss_mlp": 1.0183028,
      "epoch": 0.22004448986953645,
      "flos": 17638495971840.0,
      "grad_norm": 1.9243252228605034,
      "language_loss": 0.83171147,
      "learning_rate": 3.63309951127462e-06,
      "loss": 0.85415107,
      "num_input_tokens_seen": 38945275,
      "step": 1830,
      "time_per_iteration": 2.658545970916748
    },
    {
      "auxiliary_loss_clip": 0.01221516,
      "auxiliary_loss_mlp": 0.0103572,
      "balance_loss_clip": 1.0568645,
      "balance_loss_mlp": 1.02540243,
      "epoch": 0.22016473276017556,
      "flos": 22275533203200.0,
      "grad_norm": 2.0287778908455136,
      "language_loss": 0.7523526,
      "learning_rate": 3.6326497065104757e-06,
      "loss": 0.77492499,
      "num_input_tokens_seen": 38965740,
      "step": 1831,
      "time_per_iteration": 3.673292875289917
    },
    {
      "auxiliary_loss_clip": 0.01223961,
      "auxiliary_loss_mlp": 0.01032226,
      "balance_loss_clip": 1.06006849,
      "balance_loss_mlp": 1.02295113,
      "epoch": 0.22028497565081465,
      "flos": 25556259859200.0,
      "grad_norm": 2.1359615279413724,
      "language_loss": 0.77825105,
      "learning_rate": 3.6321996540746697e-06,
      "loss": 0.80081284,
      "num_input_tokens_seen": 38984815,
      "step": 1832,
      "time_per_iteration": 3.6236958503723145
    },
    {
      "auxiliary_loss_clip": 0.01220468,
      "auxiliary_loss_mlp": 0.01030967,
      "balance_loss_clip": 1.05522418,
      "balance_loss_mlp": 1.02099514,
      "epoch": 0.22040521854145373,
      "flos": 36247332925440.0,
      "grad_norm": 1.8519707945402175,
      "language_loss": 0.80679476,
      "learning_rate": 3.6317493540354733e-06,
      "loss": 0.82930911,
      "num_input_tokens_seen": 39008230,
      "step": 1833,
      "time_per_iteration": 2.7861385345458984
    },
    {
      "auxiliary_loss_clip": 0.01214313,
      "auxiliary_loss_mlp": 0.01041831,
      "balance_loss_clip": 1.0566467,
      "balance_loss_mlp": 1.03221726,
      "epoch": 0.22052546143209284,
      "flos": 11838420270720.0,
      "grad_norm": 1.9728950837556138,
      "language_loss": 0.76788604,
      "learning_rate": 3.6312988064611976e-06,
      "loss": 0.79044753,
      "num_input_tokens_seen": 39026540,
      "step": 1834,
      "time_per_iteration": 2.6826744079589844
    },
    {
      "auxiliary_loss_clip": 0.0122526,
      "auxiliary_loss_mlp": 0.01033338,
      "balance_loss_clip": 1.05374789,
      "balance_loss_mlp": 1.02377129,
      "epoch": 0.22064570432273192,
      "flos": 24209250906240.0,
      "grad_norm": 1.8325483718446187,
      "language_loss": 0.81637704,
      "learning_rate": 3.6308480114201896e-06,
      "loss": 0.83896309,
      "num_input_tokens_seen": 39048460,
      "step": 1835,
      "time_per_iteration": 2.655083179473877
    },
    {
      "auxiliary_loss_clip": 0.01221217,
      "auxiliary_loss_mlp": 0.01035409,
      "balance_loss_clip": 1.06461763,
      "balance_loss_mlp": 1.02544296,
      "epoch": 0.220765947213371,
      "flos": 17931347556480.0,
      "grad_norm": 2.0068574833919137,
      "language_loss": 0.76552343,
      "learning_rate": 3.630396968980835e-06,
      "loss": 0.78808969,
      "num_input_tokens_seen": 39066335,
      "step": 1836,
      "time_per_iteration": 2.636457920074463
    },
    {
      "auxiliary_loss_clip": 0.01221123,
      "auxiliary_loss_mlp": 0.01036863,
      "balance_loss_clip": 1.05576885,
      "balance_loss_mlp": 1.02676058,
      "epoch": 0.2208861901040101,
      "flos": 26757040544640.0,
      "grad_norm": 2.5533088489310827,
      "language_loss": 0.83470827,
      "learning_rate": 3.6299456792115575e-06,
      "loss": 0.85728812,
      "num_input_tokens_seen": 39087590,
      "step": 1837,
      "time_per_iteration": 3.6463561058044434
    },
    {
      "auxiliary_loss_clip": 0.01211083,
      "auxiliary_loss_mlp": 0.01029332,
      "balance_loss_clip": 1.04395461,
      "balance_loss_mlp": 1.01942003,
      "epoch": 0.2210064329946492,
      "flos": 17817977255040.0,
      "grad_norm": 1.9464351804712194,
      "language_loss": 0.81265938,
      "learning_rate": 3.629494142180815e-06,
      "loss": 0.83506346,
      "num_input_tokens_seen": 39106335,
      "step": 1838,
      "time_per_iteration": 3.607715606689453
    },
    {
      "auxiliary_loss_clip": 0.01216797,
      "auxiliary_loss_mlp": 0.01032748,
      "balance_loss_clip": 1.06109786,
      "balance_loss_mlp": 1.02236509,
      "epoch": 0.22112667588528828,
      "flos": 17967401832960.0,
      "grad_norm": 2.230870629503133,
      "language_loss": 0.85195339,
      "learning_rate": 3.6290423579571075e-06,
      "loss": 0.87444878,
      "num_input_tokens_seen": 39122875,
      "step": 1839,
      "time_per_iteration": 2.6501662731170654
    },
    {
      "auxiliary_loss_clip": 0.01215541,
      "auxiliary_loss_mlp": 0.01032754,
      "balance_loss_clip": 1.05899167,
      "balance_loss_mlp": 1.02243638,
      "epoch": 0.22124691877592736,
      "flos": 18369206346240.0,
      "grad_norm": 1.8911746157974354,
      "language_loss": 0.80484885,
      "learning_rate": 3.6285903266089694e-06,
      "loss": 0.82733178,
      "num_input_tokens_seen": 39142150,
      "step": 1840,
      "time_per_iteration": 2.606210947036743
    },
    {
      "auxiliary_loss_clip": 0.01222915,
      "auxiliary_loss_mlp": 0.01028571,
      "balance_loss_clip": 1.05806613,
      "balance_loss_mlp": 1.01871204,
      "epoch": 0.22136716166656648,
      "flos": 20813286441600.0,
      "grad_norm": 2.485165042037828,
      "language_loss": 0.77310348,
      "learning_rate": 3.628138048204974e-06,
      "loss": 0.7956183,
      "num_input_tokens_seen": 39162835,
      "step": 1841,
      "time_per_iteration": 2.6699488162994385
    },
    {
      "auxiliary_loss_clip": 0.01213919,
      "auxiliary_loss_mlp": 0.01036936,
      "balance_loss_clip": 1.05146217,
      "balance_loss_mlp": 1.02580822,
      "epoch": 0.22148740455720556,
      "flos": 17675699483520.0,
      "grad_norm": 2.0863867652775996,
      "language_loss": 0.7654025,
      "learning_rate": 3.6276855228137304e-06,
      "loss": 0.78791106,
      "num_input_tokens_seen": 39181040,
      "step": 1842,
      "time_per_iteration": 2.62434458732605
    },
    {
      "auxiliary_loss_clip": 0.01217137,
      "auxiliary_loss_mlp": 0.01209147,
      "balance_loss_clip": 1.06072259,
      "balance_loss_mlp": 1.00012422,
      "epoch": 0.22160764744784464,
      "flos": 21726710323200.0,
      "grad_norm": 2.3624978677516846,
      "language_loss": 0.82074606,
      "learning_rate": 3.6272327505038874e-06,
      "loss": 0.84500885,
      "num_input_tokens_seen": 39197505,
      "step": 1843,
      "time_per_iteration": 2.618577480316162
    },
    {
      "auxiliary_loss_clip": 0.01226519,
      "auxiliary_loss_mlp": 0.01032235,
      "balance_loss_clip": 1.05232382,
      "balance_loss_mlp": 1.02340114,
      "epoch": 0.22172789033848372,
      "flos": 23764712186880.0,
      "grad_norm": 1.894231288858675,
      "language_loss": 0.78696012,
      "learning_rate": 3.626779731344131e-06,
      "loss": 0.80954766,
      "num_input_tokens_seen": 39217295,
      "step": 1844,
      "time_per_iteration": 2.7636678218841553
    },
    {
      "auxiliary_loss_clip": 0.01210776,
      "auxiliary_loss_mlp": 0.01032654,
      "balance_loss_clip": 1.05812812,
      "balance_loss_mlp": 1.02340329,
      "epoch": 0.22184813322912283,
      "flos": 16982300361600.0,
      "grad_norm": 2.946907513142115,
      "language_loss": 0.851403,
      "learning_rate": 3.6263264654031814e-06,
      "loss": 0.87383723,
      "num_input_tokens_seen": 39234195,
      "step": 1845,
      "time_per_iteration": 2.6327600479125977
    },
    {
      "auxiliary_loss_clip": 0.0113134,
      "auxiliary_loss_mlp": 0.01005483,
      "balance_loss_clip": 1.02731562,
      "balance_loss_mlp": 1.00311089,
      "epoch": 0.22196837611976192,
      "flos": 61823740314240.0,
      "grad_norm": 0.7003793230507843,
      "language_loss": 0.59170341,
      "learning_rate": 3.6258729527498008e-06,
      "loss": 0.61307162,
      "num_input_tokens_seen": 39295040,
      "step": 1846,
      "time_per_iteration": 3.1899774074554443
    },
    {
      "auxiliary_loss_clip": 0.01224327,
      "auxiliary_loss_mlp": 0.0103201,
      "balance_loss_clip": 1.0596323,
      "balance_loss_mlp": 1.02293789,
      "epoch": 0.222088619010401,
      "flos": 25558019625600.0,
      "grad_norm": 2.470914228635273,
      "language_loss": 0.6493457,
      "learning_rate": 3.6254191934527854e-06,
      "loss": 0.67190903,
      "num_input_tokens_seen": 39314395,
      "step": 1847,
      "time_per_iteration": 2.713242530822754
    },
    {
      "auxiliary_loss_clip": 0.01216355,
      "auxiliary_loss_mlp": 0.01030225,
      "balance_loss_clip": 1.05626154,
      "balance_loss_mlp": 1.01976454,
      "epoch": 0.2222088619010401,
      "flos": 19318612677120.0,
      "grad_norm": 2.0523800981222555,
      "language_loss": 0.65169853,
      "learning_rate": 3.6249651875809715e-06,
      "loss": 0.6741643,
      "num_input_tokens_seen": 39334275,
      "step": 1848,
      "time_per_iteration": 2.6988232135772705
    },
    {
      "auxiliary_loss_clip": 0.01213621,
      "auxiliary_loss_mlp": 0.01028059,
      "balance_loss_clip": 1.05864501,
      "balance_loss_mlp": 1.01869512,
      "epoch": 0.2223291047916792,
      "flos": 19099342103040.0,
      "grad_norm": 1.8423802346049238,
      "language_loss": 0.89092565,
      "learning_rate": 3.62451093520323e-06,
      "loss": 0.91334242,
      "num_input_tokens_seen": 39352180,
      "step": 1849,
      "time_per_iteration": 2.6512904167175293
    },
    {
      "auxiliary_loss_clip": 0.01217377,
      "auxiliary_loss_mlp": 0.0103797,
      "balance_loss_clip": 1.0497793,
      "balance_loss_mlp": 1.02879739,
      "epoch": 0.22244934768231828,
      "flos": 20850418126080.0,
      "grad_norm": 2.108781973244178,
      "language_loss": 0.90580487,
      "learning_rate": 3.6240564363884714e-06,
      "loss": 0.92835832,
      "num_input_tokens_seen": 39372125,
      "step": 1850,
      "time_per_iteration": 2.7792327404022217
    },
    {
      "auxiliary_loss_clip": 0.01219011,
      "auxiliary_loss_mlp": 0.01032553,
      "balance_loss_clip": 1.05612171,
      "balance_loss_mlp": 1.02291512,
      "epoch": 0.2225695905729574,
      "flos": 15632921111040.0,
      "grad_norm": 3.4536735520388606,
      "language_loss": 0.7038213,
      "learning_rate": 3.623601691205643e-06,
      "loss": 0.7263369,
      "num_input_tokens_seen": 39391200,
      "step": 1851,
      "time_per_iteration": 2.625060558319092
    },
    {
      "auxiliary_loss_clip": 0.01215247,
      "auxiliary_loss_mlp": 0.01027303,
      "balance_loss_clip": 1.05651617,
      "balance_loss_mlp": 1.01820159,
      "epoch": 0.22268983346359647,
      "flos": 25373582265600.0,
      "grad_norm": 2.089770587419895,
      "language_loss": 0.81508952,
      "learning_rate": 3.623146699723729e-06,
      "loss": 0.83751506,
      "num_input_tokens_seen": 39410660,
      "step": 1852,
      "time_per_iteration": 2.692967176437378
    },
    {
      "auxiliary_loss_clip": 0.01220202,
      "auxiliary_loss_mlp": 0.01035388,
      "balance_loss_clip": 1.06158471,
      "balance_loss_mlp": 1.02556467,
      "epoch": 0.22281007635423555,
      "flos": 13261452359040.0,
      "grad_norm": 4.081229086144897,
      "language_loss": 0.77582121,
      "learning_rate": 3.6226914620117507e-06,
      "loss": 0.79837716,
      "num_input_tokens_seen": 39429280,
      "step": 1853,
      "time_per_iteration": 2.692613124847412
    },
    {
      "auxiliary_loss_clip": 0.0122021,
      "auxiliary_loss_mlp": 0.01032206,
      "balance_loss_clip": 1.05019367,
      "balance_loss_mlp": 1.02322364,
      "epoch": 0.22293031924487464,
      "flos": 15340536403200.0,
      "grad_norm": 2.1251952290798415,
      "language_loss": 0.81077379,
      "learning_rate": 3.622235978138768e-06,
      "loss": 0.83329797,
      "num_input_tokens_seen": 39446905,
      "step": 1854,
      "time_per_iteration": 2.6627817153930664
    },
    {
      "auxiliary_loss_clip": 0.01218139,
      "auxiliary_loss_mlp": 0.01033729,
      "balance_loss_clip": 1.06110609,
      "balance_loss_mlp": 1.02406073,
      "epoch": 0.22305056213551375,
      "flos": 22564649773440.0,
      "grad_norm": 2.0067584535435783,
      "language_loss": 0.81605309,
      "learning_rate": 3.621780248173877e-06,
      "loss": 0.83857179,
      "num_input_tokens_seen": 39465105,
      "step": 1855,
      "time_per_iteration": 2.5872530937194824
    },
    {
      "auxiliary_loss_clip": 0.0112642,
      "auxiliary_loss_mlp": 0.0100748,
      "balance_loss_clip": 1.02769351,
      "balance_loss_mlp": 1.00535822,
      "epoch": 0.22317080502615283,
      "flos": 64880419887360.0,
      "grad_norm": 0.8253559533053295,
      "language_loss": 0.61064762,
      "learning_rate": 3.6213242721862125e-06,
      "loss": 0.63198662,
      "num_input_tokens_seen": 39523560,
      "step": 1856,
      "time_per_iteration": 3.1748814582824707
    },
    {
      "auxiliary_loss_clip": 0.01209259,
      "auxiliary_loss_mlp": 0.01032867,
      "balance_loss_clip": 1.05635548,
      "balance_loss_mlp": 1.02359796,
      "epoch": 0.2232910479167919,
      "flos": 25775997310080.0,
      "grad_norm": 1.6142234971446991,
      "language_loss": 0.75389671,
      "learning_rate": 3.620868050244945e-06,
      "loss": 0.77631789,
      "num_input_tokens_seen": 39544040,
      "step": 1857,
      "time_per_iteration": 2.742816925048828
    },
    {
      "auxiliary_loss_clip": 0.01213882,
      "auxiliary_loss_mlp": 0.01030801,
      "balance_loss_clip": 1.05633163,
      "balance_loss_mlp": 1.02103782,
      "epoch": 0.22341129080743102,
      "flos": 23251799928960.0,
      "grad_norm": 2.3729431335919307,
      "language_loss": 0.77619958,
      "learning_rate": 3.6204115824192817e-06,
      "loss": 0.79864645,
      "num_input_tokens_seen": 39561515,
      "step": 1858,
      "time_per_iteration": 3.536945343017578
    },
    {
      "auxiliary_loss_clip": 0.01210131,
      "auxiliary_loss_mlp": 0.01031243,
      "balance_loss_clip": 1.05378151,
      "balance_loss_mlp": 1.02124107,
      "epoch": 0.2235315336980701,
      "flos": 21214552250880.0,
      "grad_norm": 2.8033201441293745,
      "language_loss": 0.76765037,
      "learning_rate": 3.619954868778471e-06,
      "loss": 0.7900641,
      "num_input_tokens_seen": 39578210,
      "step": 1859,
      "time_per_iteration": 3.6051881313323975
    },
    {
      "auxiliary_loss_clip": 0.01216585,
      "auxiliary_loss_mlp": 0.0103267,
      "balance_loss_clip": 1.0540725,
      "balance_loss_mlp": 1.02418256,
      "epoch": 0.2236517765887092,
      "flos": 19901945548800.0,
      "grad_norm": 1.994790471702896,
      "language_loss": 0.82597929,
      "learning_rate": 3.6194979093917944e-06,
      "loss": 0.84847188,
      "num_input_tokens_seen": 39597625,
      "step": 1860,
      "time_per_iteration": 2.6403138637542725
    },
    {
      "auxiliary_loss_clip": 0.01212034,
      "auxiliary_loss_mlp": 0.01037384,
      "balance_loss_clip": 1.05411613,
      "balance_loss_mlp": 1.02820492,
      "epoch": 0.22377201947934827,
      "flos": 23214847812480.0,
      "grad_norm": 2.0264452294672557,
      "language_loss": 0.87169743,
      "learning_rate": 3.6190407043285724e-06,
      "loss": 0.89419162,
      "num_input_tokens_seen": 39615360,
      "step": 1861,
      "time_per_iteration": 2.6616463661193848
    },
    {
      "auxiliary_loss_clip": 0.01217499,
      "auxiliary_loss_mlp": 0.01036876,
      "balance_loss_clip": 1.0605514,
      "balance_loss_mlp": 1.02724385,
      "epoch": 0.22389226236998738,
      "flos": 26794244056320.0,
      "grad_norm": 1.828281361139331,
      "language_loss": 0.75838667,
      "learning_rate": 3.618583253658163e-06,
      "loss": 0.7809304,
      "num_input_tokens_seen": 39635460,
      "step": 1862,
      "time_per_iteration": 2.6732823848724365
    },
    {
      "auxiliary_loss_clip": 0.01219891,
      "auxiliary_loss_mlp": 0.01208868,
      "balance_loss_clip": 1.05232382,
      "balance_loss_mlp": 1.00007844,
      "epoch": 0.22401250526062647,
      "flos": 24170359455360.0,
      "grad_norm": 2.194250937179635,
      "language_loss": 0.86521208,
      "learning_rate": 3.618125557449961e-06,
      "loss": 0.88949966,
      "num_input_tokens_seen": 39653515,
      "step": 1863,
      "time_per_iteration": 3.645416498184204
    },
    {
      "auxiliary_loss_clip": 0.01212572,
      "auxiliary_loss_mlp": 0.01028533,
      "balance_loss_clip": 1.05805337,
      "balance_loss_mlp": 1.01894879,
      "epoch": 0.22413274815126555,
      "flos": 16759761649920.0,
      "grad_norm": 2.1005107390900233,
      "language_loss": 0.83436537,
      "learning_rate": 3.6176676157733983e-06,
      "loss": 0.85677642,
      "num_input_tokens_seen": 39668525,
      "step": 1864,
      "time_per_iteration": 3.510066270828247
    },
    {
      "auxiliary_loss_clip": 0.01213201,
      "auxiliary_loss_mlp": 0.01032939,
      "balance_loss_clip": 1.05318832,
      "balance_loss_mlp": 1.02355754,
      "epoch": 0.22425299104190466,
      "flos": 21360205900800.0,
      "grad_norm": 2.0695820537078857,
      "language_loss": 0.76520181,
      "learning_rate": 3.6172094286979443e-06,
      "loss": 0.78766322,
      "num_input_tokens_seen": 39685895,
      "step": 1865,
      "time_per_iteration": 2.679082155227661
    },
    {
      "auxiliary_loss_clip": 0.01216592,
      "auxiliary_loss_mlp": 0.01032045,
      "balance_loss_clip": 1.05393636,
      "balance_loss_mlp": 1.0224843,
      "epoch": 0.22437323393254374,
      "flos": 32165547108480.0,
      "grad_norm": 1.4169690689711072,
      "language_loss": 0.81300473,
      "learning_rate": 3.6167509962931064e-06,
      "loss": 0.83549112,
      "num_input_tokens_seen": 39711595,
      "step": 1866,
      "time_per_iteration": 2.7892563343048096
    },
    {
      "auxiliary_loss_clip": 0.01228682,
      "auxiliary_loss_mlp": 0.01028393,
      "balance_loss_clip": 1.05656838,
      "balance_loss_mlp": 1.0187254,
      "epoch": 0.22449347682318282,
      "flos": 18002809664640.0,
      "grad_norm": 2.594024444627303,
      "language_loss": 0.77075207,
      "learning_rate": 3.6162923186284276e-06,
      "loss": 0.79332286,
      "num_input_tokens_seen": 39727555,
      "step": 1867,
      "time_per_iteration": 2.773339033126831
    },
    {
      "auxiliary_loss_clip": 0.01216329,
      "auxiliary_loss_mlp": 0.01032356,
      "balance_loss_clip": 1.05441177,
      "balance_loss_mlp": 1.02299225,
      "epoch": 0.2246137197138219,
      "flos": 18697286194560.0,
      "grad_norm": 2.4488091936904786,
      "language_loss": 0.86029094,
      "learning_rate": 3.6158333957734888e-06,
      "loss": 0.88277781,
      "num_input_tokens_seen": 39746145,
      "step": 1868,
      "time_per_iteration": 2.576589584350586
    },
    {
      "auxiliary_loss_clip": 0.01221848,
      "auxiliary_loss_mlp": 0.0103115,
      "balance_loss_clip": 1.05341601,
      "balance_loss_mlp": 1.02181017,
      "epoch": 0.22473396260446102,
      "flos": 15590653781760.0,
      "grad_norm": 2.0679611603695958,
      "language_loss": 0.82881302,
      "learning_rate": 3.6153742277979088e-06,
      "loss": 0.85134304,
      "num_input_tokens_seen": 39763575,
      "step": 1869,
      "time_per_iteration": 2.871107816696167
    },
    {
      "auxiliary_loss_clip": 0.01220922,
      "auxiliary_loss_mlp": 0.01035167,
      "balance_loss_clip": 1.05683112,
      "balance_loss_mlp": 1.02599931,
      "epoch": 0.2248542054951001,
      "flos": 14465501182080.0,
      "grad_norm": 1.9221149640202455,
      "language_loss": 0.78209025,
      "learning_rate": 3.6149148147713434e-06,
      "loss": 0.80465114,
      "num_input_tokens_seen": 39781810,
      "step": 1870,
      "time_per_iteration": 2.6299889087677
    },
    {
      "auxiliary_loss_clip": 0.01224422,
      "auxiliary_loss_mlp": 0.01033481,
      "balance_loss_clip": 1.06443596,
      "balance_loss_mlp": 1.024719,
      "epoch": 0.22497444838573918,
      "flos": 19243882431360.0,
      "grad_norm": 2.316110392149575,
      "language_loss": 0.87116539,
      "learning_rate": 3.614455156763484e-06,
      "loss": 0.89374447,
      "num_input_tokens_seen": 39800115,
      "step": 1871,
      "time_per_iteration": 2.6453540325164795
    },
    {
      "auxiliary_loss_clip": 0.01215695,
      "auxiliary_loss_mlp": 0.01031945,
      "balance_loss_clip": 1.04843116,
      "balance_loss_mlp": 1.02324879,
      "epoch": 0.2250946912763783,
      "flos": 16910299549440.0,
      "grad_norm": 1.9631948801211274,
      "language_loss": 0.71164924,
      "learning_rate": 3.613995253844061e-06,
      "loss": 0.73412561,
      "num_input_tokens_seen": 39817795,
      "step": 1872,
      "time_per_iteration": 2.7040233612060547
    },
    {
      "auxiliary_loss_clip": 0.01214638,
      "auxiliary_loss_mlp": 0.01035175,
      "balance_loss_clip": 1.06001377,
      "balance_loss_mlp": 1.02582276,
      "epoch": 0.22521493416701738,
      "flos": 24681368292480.0,
      "grad_norm": 2.8541478995351746,
      "language_loss": 0.80948627,
      "learning_rate": 3.6135351060828414e-06,
      "loss": 0.8319844,
      "num_input_tokens_seen": 39838270,
      "step": 1873,
      "time_per_iteration": 2.5825347900390625
    },
    {
      "auxiliary_loss_clip": 0.01220524,
      "auxiliary_loss_mlp": 0.01038483,
      "balance_loss_clip": 1.06297922,
      "balance_loss_mlp": 1.0283798,
      "epoch": 0.22533517705765646,
      "flos": 17821963664640.0,
      "grad_norm": 3.797224735631961,
      "language_loss": 0.69673324,
      "learning_rate": 3.6130747135496285e-06,
      "loss": 0.71932328,
      "num_input_tokens_seen": 39857270,
      "step": 1874,
      "time_per_iteration": 2.6695475578308105
    },
    {
      "auxiliary_loss_clip": 0.01212297,
      "auxiliary_loss_mlp": 0.01030792,
      "balance_loss_clip": 1.05897582,
      "balance_loss_mlp": 1.02102852,
      "epoch": 0.22545541994829554,
      "flos": 33691390899840.0,
      "grad_norm": 1.910025162903034,
      "language_loss": 0.66178375,
      "learning_rate": 3.6126140763142646e-06,
      "loss": 0.68421459,
      "num_input_tokens_seen": 39882300,
      "step": 1875,
      "time_per_iteration": 2.722743511199951
    },
    {
      "auxiliary_loss_clip": 0.01215138,
      "auxiliary_loss_mlp": 0.01035607,
      "balance_loss_clip": 1.06138396,
      "balance_loss_mlp": 1.02580225,
      "epoch": 0.22557566283893465,
      "flos": 19171594310400.0,
      "grad_norm": 2.567027959338004,
      "language_loss": 0.86070526,
      "learning_rate": 3.6121531944466275e-06,
      "loss": 0.88321269,
      "num_input_tokens_seen": 39899625,
      "step": 1876,
      "time_per_iteration": 2.702401876449585
    },
    {
      "auxiliary_loss_clip": 0.01212386,
      "auxiliary_loss_mlp": 0.01034896,
      "balance_loss_clip": 1.05874634,
      "balance_loss_mlp": 1.02616429,
      "epoch": 0.22569590572957374,
      "flos": 20773281669120.0,
      "grad_norm": 2.69019263592836,
      "language_loss": 0.78123438,
      "learning_rate": 3.611692068016633e-06,
      "loss": 0.80370712,
      "num_input_tokens_seen": 39915955,
      "step": 1877,
      "time_per_iteration": 2.5996057987213135
    },
    {
      "auxiliary_loss_clip": 0.01213464,
      "auxiliary_loss_mlp": 0.01034509,
      "balance_loss_clip": 1.05072999,
      "balance_loss_mlp": 1.02377987,
      "epoch": 0.22581614862021282,
      "flos": 18442715529600.0,
      "grad_norm": 2.335596386192277,
      "language_loss": 0.74875998,
      "learning_rate": 3.611230697094233e-06,
      "loss": 0.77123976,
      "num_input_tokens_seen": 39932655,
      "step": 1878,
      "time_per_iteration": 2.7121853828430176
    },
    {
      "auxiliary_loss_clip": 0.01221084,
      "auxiliary_loss_mlp": 0.01031913,
      "balance_loss_clip": 1.05621588,
      "balance_loss_mlp": 1.02312112,
      "epoch": 0.22593639151085193,
      "flos": 20048389297920.0,
      "grad_norm": 1.795196420352208,
      "language_loss": 0.87289447,
      "learning_rate": 3.6107690817494173e-06,
      "loss": 0.89542443,
      "num_input_tokens_seen": 39952875,
      "step": 1879,
      "time_per_iteration": 2.6252315044403076
    },
    {
      "auxiliary_loss_clip": 0.0121686,
      "auxiliary_loss_mlp": 0.01027607,
      "balance_loss_clip": 1.05187714,
      "balance_loss_mlp": 1.01868415,
      "epoch": 0.226056634401491,
      "flos": 13115116350720.0,
      "grad_norm": 2.332887587454691,
      "language_loss": 0.71253163,
      "learning_rate": 3.6103072220522117e-06,
      "loss": 0.73497629,
      "num_input_tokens_seen": 39968405,
      "step": 1880,
      "time_per_iteration": 2.6868486404418945
    },
    {
      "auxiliary_loss_clip": 0.01222425,
      "auxiliary_loss_mlp": 0.01033708,
      "balance_loss_clip": 1.05412483,
      "balance_loss_mlp": 1.02457094,
      "epoch": 0.2261768772921301,
      "flos": 18988378012800.0,
      "grad_norm": 2.737825310417851,
      "language_loss": 0.91856778,
      "learning_rate": 3.609845118072682e-06,
      "loss": 0.94112909,
      "num_input_tokens_seen": 39987075,
      "step": 1881,
      "time_per_iteration": 2.7182607650756836
    },
    {
      "auxiliary_loss_clip": 0.01219073,
      "auxiliary_loss_mlp": 0.01208948,
      "balance_loss_clip": 1.05762589,
      "balance_loss_mlp": 1.00019431,
      "epoch": 0.2262971201827692,
      "flos": 19974054101760.0,
      "grad_norm": 1.9193155411368095,
      "language_loss": 0.79834378,
      "learning_rate": 3.6093827698809276e-06,
      "loss": 0.82262409,
      "num_input_tokens_seen": 40006175,
      "step": 1882,
      "time_per_iteration": 2.6438636779785156
    },
    {
      "auxiliary_loss_clip": 0.01213122,
      "auxiliary_loss_mlp": 0.01031619,
      "balance_loss_clip": 1.05583918,
      "balance_loss_mlp": 1.02248168,
      "epoch": 0.2264173630734083,
      "flos": 16654543735680.0,
      "grad_norm": 3.1211229954369415,
      "language_loss": 0.84951127,
      "learning_rate": 3.6089201775470864e-06,
      "loss": 0.87195867,
      "num_input_tokens_seen": 40021630,
      "step": 1883,
      "time_per_iteration": 2.6097843647003174
    },
    {
      "auxiliary_loss_clip": 0.01206997,
      "auxiliary_loss_mlp": 0.01030978,
      "balance_loss_clip": 1.05285478,
      "balance_loss_mlp": 1.02195406,
      "epoch": 0.22653760596404737,
      "flos": 24389809597440.0,
      "grad_norm": 1.364743278800436,
      "language_loss": 0.77732623,
      "learning_rate": 3.6084573411413334e-06,
      "loss": 0.79970598,
      "num_input_tokens_seen": 40041025,
      "step": 1884,
      "time_per_iteration": 3.6512978076934814
    },
    {
      "auxiliary_loss_clip": 0.01219441,
      "auxiliary_loss_mlp": 0.01034484,
      "balance_loss_clip": 1.05481243,
      "balance_loss_mlp": 1.02444625,
      "epoch": 0.22665784885468646,
      "flos": 18332541538560.0,
      "grad_norm": 5.312441414884545,
      "language_loss": 0.8104254,
      "learning_rate": 3.607994260733881e-06,
      "loss": 0.83296466,
      "num_input_tokens_seen": 40060265,
      "step": 1885,
      "time_per_iteration": 2.650224208831787
    },
    {
      "auxiliary_loss_clip": 0.01206067,
      "auxiliary_loss_mlp": 0.01032075,
      "balance_loss_clip": 1.05648971,
      "balance_loss_mlp": 1.02341461,
      "epoch": 0.22677809174532557,
      "flos": 24058102475520.0,
      "grad_norm": 2.1003892399286657,
      "language_loss": 0.74696231,
      "learning_rate": 3.6075309363949776e-06,
      "loss": 0.76934373,
      "num_input_tokens_seen": 40079435,
      "step": 1886,
      "time_per_iteration": 3.5573995113372803
    },
    {
      "auxiliary_loss_clip": 0.01214012,
      "auxiliary_loss_mlp": 0.01028022,
      "balance_loss_clip": 1.05960178,
      "balance_loss_mlp": 1.01834774,
      "epoch": 0.22689833463596465,
      "flos": 20374242503040.0,
      "grad_norm": 2.689932129400688,
      "language_loss": 0.81681657,
      "learning_rate": 3.6070673681949094e-06,
      "loss": 0.83923686,
      "num_input_tokens_seen": 40097800,
      "step": 1887,
      "time_per_iteration": 2.649779796600342
    },
    {
      "auxiliary_loss_clip": 0.01219315,
      "auxiliary_loss_mlp": 0.01208986,
      "balance_loss_clip": 1.05833268,
      "balance_loss_mlp": 1.0001241,
      "epoch": 0.22701857752660373,
      "flos": 30120398438400.0,
      "grad_norm": 1.7900613837259376,
      "language_loss": 0.81604052,
      "learning_rate": 3.606603556203999e-06,
      "loss": 0.84032345,
      "num_input_tokens_seen": 40122745,
      "step": 1888,
      "time_per_iteration": 2.7909984588623047
    },
    {
      "auxiliary_loss_clip": 0.0121539,
      "auxiliary_loss_mlp": 0.01034592,
      "balance_loss_clip": 1.05663705,
      "balance_loss_mlp": 1.02562761,
      "epoch": 0.22713882041724284,
      "flos": 22492182084480.0,
      "grad_norm": 1.799678474143235,
      "language_loss": 0.83723307,
      "learning_rate": 3.6061395004926066e-06,
      "loss": 0.85973293,
      "num_input_tokens_seen": 40141680,
      "step": 1889,
      "time_per_iteration": 2.700709104537964
    },
    {
      "auxiliary_loss_clip": 0.01212947,
      "auxiliary_loss_mlp": 0.0103166,
      "balance_loss_clip": 1.05813944,
      "balance_loss_mlp": 1.02262962,
      "epoch": 0.22725906330788193,
      "flos": 20521548178560.0,
      "grad_norm": 2.4809041681935025,
      "language_loss": 0.85303628,
      "learning_rate": 3.605675201131129e-06,
      "loss": 0.87548232,
      "num_input_tokens_seen": 40160140,
      "step": 1890,
      "time_per_iteration": 3.5213611125946045
    },
    {
      "auxiliary_loss_clip": 0.01222325,
      "auxiliary_loss_mlp": 0.0103141,
      "balance_loss_clip": 1.06246817,
      "balance_loss_mlp": 1.02283859,
      "epoch": 0.227379306198521,
      "flos": 18989922297600.0,
      "grad_norm": 2.5644177167581246,
      "language_loss": 0.80021006,
      "learning_rate": 3.60521065819e-06,
      "loss": 0.82274747,
      "num_input_tokens_seen": 40177450,
      "step": 1891,
      "time_per_iteration": 3.5281479358673096
    },
    {
      "auxiliary_loss_clip": 0.01220604,
      "auxiliary_loss_mlp": 0.010311,
      "balance_loss_clip": 1.0567286,
      "balance_loss_mlp": 1.02265406,
      "epoch": 0.2274995490891601,
      "flos": 21798351999360.0,
      "grad_norm": 1.8552381886151803,
      "language_loss": 0.88088393,
      "learning_rate": 3.60474587173969e-06,
      "loss": 0.90340102,
      "num_input_tokens_seen": 40195935,
      "step": 1892,
      "time_per_iteration": 2.7022087574005127
    },
    {
      "auxiliary_loss_clip": 0.01213903,
      "auxiliary_loss_mlp": 0.0103431,
      "balance_loss_clip": 1.06113851,
      "balance_loss_mlp": 1.02532768,
      "epoch": 0.2276197919797992,
      "flos": 19058654972160.0,
      "grad_norm": 2.2606623111841184,
      "language_loss": 0.84417117,
      "learning_rate": 3.6042808418507084e-06,
      "loss": 0.86665326,
      "num_input_tokens_seen": 40213620,
      "step": 1893,
      "time_per_iteration": 2.5979115962982178
    },
    {
      "auxiliary_loss_clip": 0.01215419,
      "auxiliary_loss_mlp": 0.01031217,
      "balance_loss_clip": 1.05914497,
      "balance_loss_mlp": 1.02197862,
      "epoch": 0.22774003487043828,
      "flos": 18806777827200.0,
      "grad_norm": 2.0389096817009604,
      "language_loss": 0.76811957,
      "learning_rate": 3.6038155685935976e-06,
      "loss": 0.79058588,
      "num_input_tokens_seen": 40230190,
      "step": 1894,
      "time_per_iteration": 2.620732307434082
    },
    {
      "auxiliary_loss_clip": 0.01211594,
      "auxiliary_loss_mlp": 0.01029984,
      "balance_loss_clip": 1.05708957,
      "balance_loss_mlp": 1.02142525,
      "epoch": 0.22786027776107737,
      "flos": 23002544476800.0,
      "grad_norm": 2.0457632979992657,
      "language_loss": 0.70492387,
      "learning_rate": 3.6033500520389404e-06,
      "loss": 0.72733974,
      "num_input_tokens_seen": 40246860,
      "step": 1895,
      "time_per_iteration": 2.6223254203796387
    },
    {
      "auxiliary_loss_clip": 0.01137765,
      "auxiliary_loss_mlp": 0.01004321,
      "balance_loss_clip": 1.02360082,
      "balance_loss_mlp": 1.00200868,
      "epoch": 0.22798052065171648,
      "flos": 66706872600960.0,
      "grad_norm": 0.81646903347896,
      "language_loss": 0.64828414,
      "learning_rate": 3.6028842922573553e-06,
      "loss": 0.66970497,
      "num_input_tokens_seen": 40311005,
      "step": 1896,
      "time_per_iteration": 3.303868055343628
    },
    {
      "auxiliary_loss_clip": 0.01135855,
      "auxiliary_loss_mlp": 0.01200244,
      "balance_loss_clip": 1.02611792,
      "balance_loss_mlp": 0.99983007,
      "epoch": 0.22810076354235556,
      "flos": 62080896758400.0,
      "grad_norm": 0.8731104894568679,
      "language_loss": 0.62931621,
      "learning_rate": 3.602418289319497e-06,
      "loss": 0.65267718,
      "num_input_tokens_seen": 40369560,
      "step": 1897,
      "time_per_iteration": 3.2259762287139893
    },
    {
      "auxiliary_loss_clip": 0.01217238,
      "auxiliary_loss_mlp": 0.01031609,
      "balance_loss_clip": 1.05208731,
      "balance_loss_mlp": 1.02219093,
      "epoch": 0.22822100643299464,
      "flos": 23876358635520.0,
      "grad_norm": 2.4897872387872777,
      "language_loss": 0.73396444,
      "learning_rate": 3.601952043296059e-06,
      "loss": 0.75645292,
      "num_input_tokens_seen": 40389555,
      "step": 1898,
      "time_per_iteration": 2.7066283226013184
    },
    {
      "auxiliary_loss_clip": 0.01221377,
      "auxiliary_loss_mlp": 0.01028862,
      "balance_loss_clip": 1.05420899,
      "balance_loss_mlp": 1.01984334,
      "epoch": 0.22834124932363373,
      "flos": 20991331180800.0,
      "grad_norm": 2.135791922314558,
      "language_loss": 0.80796242,
      "learning_rate": 3.6014855542577696e-06,
      "loss": 0.83046484,
      "num_input_tokens_seen": 40406765,
      "step": 1899,
      "time_per_iteration": 2.659780740737915
    },
    {
      "auxiliary_loss_clip": 0.01216899,
      "auxiliary_loss_mlp": 0.01027011,
      "balance_loss_clip": 1.05734348,
      "balance_loss_mlp": 1.01768243,
      "epoch": 0.22846149221427284,
      "flos": 24901572620160.0,
      "grad_norm": 1.9558935387067438,
      "language_loss": 0.84017736,
      "learning_rate": 3.6010188222753943e-06,
      "loss": 0.86261648,
      "num_input_tokens_seen": 40427535,
      "step": 1900,
      "time_per_iteration": 2.7299654483795166
    },
    {
      "auxiliary_loss_clip": 0.01123143,
      "auxiliary_loss_mlp": 0.0100218,
      "balance_loss_clip": 1.02556896,
      "balance_loss_mlp": 0.99970037,
      "epoch": 0.22858173510491192,
      "flos": 56132294319360.0,
      "grad_norm": 0.899843816102502,
      "language_loss": 0.64184552,
      "learning_rate": 3.6005518474197372e-06,
      "loss": 0.66309875,
      "num_input_tokens_seen": 40479580,
      "step": 1901,
      "time_per_iteration": 3.1144556999206543
    },
    {
      "auxiliary_loss_clip": 0.01216421,
      "auxiliary_loss_mlp": 0.01028107,
      "balance_loss_clip": 1.0610292,
      "balance_loss_mlp": 1.01852822,
      "epoch": 0.228701977995551,
      "flos": 24170826332160.0,
      "grad_norm": 1.9408774264174826,
      "language_loss": 0.78481615,
      "learning_rate": 3.6000846297616373e-06,
      "loss": 0.80726141,
      "num_input_tokens_seen": 40497880,
      "step": 1902,
      "time_per_iteration": 2.7256112098693848
    },
    {
      "auxiliary_loss_clip": 0.01218367,
      "auxiliary_loss_mlp": 0.01031606,
      "balance_loss_clip": 1.06307721,
      "balance_loss_mlp": 1.02177143,
      "epoch": 0.22882222088619011,
      "flos": 21387892308480.0,
      "grad_norm": 2.468422768704748,
      "language_loss": 0.73085284,
      "learning_rate": 3.5996171693719717e-06,
      "loss": 0.75335258,
      "num_input_tokens_seen": 40513975,
      "step": 1903,
      "time_per_iteration": 2.5940444469451904
    },
    {
      "auxiliary_loss_clip": 0.01123432,
      "auxiliary_loss_mlp": 0.01003873,
      "balance_loss_clip": 1.02674568,
      "balance_loss_mlp": 1.00154865,
      "epoch": 0.2289424637768292,
      "flos": 64589615377920.0,
      "grad_norm": 0.827171787681534,
      "language_loss": 0.64752692,
      "learning_rate": 3.5991494663216528e-06,
      "loss": 0.6688,
      "num_input_tokens_seen": 40576960,
      "step": 1904,
      "time_per_iteration": 3.225473642349243
    },
    {
      "auxiliary_loss_clip": 0.01212771,
      "auxiliary_loss_mlp": 0.01028957,
      "balance_loss_clip": 1.06085634,
      "balance_loss_mlp": 1.01975441,
      "epoch": 0.22906270666746828,
      "flos": 22163419877760.0,
      "grad_norm": 2.2117929295247447,
      "language_loss": 0.87846392,
      "learning_rate": 3.5986815206816314e-06,
      "loss": 0.90088117,
      "num_input_tokens_seen": 40595780,
      "step": 1905,
      "time_per_iteration": 2.5935299396514893
    },
    {
      "auxiliary_loss_clip": 0.0121512,
      "auxiliary_loss_mlp": 0.01030503,
      "balance_loss_clip": 1.06249034,
      "balance_loss_mlp": 1.02172339,
      "epoch": 0.2291829495581074,
      "flos": 25772334122880.0,
      "grad_norm": 2.087359240613912,
      "language_loss": 0.74864686,
      "learning_rate": 3.598213332522895e-06,
      "loss": 0.77110314,
      "num_input_tokens_seen": 40615810,
      "step": 1906,
      "time_per_iteration": 2.6517746448516846
    },
    {
      "auxiliary_loss_clip": 0.01213924,
      "auxiliary_loss_mlp": 0.01029722,
      "balance_loss_clip": 1.05799842,
      "balance_loss_mlp": 1.02032888,
      "epoch": 0.22930319244874647,
      "flos": 31172760126720.0,
      "grad_norm": 1.7665619418395957,
      "language_loss": 0.77506274,
      "learning_rate": 3.597744901916466e-06,
      "loss": 0.79749924,
      "num_input_tokens_seen": 40637095,
      "step": 1907,
      "time_per_iteration": 2.655111789703369
    },
    {
      "auxiliary_loss_clip": 0.01218466,
      "auxiliary_loss_mlp": 0.01029215,
      "balance_loss_clip": 1.06154013,
      "balance_loss_mlp": 1.01954746,
      "epoch": 0.22942343533938556,
      "flos": 23254098399360.0,
      "grad_norm": 2.8609087076435307,
      "language_loss": 0.77201056,
      "learning_rate": 3.5972762289334058e-06,
      "loss": 0.79448736,
      "num_input_tokens_seen": 40656725,
      "step": 1908,
      "time_per_iteration": 2.619905710220337
    },
    {
      "auxiliary_loss_clip": 0.01219378,
      "auxiliary_loss_mlp": 0.01029638,
      "balance_loss_clip": 1.05432534,
      "balance_loss_mlp": 1.01957703,
      "epoch": 0.22954367823002464,
      "flos": 14610903436800.0,
      "grad_norm": 2.973501946634718,
      "language_loss": 0.8567766,
      "learning_rate": 3.5968073136448116e-06,
      "loss": 0.87926674,
      "num_input_tokens_seen": 40674745,
      "step": 1909,
      "time_per_iteration": 2.6268370151519775
    },
    {
      "auxiliary_loss_clip": 0.01221245,
      "auxiliary_loss_mlp": 0.01032565,
      "balance_loss_clip": 1.05986595,
      "balance_loss_mlp": 1.02256298,
      "epoch": 0.22966392112066375,
      "flos": 16763604405120.0,
      "grad_norm": 1.8302953057041136,
      "language_loss": 0.91490436,
      "learning_rate": 3.596338156121818e-06,
      "loss": 0.93744242,
      "num_input_tokens_seen": 40693630,
      "step": 1910,
      "time_per_iteration": 2.660398483276367
    },
    {
      "auxiliary_loss_clip": 0.01122326,
      "auxiliary_loss_mlp": 0.01001759,
      "balance_loss_clip": 1.02345634,
      "balance_loss_mlp": 0.99949378,
      "epoch": 0.22978416401130283,
      "flos": 67474247783040.0,
      "grad_norm": 0.743260775772671,
      "language_loss": 0.59366429,
      "learning_rate": 3.595868756435595e-06,
      "loss": 0.61490512,
      "num_input_tokens_seen": 40761310,
      "step": 1911,
      "time_per_iteration": 4.260406017303467
    },
    {
      "auxiliary_loss_clip": 0.01222986,
      "auxiliary_loss_mlp": 0.01030542,
      "balance_loss_clip": 1.0606966,
      "balance_loss_mlp": 1.02098691,
      "epoch": 0.22990440690194192,
      "flos": 19865137086720.0,
      "grad_norm": 1.9736840673696427,
      "language_loss": 0.80846655,
      "learning_rate": 3.5953991146573504e-06,
      "loss": 0.83100182,
      "num_input_tokens_seen": 40779955,
      "step": 1912,
      "time_per_iteration": 2.6221730709075928
    },
    {
      "auxiliary_loss_clip": 0.0121802,
      "auxiliary_loss_mlp": 0.01029381,
      "balance_loss_clip": 1.05776846,
      "balance_loss_mlp": 1.01957011,
      "epoch": 0.23002464979258103,
      "flos": 13289246507520.0,
      "grad_norm": 3.394923793595634,
      "language_loss": 0.83406448,
      "learning_rate": 3.5949292308583294e-06,
      "loss": 0.85653853,
      "num_input_tokens_seen": 40793200,
      "step": 1913,
      "time_per_iteration": 3.534404754638672
    },
    {
      "auxiliary_loss_clip": 0.01217488,
      "auxiliary_loss_mlp": 0.01033757,
      "balance_loss_clip": 1.06498718,
      "balance_loss_mlp": 1.02314115,
      "epoch": 0.2301448926832201,
      "flos": 22163779013760.0,
      "grad_norm": 2.0468681473443358,
      "language_loss": 0.8096137,
      "learning_rate": 3.594459105109811e-06,
      "loss": 0.83212614,
      "num_input_tokens_seen": 40812380,
      "step": 1914,
      "time_per_iteration": 2.630662441253662
    },
    {
      "auxiliary_loss_clip": 0.01219639,
      "auxiliary_loss_mlp": 0.01029868,
      "balance_loss_clip": 1.06311011,
      "balance_loss_mlp": 1.02136266,
      "epoch": 0.2302651355738592,
      "flos": 20704477167360.0,
      "grad_norm": 5.232972573524458,
      "language_loss": 0.81582773,
      "learning_rate": 3.593988737483115e-06,
      "loss": 0.83832282,
      "num_input_tokens_seen": 40832320,
      "step": 1915,
      "time_per_iteration": 2.588772773742676
    },
    {
      "auxiliary_loss_clip": 0.01217431,
      "auxiliary_loss_mlp": 0.01026897,
      "balance_loss_clip": 1.05813217,
      "balance_loss_mlp": 1.01757467,
      "epoch": 0.23038537846449827,
      "flos": 18588943797120.0,
      "grad_norm": 1.9665442021357065,
      "language_loss": 0.78234583,
      "learning_rate": 3.5935181280495947e-06,
      "loss": 0.80478913,
      "num_input_tokens_seen": 40850900,
      "step": 1916,
      "time_per_iteration": 2.641099691390991
    },
    {
      "auxiliary_loss_clip": 0.01114235,
      "auxiliary_loss_mlp": 0.01003061,
      "balance_loss_clip": 1.02051401,
      "balance_loss_mlp": 1.00105846,
      "epoch": 0.23050562135513739,
      "flos": 64224260190720.0,
      "grad_norm": 0.7986819152977918,
      "language_loss": 0.54271692,
      "learning_rate": 3.5930472768806412e-06,
      "loss": 0.56388992,
      "num_input_tokens_seen": 40909570,
      "step": 1917,
      "time_per_iteration": 4.076315879821777
    },
    {
      "auxiliary_loss_clip": 0.01215149,
      "auxiliary_loss_mlp": 0.01030463,
      "balance_loss_clip": 1.06451535,
      "balance_loss_mlp": 1.02100396,
      "epoch": 0.23062586424577647,
      "flos": 17313396952320.0,
      "grad_norm": 1.8155229533351964,
      "language_loss": 0.7705332,
      "learning_rate": 3.5925761840476826e-06,
      "loss": 0.79298931,
      "num_input_tokens_seen": 40928180,
      "step": 1918,
      "time_per_iteration": 3.527216911315918
    },
    {
      "auxiliary_loss_clip": 0.01214632,
      "auxiliary_loss_mlp": 0.01029073,
      "balance_loss_clip": 1.05917561,
      "balance_loss_mlp": 1.02008474,
      "epoch": 0.23074610713641555,
      "flos": 27855979194240.0,
      "grad_norm": 3.1299647825420758,
      "language_loss": 0.8162185,
      "learning_rate": 3.592104849622183e-06,
      "loss": 0.83865547,
      "num_input_tokens_seen": 40950435,
      "step": 1919,
      "time_per_iteration": 2.7665202617645264
    },
    {
      "auxiliary_loss_clip": 0.01213044,
      "auxiliary_loss_mlp": 0.01029618,
      "balance_loss_clip": 1.05404842,
      "balance_loss_mlp": 1.02020681,
      "epoch": 0.23086635002705466,
      "flos": 28841798937600.0,
      "grad_norm": 1.5344139585339567,
      "language_loss": 0.73702264,
      "learning_rate": 3.591633273675644e-06,
      "loss": 0.75944924,
      "num_input_tokens_seen": 40972670,
      "step": 1920,
      "time_per_iteration": 2.793891191482544
    },
    {
      "auxiliary_loss_clip": 0.01124297,
      "auxiliary_loss_mlp": 0.01010203,
      "balance_loss_clip": 1.03368044,
      "balance_loss_mlp": 1.00820053,
      "epoch": 0.23098659291769374,
      "flos": 62923681566720.0,
      "grad_norm": 0.9089214705451581,
      "language_loss": 0.58166301,
      "learning_rate": 3.591161456279602e-06,
      "loss": 0.60300803,
      "num_input_tokens_seen": 41018215,
      "step": 1921,
      "time_per_iteration": 3.0299599170684814
    },
    {
      "auxiliary_loss_clip": 0.01223029,
      "auxiliary_loss_mlp": 0.01030331,
      "balance_loss_clip": 1.0585264,
      "balance_loss_mlp": 1.02105081,
      "epoch": 0.23110683580833283,
      "flos": 23476816679040.0,
      "grad_norm": 1.504359239929697,
      "language_loss": 0.80628514,
      "learning_rate": 3.590689397505633e-06,
      "loss": 0.8288188,
      "num_input_tokens_seen": 41039125,
      "step": 1922,
      "time_per_iteration": 2.635345935821533
    },
    {
      "auxiliary_loss_clip": 0.01212112,
      "auxiliary_loss_mlp": 0.01027885,
      "balance_loss_clip": 1.06169105,
      "balance_loss_mlp": 1.01874161,
      "epoch": 0.2312270786989719,
      "flos": 27271066124160.0,
      "grad_norm": 1.734212433164823,
      "language_loss": 0.86687982,
      "learning_rate": 3.590217097425347e-06,
      "loss": 0.88927984,
      "num_input_tokens_seen": 41059025,
      "step": 1923,
      "time_per_iteration": 2.629695177078247
    },
    {
      "auxiliary_loss_clip": 0.01217215,
      "auxiliary_loss_mlp": 0.01030723,
      "balance_loss_clip": 1.06371713,
      "balance_loss_mlp": 1.0210675,
      "epoch": 0.23134732158961102,
      "flos": 13261344618240.0,
      "grad_norm": 2.7601319873682963,
      "language_loss": 0.7116766,
      "learning_rate": 3.589744556110391e-06,
      "loss": 0.73415595,
      "num_input_tokens_seen": 41077015,
      "step": 1924,
      "time_per_iteration": 2.548661947250366
    },
    {
      "auxiliary_loss_clip": 0.01214756,
      "auxiliary_loss_mlp": 0.01029105,
      "balance_loss_clip": 1.05679524,
      "balance_loss_mlp": 1.02066517,
      "epoch": 0.2314675644802501,
      "flos": 36977648250240.0,
      "grad_norm": 1.86602556333926,
      "language_loss": 0.84230483,
      "learning_rate": 3.58927177363245e-06,
      "loss": 0.86474347,
      "num_input_tokens_seen": 41099840,
      "step": 1925,
      "time_per_iteration": 2.799006223678589
    },
    {
      "auxiliary_loss_clip": 0.01215759,
      "auxiliary_loss_mlp": 0.01028695,
      "balance_loss_clip": 1.05403721,
      "balance_loss_mlp": 1.01891375,
      "epoch": 0.2315878073708892,
      "flos": 23842207779840.0,
      "grad_norm": 2.263414405623644,
      "language_loss": 0.72684318,
      "learning_rate": 3.5887987500632447e-06,
      "loss": 0.74928772,
      "num_input_tokens_seen": 41117845,
      "step": 1926,
      "time_per_iteration": 2.746462345123291
    },
    {
      "auxiliary_loss_clip": 0.0122339,
      "auxiliary_loss_mlp": 0.01035233,
      "balance_loss_clip": 1.05698347,
      "balance_loss_mlp": 1.02656674,
      "epoch": 0.2317080502615283,
      "flos": 23039424766080.0,
      "grad_norm": 2.1333974823857957,
      "language_loss": 0.84212625,
      "learning_rate": 3.5883254854745325e-06,
      "loss": 0.86471248,
      "num_input_tokens_seen": 41136235,
      "step": 1927,
      "time_per_iteration": 2.6709866523742676
    },
    {
      "auxiliary_loss_clip": 0.01217508,
      "auxiliary_loss_mlp": 0.01028219,
      "balance_loss_clip": 1.05707359,
      "balance_loss_mlp": 1.01906347,
      "epoch": 0.23182829315216738,
      "flos": 11254656435840.0,
      "grad_norm": 5.000355580302149,
      "language_loss": 0.75241917,
      "learning_rate": 3.587851979938107e-06,
      "loss": 0.77487642,
      "num_input_tokens_seen": 41153125,
      "step": 1928,
      "time_per_iteration": 2.7618203163146973
    },
    {
      "auxiliary_loss_clip": 0.0121349,
      "auxiliary_loss_mlp": 0.01026549,
      "balance_loss_clip": 1.0600419,
      "balance_loss_mlp": 1.01748276,
      "epoch": 0.23194853604280646,
      "flos": 19828939155840.0,
      "grad_norm": 2.292940018716132,
      "language_loss": 0.77657163,
      "learning_rate": 3.5873782335257985e-06,
      "loss": 0.79897201,
      "num_input_tokens_seen": 41171290,
      "step": 1929,
      "time_per_iteration": 2.6061019897460938
    },
    {
      "auxiliary_loss_clip": 0.01215882,
      "auxiliary_loss_mlp": 0.01029302,
      "balance_loss_clip": 1.05768096,
      "balance_loss_mlp": 1.0198487,
      "epoch": 0.23206877893344555,
      "flos": 15305020830720.0,
      "grad_norm": 2.0121864343348665,
      "language_loss": 0.78702629,
      "learning_rate": 3.5869042463094744e-06,
      "loss": 0.8094781,
      "num_input_tokens_seen": 41189005,
      "step": 1930,
      "time_per_iteration": 2.6737401485443115
    },
    {
      "auxiliary_loss_clip": 0.01203005,
      "auxiliary_loss_mlp": 0.01034439,
      "balance_loss_clip": 1.05067647,
      "balance_loss_mlp": 1.024634,
      "epoch": 0.23218902182408466,
      "flos": 22711488572160.0,
      "grad_norm": 1.851159494668266,
      "language_loss": 0.77055407,
      "learning_rate": 3.586430018361038e-06,
      "loss": 0.79292852,
      "num_input_tokens_seen": 41208775,
      "step": 1931,
      "time_per_iteration": 2.6629269123077393
    },
    {
      "auxiliary_loss_clip": 0.01200858,
      "auxiliary_loss_mlp": 0.01032318,
      "balance_loss_clip": 1.05154204,
      "balance_loss_mlp": 1.0223757,
      "epoch": 0.23230926471472374,
      "flos": 22710734386560.0,
      "grad_norm": 2.16705454340042,
      "language_loss": 0.76317036,
      "learning_rate": 3.5859555497524283e-06,
      "loss": 0.7855022,
      "num_input_tokens_seen": 41226010,
      "step": 1932,
      "time_per_iteration": 2.6517717838287354
    },
    {
      "auxiliary_loss_clip": 0.0121576,
      "auxiliary_loss_mlp": 0.01035249,
      "balance_loss_clip": 1.0609957,
      "balance_loss_mlp": 1.02658224,
      "epoch": 0.23242950760536282,
      "flos": 20375499479040.0,
      "grad_norm": 1.816784635462696,
      "language_loss": 0.9203707,
      "learning_rate": 3.5854808405556237e-06,
      "loss": 0.94288087,
      "num_input_tokens_seen": 41245245,
      "step": 1933,
      "time_per_iteration": 2.5968031883239746
    },
    {
      "auxiliary_loss_clip": 0.01217599,
      "auxiliary_loss_mlp": 0.01032391,
      "balance_loss_clip": 1.05533624,
      "balance_loss_mlp": 1.0242548,
      "epoch": 0.23254975049600193,
      "flos": 16908324301440.0,
      "grad_norm": 2.1184200623596774,
      "language_loss": 0.7506001,
      "learning_rate": 3.5850058908426355e-06,
      "loss": 0.77310002,
      "num_input_tokens_seen": 41263795,
      "step": 1934,
      "time_per_iteration": 2.6336779594421387
    },
    {
      "auxiliary_loss_clip": 0.01215649,
      "auxiliary_loss_mlp": 0.01031928,
      "balance_loss_clip": 1.05276322,
      "balance_loss_mlp": 1.02298129,
      "epoch": 0.23266999338664102,
      "flos": 23294821443840.0,
      "grad_norm": 2.268537579096651,
      "language_loss": 0.85872054,
      "learning_rate": 3.584530700685514e-06,
      "loss": 0.88119626,
      "num_input_tokens_seen": 41284055,
      "step": 1935,
      "time_per_iteration": 2.5885331630706787
    },
    {
      "auxiliary_loss_clip": 0.01213093,
      "auxiliary_loss_mlp": 0.01028483,
      "balance_loss_clip": 1.05937624,
      "balance_loss_mlp": 1.01973259,
      "epoch": 0.2327902362772801,
      "flos": 19569987031680.0,
      "grad_norm": 2.9574837887126004,
      "language_loss": 0.89026105,
      "learning_rate": 3.5840552701563448e-06,
      "loss": 0.91267687,
      "num_input_tokens_seen": 41300255,
      "step": 1936,
      "time_per_iteration": 2.6206111907958984
    },
    {
      "auxiliary_loss_clip": 0.01207116,
      "auxiliary_loss_mlp": 0.01031134,
      "balance_loss_clip": 1.05760431,
      "balance_loss_mlp": 1.02197242,
      "epoch": 0.2329104791679192,
      "flos": 16727514215040.0,
      "grad_norm": 2.5431222159266316,
      "language_loss": 0.8200146,
      "learning_rate": 3.5835795993272513e-06,
      "loss": 0.84239709,
      "num_input_tokens_seen": 41318540,
      "step": 1937,
      "time_per_iteration": 2.545825242996216
    },
    {
      "auxiliary_loss_clip": 0.01216089,
      "auxiliary_loss_mlp": 0.01036257,
      "balance_loss_clip": 1.04481459,
      "balance_loss_mlp": 1.02710187,
      "epoch": 0.2330307220585583,
      "flos": 22163743100160.0,
      "grad_norm": 1.8866107452447778,
      "language_loss": 0.71171427,
      "learning_rate": 3.583103688270391e-06,
      "loss": 0.73423779,
      "num_input_tokens_seen": 41338320,
      "step": 1938,
      "time_per_iteration": 3.9196598529815674
    },
    {
      "auxiliary_loss_clip": 0.01201237,
      "auxiliary_loss_mlp": 0.01031425,
      "balance_loss_clip": 1.05112672,
      "balance_loss_mlp": 1.02142954,
      "epoch": 0.23315096494919738,
      "flos": 19317319787520.0,
      "grad_norm": 2.0692426696649466,
      "language_loss": 0.89463639,
      "learning_rate": 3.58262753705796e-06,
      "loss": 0.91696298,
      "num_input_tokens_seen": 41353210,
      "step": 1939,
      "time_per_iteration": 3.709167718887329
    },
    {
      "auxiliary_loss_clip": 0.01108308,
      "auxiliary_loss_mlp": 0.01020949,
      "balance_loss_clip": 1.01618433,
      "balance_loss_mlp": 1.01894629,
      "epoch": 0.23327120783983646,
      "flos": 53031048946560.0,
      "grad_norm": 0.7609889759862565,
      "language_loss": 0.55503976,
      "learning_rate": 3.5821511457621902e-06,
      "loss": 0.57633233,
      "num_input_tokens_seen": 41410510,
      "step": 1940,
      "time_per_iteration": 3.1654741764068604
    },
    {
      "auxiliary_loss_clip": 0.0120983,
      "auxiliary_loss_mlp": 0.0103089,
      "balance_loss_clip": 1.05445552,
      "balance_loss_mlp": 1.02088261,
      "epoch": 0.23339145073047557,
      "flos": 17126984344320.0,
      "grad_norm": 3.3969412260890235,
      "language_loss": 0.81733519,
      "learning_rate": 3.5816745144553497e-06,
      "loss": 0.83974242,
      "num_input_tokens_seen": 41425830,
      "step": 1941,
      "time_per_iteration": 2.625107765197754
    },
    {
      "auxiliary_loss_clip": 0.01211274,
      "auxiliary_loss_mlp": 0.01027542,
      "balance_loss_clip": 1.05207634,
      "balance_loss_mlp": 1.018399,
      "epoch": 0.23351169362111465,
      "flos": 13078918419840.0,
      "grad_norm": 2.7141974848713737,
      "language_loss": 0.75300163,
      "learning_rate": 3.5811976432097424e-06,
      "loss": 0.77538979,
      "num_input_tokens_seen": 41443500,
      "step": 1942,
      "time_per_iteration": 2.6982364654541016
    },
    {
      "auxiliary_loss_clip": 0.0121231,
      "auxiliary_loss_mlp": 0.01208027,
      "balance_loss_clip": 1.06030583,
      "balance_loss_mlp": 0.99999535,
      "epoch": 0.23363193651175373,
      "flos": 15851257931520.0,
      "grad_norm": 2.494506189131233,
      "language_loss": 0.85061622,
      "learning_rate": 3.58072053209771e-06,
      "loss": 0.87481964,
      "num_input_tokens_seen": 41460055,
      "step": 1943,
      "time_per_iteration": 3.5754058361053467
    },
    {
      "auxiliary_loss_clip": 0.01207349,
      "auxiliary_loss_mlp": 0.01034034,
      "balance_loss_clip": 1.05273092,
      "balance_loss_mlp": 1.02499831,
      "epoch": 0.23375217940239285,
      "flos": 21025769345280.0,
      "grad_norm": 1.908442232010515,
      "language_loss": 0.79380465,
      "learning_rate": 3.5802431811916296e-06,
      "loss": 0.8162185,
      "num_input_tokens_seen": 41476665,
      "step": 1944,
      "time_per_iteration": 2.6529970169067383
    },
    {
      "auxiliary_loss_clip": 0.0120988,
      "auxiliary_loss_mlp": 0.01028231,
      "balance_loss_clip": 1.0568496,
      "balance_loss_mlp": 1.01910508,
      "epoch": 0.23387242229303193,
      "flos": 20594698225920.0,
      "grad_norm": 1.7737495057284072,
      "language_loss": 0.80638951,
      "learning_rate": 3.579765590563916e-06,
      "loss": 0.82877064,
      "num_input_tokens_seen": 41496065,
      "step": 1945,
      "time_per_iteration": 3.5756115913391113
    },
    {
      "auxiliary_loss_clip": 0.01201284,
      "auxiliary_loss_mlp": 0.01034679,
      "balance_loss_clip": 1.05470896,
      "balance_loss_mlp": 1.02536309,
      "epoch": 0.233992665183671,
      "flos": 24279491952000.0,
      "grad_norm": 1.857846726301916,
      "language_loss": 0.81760395,
      "learning_rate": 3.579287760287017e-06,
      "loss": 0.83996356,
      "num_input_tokens_seen": 41516815,
      "step": 1946,
      "time_per_iteration": 2.6142051219940186
    },
    {
      "auxiliary_loss_clip": 0.01207983,
      "auxiliary_loss_mlp": 0.01034125,
      "balance_loss_clip": 1.05648005,
      "balance_loss_mlp": 1.025316,
      "epoch": 0.2341129080743101,
      "flos": 30154621121280.0,
      "grad_norm": 1.7640931857616915,
      "language_loss": 0.73454821,
      "learning_rate": 3.578809690433421e-06,
      "loss": 0.75696933,
      "num_input_tokens_seen": 41538525,
      "step": 1947,
      "time_per_iteration": 2.733745813369751
    },
    {
      "auxiliary_loss_clip": 0.01214279,
      "auxiliary_loss_mlp": 0.01033103,
      "balance_loss_clip": 1.05990243,
      "balance_loss_mlp": 1.0237453,
      "epoch": 0.2342331509649492,
      "flos": 22784135829120.0,
      "grad_norm": 3.457571619944202,
      "language_loss": 0.81661892,
      "learning_rate": 3.578331381075651e-06,
      "loss": 0.83909273,
      "num_input_tokens_seen": 41559025,
      "step": 1948,
      "time_per_iteration": 2.619236946105957
    },
    {
      "auxiliary_loss_clip": 0.01212026,
      "auxiliary_loss_mlp": 0.01028357,
      "balance_loss_clip": 1.05555356,
      "balance_loss_mlp": 1.01876092,
      "epoch": 0.2343533938555883,
      "flos": 23623152687360.0,
      "grad_norm": 2.200459704903917,
      "language_loss": 0.69752389,
      "learning_rate": 3.5778528322862646e-06,
      "loss": 0.71992767,
      "num_input_tokens_seen": 41577845,
      "step": 1949,
      "time_per_iteration": 2.681548833847046
    },
    {
      "auxiliary_loss_clip": 0.01211418,
      "auxiliary_loss_mlp": 0.01030554,
      "balance_loss_clip": 1.05491114,
      "balance_loss_mlp": 1.02146411,
      "epoch": 0.23447363674622737,
      "flos": 24570332375040.0,
      "grad_norm": 1.7482257715736402,
      "language_loss": 0.86677206,
      "learning_rate": 3.5773740441378585e-06,
      "loss": 0.88919181,
      "num_input_tokens_seen": 41598600,
      "step": 1950,
      "time_per_iteration": 2.6221923828125
    },
    {
      "auxiliary_loss_clip": 0.01206693,
      "auxiliary_loss_mlp": 0.01034571,
      "balance_loss_clip": 1.05515647,
      "balance_loss_mlp": 1.0261848,
      "epoch": 0.23459387963686648,
      "flos": 53140322119680.0,
      "grad_norm": 2.086367853132221,
      "language_loss": 0.73834646,
      "learning_rate": 3.5768950167030633e-06,
      "loss": 0.76075912,
      "num_input_tokens_seen": 41623300,
      "step": 1951,
      "time_per_iteration": 2.9083592891693115
    },
    {
      "auxiliary_loss_clip": 0.01201563,
      "auxiliary_loss_mlp": 0.01041713,
      "balance_loss_clip": 1.05116034,
      "balance_loss_mlp": 1.03174722,
      "epoch": 0.23471412252750556,
      "flos": 23951412103680.0,
      "grad_norm": 2.020221816596833,
      "language_loss": 0.7893517,
      "learning_rate": 3.576415750054548e-06,
      "loss": 0.81178451,
      "num_input_tokens_seen": 41643420,
      "step": 1952,
      "time_per_iteration": 2.636072874069214
    },
    {
      "auxiliary_loss_clip": 0.01204416,
      "auxiliary_loss_mlp": 0.01028739,
      "balance_loss_clip": 1.05339348,
      "balance_loss_mlp": 1.01958382,
      "epoch": 0.23483436541814465,
      "flos": 15706573948800.0,
      "grad_norm": 1.8661692584862084,
      "language_loss": 0.86017525,
      "learning_rate": 3.5759362442650172e-06,
      "loss": 0.88250685,
      "num_input_tokens_seen": 41660170,
      "step": 1953,
      "time_per_iteration": 2.704658269882202
    },
    {
      "auxiliary_loss_clip": 0.01210388,
      "auxiliary_loss_mlp": 0.01032535,
      "balance_loss_clip": 1.05794239,
      "balance_loss_mlp": 1.0233736,
      "epoch": 0.23495460830878373,
      "flos": 24936262179840.0,
      "grad_norm": 3.3863611603708046,
      "language_loss": 0.85398602,
      "learning_rate": 3.5754564994072113e-06,
      "loss": 0.87641525,
      "num_input_tokens_seen": 41679010,
      "step": 1954,
      "time_per_iteration": 2.6544995307922363
    },
    {
      "auxiliary_loss_clip": 0.01208012,
      "auxiliary_loss_mlp": 0.01029418,
      "balance_loss_clip": 1.05200005,
      "balance_loss_mlp": 1.01964307,
      "epoch": 0.23507485119942284,
      "flos": 30482665056000.0,
      "grad_norm": 27.810358268174795,
      "language_loss": 0.59394836,
      "learning_rate": 3.5749765155539067e-06,
      "loss": 0.6163227,
      "num_input_tokens_seen": 41699495,
      "step": 1955,
      "time_per_iteration": 2.745849132537842
    },
    {
      "auxiliary_loss_clip": 0.01209759,
      "auxiliary_loss_mlp": 0.01029886,
      "balance_loss_clip": 1.0503391,
      "balance_loss_mlp": 1.01993799,
      "epoch": 0.23519509409006192,
      "flos": 18329129746560.0,
      "grad_norm": 2.5436679750348716,
      "language_loss": 0.92604542,
      "learning_rate": 3.574496292777917e-06,
      "loss": 0.9484418,
      "num_input_tokens_seen": 41717705,
      "step": 1956,
      "time_per_iteration": 2.7078773975372314
    },
    {
      "auxiliary_loss_clip": 0.01219326,
      "auxiliary_loss_mlp": 0.01034804,
      "balance_loss_clip": 1.05624342,
      "balance_loss_mlp": 1.02425456,
      "epoch": 0.235315336980701,
      "flos": 29643217234560.0,
      "grad_norm": 2.329639119777932,
      "language_loss": 0.71620095,
      "learning_rate": 3.574015831152092e-06,
      "loss": 0.73874223,
      "num_input_tokens_seen": 41738120,
      "step": 1957,
      "time_per_iteration": 2.73600172996521
    },
    {
      "auxiliary_loss_clip": 0.01200711,
      "auxiliary_loss_mlp": 0.01025076,
      "balance_loss_clip": 1.05115235,
      "balance_loss_mlp": 1.01621842,
      "epoch": 0.23543557987134012,
      "flos": 18551704371840.0,
      "grad_norm": 2.0842838122212233,
      "language_loss": 0.8405053,
      "learning_rate": 3.573535130749316e-06,
      "loss": 0.86276317,
      "num_input_tokens_seen": 41756070,
      "step": 1958,
      "time_per_iteration": 2.61672043800354
    },
    {
      "auxiliary_loss_clip": 0.01203264,
      "auxiliary_loss_mlp": 0.01033424,
      "balance_loss_clip": 1.05353141,
      "balance_loss_mlp": 1.02431631,
      "epoch": 0.2355558227619792,
      "flos": 24679033908480.0,
      "grad_norm": 1.659244791088402,
      "language_loss": 0.73746651,
      "learning_rate": 3.5730541916425127e-06,
      "loss": 0.7598334,
      "num_input_tokens_seen": 41777550,
      "step": 1959,
      "time_per_iteration": 2.724787712097168
    },
    {
      "auxiliary_loss_clip": 0.01215808,
      "auxiliary_loss_mlp": 0.01028455,
      "balance_loss_clip": 1.05353022,
      "balance_loss_mlp": 1.01932979,
      "epoch": 0.23567606565261828,
      "flos": 21944795748480.0,
      "grad_norm": 2.0116836769827513,
      "language_loss": 0.86113,
      "learning_rate": 3.572573013904639e-06,
      "loss": 0.8835727,
      "num_input_tokens_seen": 41797460,
      "step": 1960,
      "time_per_iteration": 2.726834774017334
    },
    {
      "auxiliary_loss_clip": 0.01206761,
      "auxiliary_loss_mlp": 0.0102541,
      "balance_loss_clip": 1.05589819,
      "balance_loss_mlp": 1.01632047,
      "epoch": 0.2357963085432574,
      "flos": 13589352639360.0,
      "grad_norm": 2.482029495875349,
      "language_loss": 0.92512381,
      "learning_rate": 3.572091597608689e-06,
      "loss": 0.94744551,
      "num_input_tokens_seen": 41815585,
      "step": 1961,
      "time_per_iteration": 2.549802541732788
    },
    {
      "auxiliary_loss_clip": 0.01217194,
      "auxiliary_loss_mlp": 0.01028268,
      "balance_loss_clip": 1.05437887,
      "balance_loss_mlp": 1.01823699,
      "epoch": 0.23591655143389648,
      "flos": 22088689632000.0,
      "grad_norm": 2.298254231247352,
      "language_loss": 0.73962718,
      "learning_rate": 3.571609942827694e-06,
      "loss": 0.76208174,
      "num_input_tokens_seen": 41834700,
      "step": 1962,
      "time_per_iteration": 2.6761162281036377
    },
    {
      "auxiliary_loss_clip": 0.01209288,
      "auxiliary_loss_mlp": 0.01030066,
      "balance_loss_clip": 1.05322695,
      "balance_loss_mlp": 1.02096462,
      "epoch": 0.23603679432453556,
      "flos": 17017349057280.0,
      "grad_norm": 1.789179581918357,
      "language_loss": 0.88497972,
      "learning_rate": 3.57112804963472e-06,
      "loss": 0.90737319,
      "num_input_tokens_seen": 41852915,
      "step": 1963,
      "time_per_iteration": 2.653686285018921
    },
    {
      "auxiliary_loss_clip": 0.01209828,
      "auxiliary_loss_mlp": 0.0103211,
      "balance_loss_clip": 1.05458403,
      "balance_loss_mlp": 1.02328801,
      "epoch": 0.23615703721517464,
      "flos": 19171307001600.0,
      "grad_norm": 1.6778748337519087,
      "language_loss": 0.76473552,
      "learning_rate": 3.57064591810287e-06,
      "loss": 0.78715491,
      "num_input_tokens_seen": 41870415,
      "step": 1964,
      "time_per_iteration": 3.5611915588378906
    },
    {
      "auxiliary_loss_clip": 0.01208204,
      "auxiliary_loss_mlp": 0.01208334,
      "balance_loss_clip": 1.05908012,
      "balance_loss_mlp": 0.99998969,
      "epoch": 0.23627728010581375,
      "flos": 19098803399040.0,
      "grad_norm": 2.449856062327435,
      "language_loss": 0.80970407,
      "learning_rate": 3.570163548305284e-06,
      "loss": 0.83386952,
      "num_input_tokens_seen": 41889345,
      "step": 1965,
      "time_per_iteration": 2.589942216873169
    },
    {
      "auxiliary_loss_clip": 0.01209301,
      "auxiliary_loss_mlp": 0.01033169,
      "balance_loss_clip": 1.0530467,
      "balance_loss_mlp": 1.02348304,
      "epoch": 0.23639752299645284,
      "flos": 14282213057280.0,
      "grad_norm": 2.5975719538411988,
      "language_loss": 0.69980794,
      "learning_rate": 3.569680940315135e-06,
      "loss": 0.72223264,
      "num_input_tokens_seen": 41905745,
      "step": 1966,
      "time_per_iteration": 3.6184043884277344
    },
    {
      "auxiliary_loss_clip": 0.01217196,
      "auxiliary_loss_mlp": 0.01032855,
      "balance_loss_clip": 1.05240393,
      "balance_loss_mlp": 1.02260327,
      "epoch": 0.23651776588709192,
      "flos": 22893411980160.0,
      "grad_norm": 2.7369957593637033,
      "language_loss": 0.82169086,
      "learning_rate": 3.5691980942056356e-06,
      "loss": 0.84419143,
      "num_input_tokens_seen": 41925115,
      "step": 1967,
      "time_per_iteration": 2.7006256580352783
    },
    {
      "auxiliary_loss_clip": 0.01213721,
      "auxiliary_loss_mlp": 0.01028307,
      "balance_loss_clip": 1.05489945,
      "balance_loss_mlp": 1.01899099,
      "epoch": 0.23663800877773103,
      "flos": 18624531196800.0,
      "grad_norm": 1.7434525442563567,
      "language_loss": 0.797194,
      "learning_rate": 3.5687150100500332e-06,
      "loss": 0.81961429,
      "num_input_tokens_seen": 41944815,
      "step": 1968,
      "time_per_iteration": 2.664184808731079
    },
    {
      "auxiliary_loss_clip": 0.01210817,
      "auxiliary_loss_mlp": 0.01027234,
      "balance_loss_clip": 1.05545378,
      "balance_loss_mlp": 1.01764321,
      "epoch": 0.2367582516683701,
      "flos": 25555828896000.0,
      "grad_norm": 1.694688309954491,
      "language_loss": 0.74687374,
      "learning_rate": 3.568231687921611e-06,
      "loss": 0.76925421,
      "num_input_tokens_seen": 41964990,
      "step": 1969,
      "time_per_iteration": 2.720313310623169
    },
    {
      "auxiliary_loss_clip": 0.01206884,
      "auxiliary_loss_mlp": 0.01031705,
      "balance_loss_clip": 1.05750465,
      "balance_loss_mlp": 1.02282977,
      "epoch": 0.2368784945590092,
      "flos": 23295072839040.0,
      "grad_norm": 1.5219841054959622,
      "language_loss": 0.803159,
      "learning_rate": 3.5677481278936883e-06,
      "loss": 0.82554483,
      "num_input_tokens_seen": 41984570,
      "step": 1970,
      "time_per_iteration": 3.502931833267212
    },
    {
      "auxiliary_loss_clip": 0.01117887,
      "auxiliary_loss_mlp": 0.01014383,
      "balance_loss_clip": 1.02465641,
      "balance_loss_mlp": 1.01232076,
      "epoch": 0.23699873744964828,
      "flos": 69859291875840.0,
      "grad_norm": 0.8352433381818106,
      "language_loss": 0.57858843,
      "learning_rate": 3.5672643300396214e-06,
      "loss": 0.59991109,
      "num_input_tokens_seen": 42053715,
      "step": 1971,
      "time_per_iteration": 3.2739107608795166
    },
    {
      "auxiliary_loss_clip": 0.01214711,
      "auxiliary_loss_mlp": 0.01029115,
      "balance_loss_clip": 1.05378222,
      "balance_loss_mlp": 1.02020407,
      "epoch": 0.2371189803402874,
      "flos": 21835052720640.0,
      "grad_norm": 2.067142602745322,
      "language_loss": 0.67348433,
      "learning_rate": 3.566780294432802e-06,
      "loss": 0.69592255,
      "num_input_tokens_seen": 42070890,
      "step": 1972,
      "time_per_iteration": 3.6543796062469482
    },
    {
      "auxiliary_loss_clip": 0.01210246,
      "auxiliary_loss_mlp": 0.01030587,
      "balance_loss_clip": 1.0588243,
      "balance_loss_mlp": 1.02190256,
      "epoch": 0.23723922323092647,
      "flos": 21908490076800.0,
      "grad_norm": 2.619666726888747,
      "language_loss": 0.74516326,
      "learning_rate": 3.566296021146657e-06,
      "loss": 0.76757163,
      "num_input_tokens_seen": 42090270,
      "step": 1973,
      "time_per_iteration": 2.645012378692627
    },
    {
      "auxiliary_loss_clip": 0.01215909,
      "auxiliary_loss_mlp": 0.01026949,
      "balance_loss_clip": 1.0612576,
      "balance_loss_mlp": 1.01664948,
      "epoch": 0.23735946612156555,
      "flos": 32708803380480.0,
      "grad_norm": 1.5686921663284632,
      "language_loss": 0.72976267,
      "learning_rate": 3.565811510254652e-06,
      "loss": 0.75219131,
      "num_input_tokens_seen": 42111150,
      "step": 1974,
      "time_per_iteration": 2.731044292449951
    },
    {
      "auxiliary_loss_clip": 0.01108308,
      "auxiliary_loss_mlp": 0.00999928,
      "balance_loss_clip": 1.02417135,
      "balance_loss_mlp": 0.99809211,
      "epoch": 0.23747970901220466,
      "flos": 70546944821760.0,
      "grad_norm": 0.8371055407295794,
      "language_loss": 0.58215511,
      "learning_rate": 3.5653267618302845e-06,
      "loss": 0.60323751,
      "num_input_tokens_seen": 42178730,
      "step": 1975,
      "time_per_iteration": 3.221342086791992
    },
    {
      "auxiliary_loss_clip": 0.01208236,
      "auxiliary_loss_mlp": 0.01031053,
      "balance_loss_clip": 1.05704832,
      "balance_loss_mlp": 1.02125967,
      "epoch": 0.23759995190284375,
      "flos": 20849807594880.0,
      "grad_norm": 1.6782893501145884,
      "language_loss": 0.85716963,
      "learning_rate": 3.564841775947093e-06,
      "loss": 0.8795625,
      "num_input_tokens_seen": 42199620,
      "step": 1976,
      "time_per_iteration": 2.6842751502990723
    },
    {
      "auxiliary_loss_clip": 0.01215529,
      "auxiliary_loss_mlp": 0.01034182,
      "balance_loss_clip": 1.05275106,
      "balance_loss_mlp": 1.02431703,
      "epoch": 0.23772019479348283,
      "flos": 32921645420160.0,
      "grad_norm": 8.181512395872076,
      "language_loss": 0.76209646,
      "learning_rate": 3.5643565526786475e-06,
      "loss": 0.78459358,
      "num_input_tokens_seen": 42219560,
      "step": 1977,
      "time_per_iteration": 2.8137056827545166
    },
    {
      "auxiliary_loss_clip": 0.01209577,
      "auxiliary_loss_mlp": 0.01030617,
      "balance_loss_clip": 1.05808234,
      "balance_loss_mlp": 1.02109814,
      "epoch": 0.2378404376841219,
      "flos": 32342765834880.0,
      "grad_norm": 1.8365186296023792,
      "language_loss": 0.77395129,
      "learning_rate": 3.5638710920985574e-06,
      "loss": 0.79635322,
      "num_input_tokens_seen": 42241020,
      "step": 1978,
      "time_per_iteration": 2.6525914669036865
    },
    {
      "auxiliary_loss_clip": 0.01217359,
      "auxiliary_loss_mlp": 0.0120903,
      "balance_loss_clip": 1.05496061,
      "balance_loss_mlp": 1.00000429,
      "epoch": 0.23796068057476102,
      "flos": 22997624313600.0,
      "grad_norm": 6.461602283554373,
      "language_loss": 0.8224082,
      "learning_rate": 3.5633853942804655e-06,
      "loss": 0.84667206,
      "num_input_tokens_seen": 42259345,
      "step": 1979,
      "time_per_iteration": 2.70867919921875
    },
    {
      "auxiliary_loss_clip": 0.01215223,
      "auxiliary_loss_mlp": 0.01034705,
      "balance_loss_clip": 1.04972041,
      "balance_loss_mlp": 1.02463722,
      "epoch": 0.2380809234654001,
      "flos": 13480938414720.0,
      "grad_norm": 2.8151520626301636,
      "language_loss": 0.76863742,
      "learning_rate": 3.5628994592980527e-06,
      "loss": 0.79113668,
      "num_input_tokens_seen": 42277250,
      "step": 1980,
      "time_per_iteration": 2.6303670406341553
    },
    {
      "auxiliary_loss_clip": 0.01209125,
      "auxiliary_loss_mlp": 0.01034569,
      "balance_loss_clip": 1.0561645,
      "balance_loss_mlp": 1.0258069,
      "epoch": 0.2382011663560392,
      "flos": 16871803148160.0,
      "grad_norm": 1.879681718675801,
      "language_loss": 0.70427871,
      "learning_rate": 3.562413287225034e-06,
      "loss": 0.72671562,
      "num_input_tokens_seen": 42295360,
      "step": 1981,
      "time_per_iteration": 2.6319313049316406
    },
    {
      "auxiliary_loss_clip": 0.0120814,
      "auxiliary_loss_mlp": 0.0102825,
      "balance_loss_clip": 1.05691171,
      "balance_loss_mlp": 1.01876652,
      "epoch": 0.2383214092466783,
      "flos": 18441135331200.0,
      "grad_norm": 3.94564287351604,
      "language_loss": 0.89472961,
      "learning_rate": 3.5619268781351623e-06,
      "loss": 0.91709363,
      "num_input_tokens_seen": 42313430,
      "step": 1982,
      "time_per_iteration": 2.5890278816223145
    },
    {
      "auxiliary_loss_clip": 0.01207433,
      "auxiliary_loss_mlp": 0.01032238,
      "balance_loss_clip": 1.05528462,
      "balance_loss_mlp": 1.02355957,
      "epoch": 0.23844165213731738,
      "flos": 19755717281280.0,
      "grad_norm": 2.4172652616414125,
      "language_loss": 0.76820314,
      "learning_rate": 3.5614402321022256e-06,
      "loss": 0.79059982,
      "num_input_tokens_seen": 42331260,
      "step": 1983,
      "time_per_iteration": 2.6046524047851562
    },
    {
      "auxiliary_loss_clip": 0.0121184,
      "auxiliary_loss_mlp": 0.01032538,
      "balance_loss_clip": 1.05043244,
      "balance_loss_mlp": 1.02310288,
      "epoch": 0.23856189502795647,
      "flos": 23367360960000.0,
      "grad_norm": 1.973405266889139,
      "language_loss": 0.87521636,
      "learning_rate": 3.5609533492000463e-06,
      "loss": 0.89766014,
      "num_input_tokens_seen": 42350150,
      "step": 1984,
      "time_per_iteration": 2.662203550338745
    },
    {
      "auxiliary_loss_clip": 0.0120909,
      "auxiliary_loss_mlp": 0.01030717,
      "balance_loss_clip": 1.05568004,
      "balance_loss_mlp": 1.02099574,
      "epoch": 0.23868213791859555,
      "flos": 23475056912640.0,
      "grad_norm": 2.7668002538826686,
      "language_loss": 0.78644061,
      "learning_rate": 3.560466229502485e-06,
      "loss": 0.80883873,
      "num_input_tokens_seen": 42369495,
      "step": 1985,
      "time_per_iteration": 2.686342716217041
    },
    {
      "auxiliary_loss_clip": 0.01210326,
      "auxiliary_loss_mlp": 0.01208481,
      "balance_loss_clip": 1.05649066,
      "balance_loss_mlp": 1.00011957,
      "epoch": 0.23880238080923466,
      "flos": 16617340224000.0,
      "grad_norm": 4.289812902452912,
      "language_loss": 0.89867914,
      "learning_rate": 3.5599788730834384e-06,
      "loss": 0.92286724,
      "num_input_tokens_seen": 42387455,
      "step": 1986,
      "time_per_iteration": 2.581331491470337
    },
    {
      "auxiliary_loss_clip": 0.01214129,
      "auxiliary_loss_mlp": 0.0102918,
      "balance_loss_clip": 1.05664372,
      "balance_loss_mlp": 1.0197444,
      "epoch": 0.23892262369987374,
      "flos": 17348409734400.0,
      "grad_norm": 5.491457926476149,
      "language_loss": 0.78597879,
      "learning_rate": 3.559491280016836e-06,
      "loss": 0.8084119,
      "num_input_tokens_seen": 42405400,
      "step": 1987,
      "time_per_iteration": 2.6442363262176514
    },
    {
      "auxiliary_loss_clip": 0.01213881,
      "auxiliary_loss_mlp": 0.01035851,
      "balance_loss_clip": 1.05603957,
      "balance_loss_mlp": 1.02578974,
      "epoch": 0.23904286659051283,
      "flos": 22309899540480.0,
      "grad_norm": 2.859890193729336,
      "language_loss": 0.70636207,
      "learning_rate": 3.5590034503766465e-06,
      "loss": 0.72885936,
      "num_input_tokens_seen": 42425065,
      "step": 1988,
      "time_per_iteration": 2.7161922454833984
    },
    {
      "auxiliary_loss_clip": 0.01207817,
      "auxiliary_loss_mlp": 0.01033597,
      "balance_loss_clip": 1.05697799,
      "balance_loss_mlp": 1.0249064,
      "epoch": 0.23916310948115194,
      "flos": 21178246579200.0,
      "grad_norm": 2.6762495627025045,
      "language_loss": 0.81183237,
      "learning_rate": 3.558515384236874e-06,
      "loss": 0.83424652,
      "num_input_tokens_seen": 42442495,
      "step": 1989,
      "time_per_iteration": 2.568535804748535
    },
    {
      "auxiliary_loss_clip": 0.01207621,
      "auxiliary_loss_mlp": 0.01209071,
      "balance_loss_clip": 1.05383098,
      "balance_loss_mlp": 1.00021946,
      "epoch": 0.23928335237179102,
      "flos": 14137349506560.0,
      "grad_norm": 1.8368474685685499,
      "language_loss": 0.84360433,
      "learning_rate": 3.558027081671556e-06,
      "loss": 0.86777127,
      "num_input_tokens_seen": 42459480,
      "step": 1990,
      "time_per_iteration": 2.6334457397460938
    },
    {
      "auxiliary_loss_clip": 0.01213259,
      "auxiliary_loss_mlp": 0.01038958,
      "balance_loss_clip": 1.05458736,
      "balance_loss_mlp": 1.02843153,
      "epoch": 0.2394035952624301,
      "flos": 23769596436480.0,
      "grad_norm": 1.9013754361224622,
      "language_loss": 0.69229674,
      "learning_rate": 3.557538542754769e-06,
      "loss": 0.71481889,
      "num_input_tokens_seen": 42479175,
      "step": 1991,
      "time_per_iteration": 2.633265733718872
    },
    {
      "auxiliary_loss_clip": 0.01211125,
      "auxiliary_loss_mlp": 0.01036948,
      "balance_loss_clip": 1.0590229,
      "balance_loss_mlp": 1.02714336,
      "epoch": 0.2395238381530692,
      "flos": 24206198250240.0,
      "grad_norm": 2.083829283477848,
      "language_loss": 0.67103982,
      "learning_rate": 3.557049767560623e-06,
      "loss": 0.69352055,
      "num_input_tokens_seen": 42498090,
      "step": 1992,
      "time_per_iteration": 3.485339403152466
    },
    {
      "auxiliary_loss_clip": 0.01220004,
      "auxiliary_loss_mlp": 0.01031509,
      "balance_loss_clip": 1.05581224,
      "balance_loss_mlp": 1.02191293,
      "epoch": 0.2396440810437083,
      "flos": 25295763450240.0,
      "grad_norm": 3.9284431284363275,
      "language_loss": 0.85650659,
      "learning_rate": 3.5565607561632655e-06,
      "loss": 0.87902176,
      "num_input_tokens_seen": 42516930,
      "step": 1993,
      "time_per_iteration": 3.6558597087860107
    },
    {
      "auxiliary_loss_clip": 0.01207804,
      "auxiliary_loss_mlp": 0.01032173,
      "balance_loss_clip": 1.05284071,
      "balance_loss_mlp": 1.02185583,
      "epoch": 0.23976432393434738,
      "flos": 28543093436160.0,
      "grad_norm": 2.7485498153064047,
      "language_loss": 0.79624593,
      "learning_rate": 3.5560715086368787e-06,
      "loss": 0.81864572,
      "num_input_tokens_seen": 42534800,
      "step": 1994,
      "time_per_iteration": 2.729137897491455
    },
    {
      "auxiliary_loss_clip": 0.01205575,
      "auxiliary_loss_mlp": 0.01032967,
      "balance_loss_clip": 1.05347836,
      "balance_loss_mlp": 1.02359676,
      "epoch": 0.23988456682498646,
      "flos": 19494358945920.0,
      "grad_norm": 2.4391840979013013,
      "language_loss": 0.82878745,
      "learning_rate": 3.5555820250556816e-06,
      "loss": 0.85117286,
      "num_input_tokens_seen": 42552000,
      "step": 1995,
      "time_per_iteration": 2.6582791805267334
    },
    {
      "auxiliary_loss_clip": 0.01218266,
      "auxiliary_loss_mlp": 0.01032146,
      "balance_loss_clip": 1.05629778,
      "balance_loss_mlp": 1.02285933,
      "epoch": 0.24000480971562557,
      "flos": 20266331068800.0,
      "grad_norm": 2.2839934429882196,
      "language_loss": 0.69697577,
      "learning_rate": 3.5550923054939278e-06,
      "loss": 0.71947986,
      "num_input_tokens_seen": 42571455,
      "step": 1996,
      "time_per_iteration": 2.650437593460083
    },
    {
      "auxiliary_loss_clip": 0.01208816,
      "auxiliary_loss_mlp": 0.01033163,
      "balance_loss_clip": 1.04920673,
      "balance_loss_mlp": 1.0237155,
      "epoch": 0.24012505260626466,
      "flos": 25443176866560.0,
      "grad_norm": 2.1186219447786527,
      "language_loss": 0.75013912,
      "learning_rate": 3.5546023500259083e-06,
      "loss": 0.77255887,
      "num_input_tokens_seen": 42592550,
      "step": 1997,
      "time_per_iteration": 3.6498725414276123
    },
    {
      "auxiliary_loss_clip": 0.01220204,
      "auxiliary_loss_mlp": 0.01032082,
      "balance_loss_clip": 1.05093551,
      "balance_loss_mlp": 1.02258718,
      "epoch": 0.24024529549690374,
      "flos": 15553342529280.0,
      "grad_norm": 2.891583981933615,
      "language_loss": 0.8102448,
      "learning_rate": 3.5541121587259477e-06,
      "loss": 0.83276761,
      "num_input_tokens_seen": 42610385,
      "step": 1998,
      "time_per_iteration": 3.645998477935791
    },
    {
      "auxiliary_loss_clip": 0.01099717,
      "auxiliary_loss_mlp": 0.01001361,
      "balance_loss_clip": 1.01529431,
      "balance_loss_mlp": 0.99960822,
      "epoch": 0.24036553838754285,
      "flos": 57122351867520.0,
      "grad_norm": 0.8503955493745117,
      "language_loss": 0.5788703,
      "learning_rate": 3.553621731668408e-06,
      "loss": 0.59988105,
      "num_input_tokens_seen": 42673595,
      "step": 1999,
      "time_per_iteration": 3.1353273391723633
    },
    {
      "auxiliary_loss_clip": 0.01203228,
      "auxiliary_loss_mlp": 0.0103406,
      "balance_loss_clip": 1.051548,
      "balance_loss_mlp": 1.02424943,
      "epoch": 0.24048578127818193,
      "flos": 24969946158720.0,
      "grad_norm": 2.3093359854813142,
      "language_loss": 0.83579171,
      "learning_rate": 3.553131068927688e-06,
      "loss": 0.85816455,
      "num_input_tokens_seen": 42692000,
      "step": 2000,
      "time_per_iteration": 2.6747264862060547
    },
    {
      "auxiliary_loss_clip": 0.01211351,
      "auxiliary_loss_mlp": 0.01029827,
      "balance_loss_clip": 1.05185509,
      "balance_loss_mlp": 1.02133346,
      "epoch": 0.24060602416882101,
      "flos": 23330947547520.0,
      "grad_norm": 2.359245514430631,
      "language_loss": 0.80469602,
      "learning_rate": 3.552640170578219e-06,
      "loss": 0.82710779,
      "num_input_tokens_seen": 42712250,
      "step": 2001,
      "time_per_iteration": 2.6770575046539307
    },
    {
      "auxiliary_loss_clip": 0.01213151,
      "auxiliary_loss_mlp": 0.0103656,
      "balance_loss_clip": 1.05572987,
      "balance_loss_mlp": 1.02771437,
      "epoch": 0.2407262670594601,
      "flos": 14173260128640.0,
      "grad_norm": 2.317250608264155,
      "language_loss": 0.78041178,
      "learning_rate": 3.5521490366944703e-06,
      "loss": 0.8029089,
      "num_input_tokens_seen": 42729900,
      "step": 2002,
      "time_per_iteration": 2.6985318660736084
    },
    {
      "auxiliary_loss_clip": 0.01214343,
      "auxiliary_loss_mlp": 0.01028872,
      "balance_loss_clip": 1.0521915,
      "balance_loss_mlp": 1.01965117,
      "epoch": 0.2408465099500992,
      "flos": 13663113217920.0,
      "grad_norm": 2.361069060906144,
      "language_loss": 0.80016148,
      "learning_rate": 3.5516576673509474e-06,
      "loss": 0.82259357,
      "num_input_tokens_seen": 42747900,
      "step": 2003,
      "time_per_iteration": 2.630446434020996
    },
    {
      "auxiliary_loss_clip": 0.01210731,
      "auxiliary_loss_mlp": 0.01030408,
      "balance_loss_clip": 1.05874538,
      "balance_loss_mlp": 1.02129412,
      "epoch": 0.2409667528407383,
      "flos": 31248029076480.0,
      "grad_norm": 1.8328660258244953,
      "language_loss": 0.86113763,
      "learning_rate": 3.5511660626221896e-06,
      "loss": 0.88354897,
      "num_input_tokens_seen": 42768540,
      "step": 2004,
      "time_per_iteration": 2.632747173309326
    },
    {
      "auxiliary_loss_clip": 0.01210393,
      "auxiliary_loss_mlp": 0.01208708,
      "balance_loss_clip": 1.05469406,
      "balance_loss_mlp": 1.00023198,
      "epoch": 0.24108699573137737,
      "flos": 22199941031040.0,
      "grad_norm": 12.295903903585568,
      "language_loss": 0.89372712,
      "learning_rate": 3.5506742225827744e-06,
      "loss": 0.91791815,
      "num_input_tokens_seen": 42785395,
      "step": 2005,
      "time_per_iteration": 2.6655304431915283
    },
    {
      "auxiliary_loss_clip": 0.01215145,
      "auxiliary_loss_mlp": 0.0103034,
      "balance_loss_clip": 1.05463207,
      "balance_loss_mlp": 1.02057052,
      "epoch": 0.24120723862201648,
      "flos": 26103035664000.0,
      "grad_norm": 2.8612653632290637,
      "language_loss": 0.90521991,
      "learning_rate": 3.5501821473073116e-06,
      "loss": 0.92767471,
      "num_input_tokens_seen": 42801980,
      "step": 2006,
      "time_per_iteration": 2.7086870670318604
    },
    {
      "auxiliary_loss_clip": 0.01211023,
      "auxiliary_loss_mlp": 0.01038949,
      "balance_loss_clip": 1.05568039,
      "balance_loss_mlp": 1.02817845,
      "epoch": 0.24132748151265557,
      "flos": 18624926246400.0,
      "grad_norm": 1.9634701940044794,
      "language_loss": 0.86650491,
      "learning_rate": 3.54968983687045e-06,
      "loss": 0.88900465,
      "num_input_tokens_seen": 42818850,
      "step": 2007,
      "time_per_iteration": 2.728384017944336
    },
    {
      "auxiliary_loss_clip": 0.01215359,
      "auxiliary_loss_mlp": 0.01036716,
      "balance_loss_clip": 1.05587018,
      "balance_loss_mlp": 1.02606499,
      "epoch": 0.24144772440329465,
      "flos": 15267673664640.0,
      "grad_norm": 13.086469603359351,
      "language_loss": 0.893915,
      "learning_rate": 3.549197291346872e-06,
      "loss": 0.91643578,
      "num_input_tokens_seen": 42835375,
      "step": 2008,
      "time_per_iteration": 2.6438958644866943
    },
    {
      "auxiliary_loss_clip": 0.01211836,
      "auxiliary_loss_mlp": 0.01030985,
      "balance_loss_clip": 1.0550282,
      "balance_loss_mlp": 1.02186537,
      "epoch": 0.24156796729393373,
      "flos": 24024274842240.0,
      "grad_norm": 2.5617154871441388,
      "language_loss": 0.79292953,
      "learning_rate": 3.548704510811297e-06,
      "loss": 0.81535774,
      "num_input_tokens_seen": 42854570,
      "step": 2009,
      "time_per_iteration": 2.6731722354888916
    },
    {
      "auxiliary_loss_clip": 0.01223395,
      "auxiliary_loss_mlp": 0.01040424,
      "balance_loss_clip": 1.05238593,
      "balance_loss_mlp": 1.030267,
      "epoch": 0.24168821018457284,
      "flos": 26286790665600.0,
      "grad_norm": 2.730094249691029,
      "language_loss": 0.74570203,
      "learning_rate": 3.5482114953384787e-06,
      "loss": 0.76834023,
      "num_input_tokens_seen": 42873800,
      "step": 2010,
      "time_per_iteration": 2.783236265182495
    },
    {
      "auxiliary_loss_clip": 0.01213389,
      "auxiliary_loss_mlp": 0.01033608,
      "balance_loss_clip": 1.05648386,
      "balance_loss_mlp": 1.02362394,
      "epoch": 0.24180845307521193,
      "flos": 18223193560320.0,
      "grad_norm": 2.19185515934731,
      "language_loss": 0.84208,
      "learning_rate": 3.5477182450032077e-06,
      "loss": 0.86454999,
      "num_input_tokens_seen": 42892400,
      "step": 2011,
      "time_per_iteration": 2.7060554027557373
    },
    {
      "auxiliary_loss_clip": 0.01207357,
      "auxiliary_loss_mlp": 0.01038852,
      "balance_loss_clip": 1.05546904,
      "balance_loss_mlp": 1.02918983,
      "epoch": 0.241928695965851,
      "flos": 20449260057600.0,
      "grad_norm": 2.313899188583222,
      "language_loss": 0.8365705,
      "learning_rate": 3.5472247598803097e-06,
      "loss": 0.85903257,
      "num_input_tokens_seen": 42911745,
      "step": 2012,
      "time_per_iteration": 2.633824110031128
    },
    {
      "auxiliary_loss_clip": 0.01210024,
      "auxiliary_loss_mlp": 0.01035036,
      "balance_loss_clip": 1.05746067,
      "balance_loss_mlp": 1.0249747,
      "epoch": 0.24204893885649012,
      "flos": 25556475340800.0,
      "grad_norm": 2.8338950540326415,
      "language_loss": 0.85828364,
      "learning_rate": 3.546731040044645e-06,
      "loss": 0.88073421,
      "num_input_tokens_seen": 42926915,
      "step": 2013,
      "time_per_iteration": 2.67852520942688
    },
    {
      "auxiliary_loss_clip": 0.01209378,
      "auxiliary_loss_mlp": 0.01025896,
      "balance_loss_clip": 1.0577569,
      "balance_loss_mlp": 1.01655614,
      "epoch": 0.2421691817471292,
      "flos": 30660207004800.0,
      "grad_norm": 1.9727544241212158,
      "language_loss": 0.7535007,
      "learning_rate": 3.546237085571112e-06,
      "loss": 0.77585346,
      "num_input_tokens_seen": 42945350,
      "step": 2014,
      "time_per_iteration": 2.628012180328369
    },
    {
      "auxiliary_loss_clip": 0.01211348,
      "auxiliary_loss_mlp": 0.01033209,
      "balance_loss_clip": 1.05878067,
      "balance_loss_mlp": 1.02407777,
      "epoch": 0.24228942463776829,
      "flos": 21945011230080.0,
      "grad_norm": 2.9815963698159327,
      "language_loss": 0.72433889,
      "learning_rate": 3.5457428965346425e-06,
      "loss": 0.74678451,
      "num_input_tokens_seen": 42964290,
      "step": 2015,
      "time_per_iteration": 2.717008590698242
    },
    {
      "auxiliary_loss_clip": 0.01217991,
      "auxiliary_loss_mlp": 0.01030448,
      "balance_loss_clip": 1.05101991,
      "balance_loss_mlp": 1.02108991,
      "epoch": 0.2424096675284074,
      "flos": 33984493879680.0,
      "grad_norm": 1.7728342779211328,
      "language_loss": 0.74636078,
      "learning_rate": 3.545248473010205e-06,
      "loss": 0.76884514,
      "num_input_tokens_seen": 42987095,
      "step": 2016,
      "time_per_iteration": 2.816986083984375
    },
    {
      "auxiliary_loss_clip": 0.01213507,
      "auxiliary_loss_mlp": 0.012091,
      "balance_loss_clip": 1.05919838,
      "balance_loss_mlp": 1.00039268,
      "epoch": 0.24252991041904648,
      "flos": 21653416621440.0,
      "grad_norm": 1.9108267536226162,
      "language_loss": 0.87855172,
      "learning_rate": 3.544753815072802e-06,
      "loss": 0.90277779,
      "num_input_tokens_seen": 43005750,
      "step": 2017,
      "time_per_iteration": 2.6561906337738037
    },
    {
      "auxiliary_loss_clip": 0.01213911,
      "auxiliary_loss_mlp": 0.01030554,
      "balance_loss_clip": 1.04457998,
      "balance_loss_mlp": 1.02098179,
      "epoch": 0.24265015330968556,
      "flos": 21870065502720.0,
      "grad_norm": 2.114395390876972,
      "language_loss": 0.88924193,
      "learning_rate": 3.544258922797474e-06,
      "loss": 0.9116866,
      "num_input_tokens_seen": 43023870,
      "step": 2018,
      "time_per_iteration": 3.7418391704559326
    },
    {
      "auxiliary_loss_clip": 0.01207944,
      "auxiliary_loss_mlp": 0.01033923,
      "balance_loss_clip": 1.05783057,
      "balance_loss_mlp": 1.02477932,
      "epoch": 0.24277039620032465,
      "flos": 25628260671360.0,
      "grad_norm": 2.1703573182937514,
      "language_loss": 0.77723926,
      "learning_rate": 3.543763796259295e-06,
      "loss": 0.79965794,
      "num_input_tokens_seen": 43043825,
      "step": 2019,
      "time_per_iteration": 3.6454412937164307
    },
    {
      "auxiliary_loss_clip": 0.01212031,
      "auxiliary_loss_mlp": 0.01036652,
      "balance_loss_clip": 1.05674124,
      "balance_loss_mlp": 1.02638197,
      "epoch": 0.24289063909096376,
      "flos": 26286575184000.0,
      "grad_norm": 2.242529112094567,
      "language_loss": 0.90881503,
      "learning_rate": 3.5432684355333754e-06,
      "loss": 0.93130195,
      "num_input_tokens_seen": 43062480,
      "step": 2020,
      "time_per_iteration": 2.8016164302825928
    },
    {
      "auxiliary_loss_clip": 0.01207871,
      "auxiliary_loss_mlp": 0.0103576,
      "balance_loss_clip": 1.052984,
      "balance_loss_mlp": 1.02664685,
      "epoch": 0.24301088198160284,
      "flos": 25075056332160.0,
      "grad_norm": 2.5882513458045633,
      "language_loss": 0.76755798,
      "learning_rate": 3.5427728406948613e-06,
      "loss": 0.78999436,
      "num_input_tokens_seen": 43081595,
      "step": 2021,
      "time_per_iteration": 2.713953971862793
    },
    {
      "auxiliary_loss_clip": 0.01110933,
      "auxiliary_loss_mlp": 0.01004856,
      "balance_loss_clip": 1.01546121,
      "balance_loss_mlp": 1.00287676,
      "epoch": 0.24313112487224192,
      "flos": 69900948673920.0,
      "grad_norm": 0.7500474036050045,
      "language_loss": 0.57859457,
      "learning_rate": 3.542277011818934e-06,
      "loss": 0.59975243,
      "num_input_tokens_seen": 43145430,
      "step": 2022,
      "time_per_iteration": 3.3874971866607666
    },
    {
      "auxiliary_loss_clip": 0.01216332,
      "auxiliary_loss_mlp": 0.01033226,
      "balance_loss_clip": 1.05921555,
      "balance_loss_mlp": 1.02402902,
      "epoch": 0.24325136776288103,
      "flos": 40662334235520.0,
      "grad_norm": 2.115648455132653,
      "language_loss": 0.74012411,
      "learning_rate": 3.5417809489808104e-06,
      "loss": 0.76261967,
      "num_input_tokens_seen": 43167040,
      "step": 2023,
      "time_per_iteration": 3.7064764499664307
    },
    {
      "auxiliary_loss_clip": 0.01211538,
      "auxiliary_loss_mlp": 0.01035078,
      "balance_loss_clip": 1.05767822,
      "balance_loss_mlp": 1.02670336,
      "epoch": 0.24337161065352012,
      "flos": 25046400257280.0,
      "grad_norm": 1.8946863814973494,
      "language_loss": 0.72460425,
      "learning_rate": 3.5412846522557422e-06,
      "loss": 0.74707031,
      "num_input_tokens_seen": 43187930,
      "step": 2024,
      "time_per_iteration": 3.6285665035247803
    },
    {
      "auxiliary_loss_clip": 0.01209452,
      "auxiliary_loss_mlp": 0.01033595,
      "balance_loss_clip": 1.05851579,
      "balance_loss_mlp": 1.02415407,
      "epoch": 0.2434918535441592,
      "flos": 18661160090880.0,
      "grad_norm": 2.2941261511700577,
      "language_loss": 0.74168491,
      "learning_rate": 3.540788121719018e-06,
      "loss": 0.76411545,
      "num_input_tokens_seen": 43206350,
      "step": 2025,
      "time_per_iteration": 2.5783393383026123
    },
    {
      "auxiliary_loss_clip": 0.01209023,
      "auxiliary_loss_mlp": 0.01036766,
      "balance_loss_clip": 1.05654919,
      "balance_loss_mlp": 1.02748585,
      "epoch": 0.24361209643479828,
      "flos": 23915142345600.0,
      "grad_norm": 2.7204811787636225,
      "language_loss": 0.81978989,
      "learning_rate": 3.5402913574459604e-06,
      "loss": 0.84224784,
      "num_input_tokens_seen": 43226255,
      "step": 2026,
      "time_per_iteration": 2.7639658451080322
    },
    {
      "auxiliary_loss_clip": 0.01211051,
      "auxiliary_loss_mlp": 0.0102837,
      "balance_loss_clip": 1.04563785,
      "balance_loss_mlp": 1.01950634,
      "epoch": 0.2437323393254374,
      "flos": 28657505232000.0,
      "grad_norm": 1.826152761666639,
      "language_loss": 0.86173773,
      "learning_rate": 3.5397943595119297e-06,
      "loss": 0.88413191,
      "num_input_tokens_seen": 43247675,
      "step": 2027,
      "time_per_iteration": 2.774019956588745
    },
    {
      "auxiliary_loss_clip": 0.01207901,
      "auxiliary_loss_mlp": 0.0103258,
      "balance_loss_clip": 1.0557853,
      "balance_loss_mlp": 1.02253056,
      "epoch": 0.24385258221607647,
      "flos": 23550325862400.0,
      "grad_norm": 3.6308818740985087,
      "language_loss": 0.7747438,
      "learning_rate": 3.5392971279923177e-06,
      "loss": 0.79714864,
      "num_input_tokens_seen": 43265895,
      "step": 2028,
      "time_per_iteration": 2.67301344871521
    },
    {
      "auxiliary_loss_clip": 0.01209089,
      "auxiliary_loss_mlp": 0.01031425,
      "balance_loss_clip": 1.05135489,
      "balance_loss_mlp": 1.02092874,
      "epoch": 0.24397282510671556,
      "flos": 25336091445120.0,
      "grad_norm": 5.096431456643741,
      "language_loss": 0.83165765,
      "learning_rate": 3.5387996629625557e-06,
      "loss": 0.8540628,
      "num_input_tokens_seen": 43283485,
      "step": 2029,
      "time_per_iteration": 2.691152334213257
    },
    {
      "auxiliary_loss_clip": 0.01104426,
      "auxiliary_loss_mlp": 0.01000085,
      "balance_loss_clip": 1.01899552,
      "balance_loss_mlp": 0.99818939,
      "epoch": 0.24409306799735467,
      "flos": 65187421430400.0,
      "grad_norm": 0.8154612562072909,
      "language_loss": 0.54980361,
      "learning_rate": 3.5383019644981083e-06,
      "loss": 0.5708487,
      "num_input_tokens_seen": 43347180,
      "step": 2030,
      "time_per_iteration": 3.16172194480896
    },
    {
      "auxiliary_loss_clip": 0.01212273,
      "auxiliary_loss_mlp": 0.0102854,
      "balance_loss_clip": 1.05559969,
      "balance_loss_mlp": 1.01897955,
      "epoch": 0.24421331088799375,
      "flos": 19537093152000.0,
      "grad_norm": 2.379527897347444,
      "language_loss": 0.73103631,
      "learning_rate": 3.5378040326744763e-06,
      "loss": 0.75344443,
      "num_input_tokens_seen": 43366665,
      "step": 2031,
      "time_per_iteration": 2.6527023315429688
    },
    {
      "auxiliary_loss_clip": 0.01216453,
      "auxiliary_loss_mlp": 0.010316,
      "balance_loss_clip": 1.05610716,
      "balance_loss_mlp": 1.02279663,
      "epoch": 0.24433355377863283,
      "flos": 21068575378560.0,
      "grad_norm": 2.540383504715426,
      "language_loss": 0.85462064,
      "learning_rate": 3.5373058675671946e-06,
      "loss": 0.87710118,
      "num_input_tokens_seen": 43384670,
      "step": 2032,
      "time_per_iteration": 2.723475456237793
    },
    {
      "auxiliary_loss_clip": 0.01207277,
      "auxiliary_loss_mlp": 0.01030404,
      "balance_loss_clip": 1.04927993,
      "balance_loss_mlp": 1.01995516,
      "epoch": 0.24445379666927192,
      "flos": 22637189289600.0,
      "grad_norm": 2.0845078474276986,
      "language_loss": 0.72036189,
      "learning_rate": 3.536807469251836e-06,
      "loss": 0.74273866,
      "num_input_tokens_seen": 43403825,
      "step": 2033,
      "time_per_iteration": 2.7691829204559326
    },
    {
      "auxiliary_loss_clip": 0.01218535,
      "auxiliary_loss_mlp": 0.0103012,
      "balance_loss_clip": 1.0516057,
      "balance_loss_mlp": 1.02079237,
      "epoch": 0.24457403955991103,
      "flos": 21251612108160.0,
      "grad_norm": 2.0595392367767857,
      "language_loss": 0.82269067,
      "learning_rate": 3.5363088378040055e-06,
      "loss": 0.84517717,
      "num_input_tokens_seen": 43422715,
      "step": 2034,
      "time_per_iteration": 2.679854154586792
    },
    {
      "auxiliary_loss_clip": 0.01102134,
      "auxiliary_loss_mlp": 0.01200192,
      "balance_loss_clip": 1.0166949,
      "balance_loss_mlp": 1.00027215,
      "epoch": 0.2446942824505501,
      "flos": 66997820764800.0,
      "grad_norm": 0.7664878635679587,
      "language_loss": 0.64352262,
      "learning_rate": 3.5358099732993463e-06,
      "loss": 0.66654587,
      "num_input_tokens_seen": 43481825,
      "step": 2035,
      "time_per_iteration": 3.0832338333129883
    },
    {
      "auxiliary_loss_clip": 0.01215731,
      "auxiliary_loss_mlp": 0.01030448,
      "balance_loss_clip": 1.05401373,
      "balance_loss_mlp": 1.02157903,
      "epoch": 0.2448145253411892,
      "flos": 20411122792320.0,
      "grad_norm": 2.8868163832577527,
      "language_loss": 0.89470339,
      "learning_rate": 3.535310875813535e-06,
      "loss": 0.91716516,
      "num_input_tokens_seen": 43500220,
      "step": 2036,
      "time_per_iteration": 2.7083418369293213
    },
    {
      "auxiliary_loss_clip": 0.01207712,
      "auxiliary_loss_mlp": 0.0103125,
      "balance_loss_clip": 1.05515194,
      "balance_loss_mlp": 1.02215433,
      "epoch": 0.2449347682318283,
      "flos": 28804739080320.0,
      "grad_norm": 6.622066754322451,
      "language_loss": 0.81542021,
      "learning_rate": 3.5348115454222843e-06,
      "loss": 0.8378098,
      "num_input_tokens_seen": 43522805,
      "step": 2037,
      "time_per_iteration": 2.6934330463409424
    },
    {
      "auxiliary_loss_clip": 0.01207671,
      "auxiliary_loss_mlp": 0.01039281,
      "balance_loss_clip": 1.0513376,
      "balance_loss_mlp": 1.02985168,
      "epoch": 0.2450550111224674,
      "flos": 22528990546560.0,
      "grad_norm": 2.62639331625983,
      "language_loss": 0.8638441,
      "learning_rate": 3.5343119822013425e-06,
      "loss": 0.88631356,
      "num_input_tokens_seen": 43541915,
      "step": 2038,
      "time_per_iteration": 2.650214433670044
    },
    {
      "auxiliary_loss_clip": 0.01216872,
      "auxiliary_loss_mlp": 0.0103674,
      "balance_loss_clip": 1.05728102,
      "balance_loss_mlp": 1.02680969,
      "epoch": 0.24517525401310647,
      "flos": 21759137326080.0,
      "grad_norm": 1.931708485270935,
      "language_loss": 0.77474105,
      "learning_rate": 3.533812186226493e-06,
      "loss": 0.79727715,
      "num_input_tokens_seen": 43562625,
      "step": 2039,
      "time_per_iteration": 2.6633412837982178
    },
    {
      "auxiliary_loss_clip": 0.01204793,
      "auxiliary_loss_mlp": 0.01028802,
      "balance_loss_clip": 1.05628598,
      "balance_loss_mlp": 1.0198195,
      "epoch": 0.24529549690374555,
      "flos": 25043311687680.0,
      "grad_norm": 7.333619607201334,
      "language_loss": 0.75934523,
      "learning_rate": 3.5333121575735545e-06,
      "loss": 0.78168118,
      "num_input_tokens_seen": 43582265,
      "step": 2040,
      "time_per_iteration": 2.674344778060913
    },
    {
      "auxiliary_loss_clip": 0.01212709,
      "auxiliary_loss_mlp": 0.01035549,
      "balance_loss_clip": 1.05717826,
      "balance_loss_mlp": 1.02619123,
      "epoch": 0.24541573979438466,
      "flos": 32123638915200.0,
      "grad_norm": 2.102184573907691,
      "language_loss": 0.75446618,
      "learning_rate": 3.532811896318381e-06,
      "loss": 0.77694881,
      "num_input_tokens_seen": 43604335,
      "step": 2041,
      "time_per_iteration": 2.7179207801818848
    },
    {
      "auxiliary_loss_clip": 0.01219201,
      "auxiliary_loss_mlp": 0.01027787,
      "balance_loss_clip": 1.05419207,
      "balance_loss_mlp": 1.01783347,
      "epoch": 0.24553598268502375,
      "flos": 31357556622720.0,
      "grad_norm": 12.94420042733825,
      "language_loss": 0.82027924,
      "learning_rate": 3.5323114025368615e-06,
      "loss": 0.84274912,
      "num_input_tokens_seen": 43619400,
      "step": 2042,
      "time_per_iteration": 2.7606313228607178
    },
    {
      "auxiliary_loss_clip": 0.01206029,
      "auxiliary_loss_mlp": 0.01025695,
      "balance_loss_clip": 1.05387723,
      "balance_loss_mlp": 1.01621151,
      "epoch": 0.24565622557566283,
      "flos": 14027462824320.0,
      "grad_norm": 2.3747270670909097,
      "language_loss": 0.81935889,
      "learning_rate": 3.53181067630492e-06,
      "loss": 0.84167612,
      "num_input_tokens_seen": 43636870,
      "step": 2043,
      "time_per_iteration": 2.659994125366211
    },
    {
      "auxiliary_loss_clip": 0.012046,
      "auxiliary_loss_mlp": 0.01035211,
      "balance_loss_clip": 1.05355489,
      "balance_loss_mlp": 1.02567434,
      "epoch": 0.24577646846630194,
      "flos": 16581465515520.0,
      "grad_norm": 2.6746595482567557,
      "language_loss": 0.76085603,
      "learning_rate": 3.5313097176985175e-06,
      "loss": 0.78325415,
      "num_input_tokens_seen": 43655180,
      "step": 2044,
      "time_per_iteration": 2.653277635574341
    },
    {
      "auxiliary_loss_clip": 0.01211907,
      "auxiliary_loss_mlp": 0.01029374,
      "balance_loss_clip": 1.05882227,
      "balance_loss_mlp": 1.02027273,
      "epoch": 0.24589671135694102,
      "flos": 18807424272000.0,
      "grad_norm": 2.2280175080633216,
      "language_loss": 0.80907404,
      "learning_rate": 3.5308085267936482e-06,
      "loss": 0.83148682,
      "num_input_tokens_seen": 43672895,
      "step": 2045,
      "time_per_iteration": 3.5223400592803955
    },
    {
      "auxiliary_loss_clip": 0.01219539,
      "auxiliary_loss_mlp": 0.01208038,
      "balance_loss_clip": 1.05281758,
      "balance_loss_mlp": 1.00045693,
      "epoch": 0.2460169542475801,
      "flos": 19938538529280.0,
      "grad_norm": 1.8130220315566452,
      "language_loss": 0.90085781,
      "learning_rate": 3.530307103666342e-06,
      "loss": 0.92513359,
      "num_input_tokens_seen": 43691975,
      "step": 2046,
      "time_per_iteration": 3.7168118953704834
    },
    {
      "auxiliary_loss_clip": 0.01222856,
      "auxiliary_loss_mlp": 0.01034222,
      "balance_loss_clip": 1.05668151,
      "balance_loss_mlp": 1.0249536,
      "epoch": 0.24613719713821922,
      "flos": 24171221381760.0,
      "grad_norm": 1.7596224916260932,
      "language_loss": 0.80525827,
      "learning_rate": 3.5298054483926658e-06,
      "loss": 0.82782906,
      "num_input_tokens_seen": 43712670,
      "step": 2047,
      "time_per_iteration": 2.6993613243103027
    },
    {
      "auxiliary_loss_clip": 0.01220489,
      "auxiliary_loss_mlp": 0.01036148,
      "balance_loss_clip": 1.05960774,
      "balance_loss_mlp": 1.02670646,
      "epoch": 0.2462574400288583,
      "flos": 30221055325440.0,
      "grad_norm": 2.1529364312726633,
      "language_loss": 0.83091837,
      "learning_rate": 3.5293035610487187e-06,
      "loss": 0.85348469,
      "num_input_tokens_seen": 43732035,
      "step": 2048,
      "time_per_iteration": 2.733018636703491
    },
    {
      "auxiliary_loss_clip": 0.01106836,
      "auxiliary_loss_mlp": 0.01009285,
      "balance_loss_clip": 1.01626933,
      "balance_loss_mlp": 1.00723505,
      "epoch": 0.24637768291949738,
      "flos": 68943030819840.0,
      "grad_norm": 0.7233738808163339,
      "language_loss": 0.61985087,
      "learning_rate": 3.5288014417106374e-06,
      "loss": 0.64101207,
      "num_input_tokens_seen": 43798055,
      "step": 2049,
      "time_per_iteration": 4.156703948974609
    },
    {
      "auxiliary_loss_clip": 0.01213254,
      "auxiliary_loss_mlp": 0.01033643,
      "balance_loss_clip": 1.05494046,
      "balance_loss_mlp": 1.02427316,
      "epoch": 0.24649792581013646,
      "flos": 34383999922560.0,
      "grad_norm": 1.7403385432096647,
      "language_loss": 0.75370336,
      "learning_rate": 3.528299090454593e-06,
      "loss": 0.77617234,
      "num_input_tokens_seen": 43818590,
      "step": 2050,
      "time_per_iteration": 2.8563835620880127
    },
    {
      "auxiliary_loss_clip": 0.01213658,
      "auxiliary_loss_mlp": 0.01032912,
      "balance_loss_clip": 1.05528545,
      "balance_loss_mlp": 1.02320242,
      "epoch": 0.24661816870077558,
      "flos": 19680448331520.0,
      "grad_norm": 41.63314573509491,
      "language_loss": 0.82248908,
      "learning_rate": 3.527796507356792e-06,
      "loss": 0.84495473,
      "num_input_tokens_seen": 43832480,
      "step": 2051,
      "time_per_iteration": 3.5789129734039307
    },
    {
      "auxiliary_loss_clip": 0.01215832,
      "auxiliary_loss_mlp": 0.01034238,
      "balance_loss_clip": 1.05813169,
      "balance_loss_mlp": 1.0249455,
      "epoch": 0.24673841159141466,
      "flos": 20002279213440.0,
      "grad_norm": 3.051039451174164,
      "language_loss": 0.8918643,
      "learning_rate": 3.527293692493475e-06,
      "loss": 0.91436505,
      "num_input_tokens_seen": 43848345,
      "step": 2052,
      "time_per_iteration": 2.712341070175171
    },
    {
      "auxiliary_loss_clip": 0.01215232,
      "auxiliary_loss_mlp": 0.01033056,
      "balance_loss_clip": 1.05706596,
      "balance_loss_mlp": 1.02289963,
      "epoch": 0.24685865448205374,
      "flos": 21646593037440.0,
      "grad_norm": 2.543697559716457,
      "language_loss": 0.73079109,
      "learning_rate": 3.52679064594092e-06,
      "loss": 0.75327396,
      "num_input_tokens_seen": 43865685,
      "step": 2053,
      "time_per_iteration": 2.641420364379883
    },
    {
      "auxiliary_loss_clip": 0.01204267,
      "auxiliary_loss_mlp": 0.01033335,
      "balance_loss_clip": 1.04373169,
      "balance_loss_mlp": 1.02439415,
      "epoch": 0.24697889737269285,
      "flos": 17960470508160.0,
      "grad_norm": 2.221658468544031,
      "language_loss": 0.7466687,
      "learning_rate": 3.5262873677754375e-06,
      "loss": 0.76904476,
      "num_input_tokens_seen": 43883690,
      "step": 2054,
      "time_per_iteration": 2.6991140842437744
    },
    {
      "auxiliary_loss_clip": 0.01205279,
      "auxiliary_loss_mlp": 0.0103015,
      "balance_loss_clip": 1.05694532,
      "balance_loss_mlp": 1.02073812,
      "epoch": 0.24709914026333193,
      "flos": 27344611221120.0,
      "grad_norm": 2.6399234126614455,
      "language_loss": 0.80895972,
      "learning_rate": 3.5257838580733745e-06,
      "loss": 0.83131403,
      "num_input_tokens_seen": 43903295,
      "step": 2055,
      "time_per_iteration": 2.6319289207458496
    },
    {
      "auxiliary_loss_clip": 0.01214235,
      "auxiliary_loss_mlp": 0.01030537,
      "balance_loss_clip": 1.05730116,
      "balance_loss_mlp": 1.02151299,
      "epoch": 0.24721938315397102,
      "flos": 19275519335040.0,
      "grad_norm": 1.8511078797694935,
      "language_loss": 0.87408423,
      "learning_rate": 3.5252801169111138e-06,
      "loss": 0.896532,
      "num_input_tokens_seen": 43920960,
      "step": 2056,
      "time_per_iteration": 2.656096935272217
    },
    {
      "auxiliary_loss_clip": 0.01210917,
      "auxiliary_loss_mlp": 0.01035142,
      "balance_loss_clip": 1.05763566,
      "balance_loss_mlp": 1.02610004,
      "epoch": 0.2473396260446101,
      "flos": 23185796688000.0,
      "grad_norm": 1.832456628164393,
      "language_loss": 0.79771137,
      "learning_rate": 3.524776144365072e-06,
      "loss": 0.82017195,
      "num_input_tokens_seen": 43939415,
      "step": 2057,
      "time_per_iteration": 2.6510918140411377
    },
    {
      "auxiliary_loss_clip": 0.01206561,
      "auxiliary_loss_mlp": 0.01034171,
      "balance_loss_clip": 1.0577035,
      "balance_loss_mlp": 1.02469409,
      "epoch": 0.2474598689352492,
      "flos": 21142443697920.0,
      "grad_norm": 1.8377602643463375,
      "language_loss": 0.78945398,
      "learning_rate": 3.5242719405117016e-06,
      "loss": 0.81186128,
      "num_input_tokens_seen": 43959220,
      "step": 2058,
      "time_per_iteration": 2.728766441345215
    },
    {
      "auxiliary_loss_clip": 0.01215994,
      "auxiliary_loss_mlp": 0.01208927,
      "balance_loss_clip": 1.05606437,
      "balance_loss_mlp": 1.00060308,
      "epoch": 0.2475801118258883,
      "flos": 21648352803840.0,
      "grad_norm": 2.738615425455071,
      "language_loss": 0.7517345,
      "learning_rate": 3.5237675054274893e-06,
      "loss": 0.77598369,
      "num_input_tokens_seen": 43978420,
      "step": 2059,
      "time_per_iteration": 2.68226957321167
    },
    {
      "auxiliary_loss_clip": 0.01211586,
      "auxiliary_loss_mlp": 0.01034383,
      "balance_loss_clip": 1.05666971,
      "balance_loss_mlp": 1.02439952,
      "epoch": 0.24770035471652738,
      "flos": 22674500542080.0,
      "grad_norm": 2.0460978399003915,
      "language_loss": 0.80099231,
      "learning_rate": 3.5232628391889584e-06,
      "loss": 0.823452,
      "num_input_tokens_seen": 43996710,
      "step": 2060,
      "time_per_iteration": 2.681941509246826
    },
    {
      "auxiliary_loss_clip": 0.0121389,
      "auxiliary_loss_mlp": 0.01029619,
      "balance_loss_clip": 1.05356586,
      "balance_loss_mlp": 1.02042818,
      "epoch": 0.2478205976071665,
      "flos": 22163814927360.0,
      "grad_norm": 2.566985896686098,
      "language_loss": 0.64424884,
      "learning_rate": 3.522757941872666e-06,
      "loss": 0.66668391,
      "num_input_tokens_seen": 44014865,
      "step": 2061,
      "time_per_iteration": 2.7312402725219727
    },
    {
      "auxiliary_loss_clip": 0.01210279,
      "auxiliary_loss_mlp": 0.01208762,
      "balance_loss_clip": 1.06153512,
      "balance_loss_mlp": 1.00070012,
      "epoch": 0.24794084049780557,
      "flos": 24973106555520.0,
      "grad_norm": 1.5920163328221677,
      "language_loss": 0.82704961,
      "learning_rate": 3.5222528135552042e-06,
      "loss": 0.85124004,
      "num_input_tokens_seen": 44036325,
      "step": 2062,
      "time_per_iteration": 2.6128690242767334
    },
    {
      "auxiliary_loss_clip": 0.01209809,
      "auxiliary_loss_mlp": 0.01036141,
      "balance_loss_clip": 1.05916226,
      "balance_loss_mlp": 1.02685499,
      "epoch": 0.24806108338844465,
      "flos": 18296379521280.0,
      "grad_norm": 2.092144381394687,
      "language_loss": 0.8028664,
      "learning_rate": 3.521747454313201e-06,
      "loss": 0.82532591,
      "num_input_tokens_seen": 44055005,
      "step": 2063,
      "time_per_iteration": 2.6703450679779053
    },
    {
      "auxiliary_loss_clip": 0.01205148,
      "auxiliary_loss_mlp": 0.0103005,
      "balance_loss_clip": 1.04909253,
      "balance_loss_mlp": 1.02060258,
      "epoch": 0.24818132627908374,
      "flos": 19282163351040.0,
      "grad_norm": 2.213934122885665,
      "language_loss": 0.67321485,
      "learning_rate": 3.521241864223319e-06,
      "loss": 0.69556683,
      "num_input_tokens_seen": 44073965,
      "step": 2064,
      "time_per_iteration": 2.654869794845581
    },
    {
      "auxiliary_loss_clip": 0.01116852,
      "auxiliary_loss_mlp": 0.0100582,
      "balance_loss_clip": 1.01876473,
      "balance_loss_mlp": 1.00396013,
      "epoch": 0.24830156916972285,
      "flos": 70285837881600.0,
      "grad_norm": 0.7849365162619191,
      "language_loss": 0.61970866,
      "learning_rate": 3.5207360433622552e-06,
      "loss": 0.64093542,
      "num_input_tokens_seen": 44135965,
      "step": 2065,
      "time_per_iteration": 3.2354371547698975
    },
    {
      "auxiliary_loss_clip": 0.01209562,
      "auxiliary_loss_mlp": 0.01037259,
      "balance_loss_clip": 1.05713511,
      "balance_loss_mlp": 1.02802062,
      "epoch": 0.24842181206036193,
      "flos": 40409128287360.0,
      "grad_norm": 2.174028342057879,
      "language_loss": 0.7445544,
      "learning_rate": 3.5202299918067437e-06,
      "loss": 0.76702261,
      "num_input_tokens_seen": 44159560,
      "step": 2066,
      "time_per_iteration": 2.803337812423706
    },
    {
      "auxiliary_loss_clip": 0.01208014,
      "auxiliary_loss_mlp": 0.0102968,
      "balance_loss_clip": 1.05688858,
      "balance_loss_mlp": 1.02081668,
      "epoch": 0.248542054951001,
      "flos": 20082432412800.0,
      "grad_norm": 2.330555546802818,
      "language_loss": 0.69384742,
      "learning_rate": 3.519723709633551e-06,
      "loss": 0.71622437,
      "num_input_tokens_seen": 44178320,
      "step": 2067,
      "time_per_iteration": 2.648911476135254
    },
    {
      "auxiliary_loss_clip": 0.01207383,
      "auxiliary_loss_mlp": 0.01030664,
      "balance_loss_clip": 1.05699372,
      "balance_loss_mlp": 1.02073359,
      "epoch": 0.24866229784164012,
      "flos": 23513948363520.0,
      "grad_norm": 1.798088105064413,
      "language_loss": 0.83362859,
      "learning_rate": 3.519217196919479e-06,
      "loss": 0.85600907,
      "num_input_tokens_seen": 44197305,
      "step": 2068,
      "time_per_iteration": 2.648120164871216
    },
    {
      "auxiliary_loss_clip": 0.01215233,
      "auxiliary_loss_mlp": 0.0103685,
      "balance_loss_clip": 1.05876851,
      "balance_loss_mlp": 1.02812362,
      "epoch": 0.2487825407322792,
      "flos": 19865101173120.0,
      "grad_norm": 2.763891462718375,
      "language_loss": 0.73515356,
      "learning_rate": 3.518710453741367e-06,
      "loss": 0.75767446,
      "num_input_tokens_seen": 44216505,
      "step": 2069,
      "time_per_iteration": 2.6837263107299805
    },
    {
      "auxiliary_loss_clip": 0.01204054,
      "auxiliary_loss_mlp": 0.01208862,
      "balance_loss_clip": 1.05348814,
      "balance_loss_mlp": 1.00064254,
      "epoch": 0.2489027836229183,
      "flos": 22017622573440.0,
      "grad_norm": 2.3663323549543094,
      "language_loss": 0.68118012,
      "learning_rate": 3.518203480176086e-06,
      "loss": 0.70530927,
      "num_input_tokens_seen": 44235435,
      "step": 2070,
      "time_per_iteration": 2.678635597229004
    },
    {
      "auxiliary_loss_clip": 0.01214874,
      "auxiliary_loss_mlp": 0.01040452,
      "balance_loss_clip": 1.04430819,
      "balance_loss_mlp": 1.03151703,
      "epoch": 0.2490230265135574,
      "flos": 23294354567040.0,
      "grad_norm": 2.082094755925137,
      "language_loss": 0.80792463,
      "learning_rate": 3.517696276300545e-06,
      "loss": 0.83047783,
      "num_input_tokens_seen": 44256975,
      "step": 2071,
      "time_per_iteration": 3.6440792083740234
    },
    {
      "auxiliary_loss_clip": 0.01211699,
      "auxiliary_loss_mlp": 0.01037087,
      "balance_loss_clip": 1.06117713,
      "balance_loss_mlp": 1.02705514,
      "epoch": 0.24914326940419648,
      "flos": 19826784339840.0,
      "grad_norm": 3.982625550882276,
      "language_loss": 0.69320917,
      "learning_rate": 3.517188842191685e-06,
      "loss": 0.71569705,
      "num_input_tokens_seen": 44275125,
      "step": 2072,
      "time_per_iteration": 2.6499271392822266
    },
    {
      "auxiliary_loss_clip": 0.01209392,
      "auxiliary_loss_mlp": 0.01032662,
      "balance_loss_clip": 1.0573107,
      "balance_loss_mlp": 1.0231967,
      "epoch": 0.24926351229483557,
      "flos": 20229271211520.0,
      "grad_norm": 2.138999147633682,
      "language_loss": 0.74231005,
      "learning_rate": 3.5166811779264837e-06,
      "loss": 0.76473057,
      "num_input_tokens_seen": 44295445,
      "step": 2073,
      "time_per_iteration": 3.589280843734741
    },
    {
      "auxiliary_loss_clip": 0.01207187,
      "auxiliary_loss_mlp": 0.01028387,
      "balance_loss_clip": 1.05770993,
      "balance_loss_mlp": 1.01863551,
      "epoch": 0.24938375518547465,
      "flos": 23294570048640.0,
      "grad_norm": 2.5679224373516987,
      "language_loss": 0.77801692,
      "learning_rate": 3.5161732835819545e-06,
      "loss": 0.80037272,
      "num_input_tokens_seen": 44314755,
      "step": 2074,
      "time_per_iteration": 2.6175363063812256
    },
    {
      "auxiliary_loss_clip": 0.0120913,
      "auxiliary_loss_mlp": 0.01032522,
      "balance_loss_clip": 1.0605644,
      "balance_loss_mlp": 1.0235275,
      "epoch": 0.24950399807611376,
      "flos": 17311673099520.0,
      "grad_norm": 2.2140940348064264,
      "language_loss": 0.83407092,
      "learning_rate": 3.515665159235143e-06,
      "loss": 0.85648751,
      "num_input_tokens_seen": 44333640,
      "step": 2075,
      "time_per_iteration": 2.6092000007629395
    },
    {
      "auxiliary_loss_clip": 0.01205243,
      "auxiliary_loss_mlp": 0.01027503,
      "balance_loss_clip": 1.0499754,
      "balance_loss_mlp": 1.01916409,
      "epoch": 0.24962424096675284,
      "flos": 19024863252480.0,
      "grad_norm": 1.7589821296417147,
      "language_loss": 0.74829483,
      "learning_rate": 3.5151568049631318e-06,
      "loss": 0.77062225,
      "num_input_tokens_seen": 44352355,
      "step": 2076,
      "time_per_iteration": 3.6081433296203613
    },
    {
      "auxiliary_loss_clip": 0.01208511,
      "auxiliary_loss_mlp": 0.01027686,
      "balance_loss_clip": 1.05853486,
      "balance_loss_mlp": 1.01809561,
      "epoch": 0.24974448385739192,
      "flos": 33398790710400.0,
      "grad_norm": 2.2481000697836935,
      "language_loss": 0.80149817,
      "learning_rate": 3.5146482208430385e-06,
      "loss": 0.82386017,
      "num_input_tokens_seen": 44374185,
      "step": 2077,
      "time_per_iteration": 2.7271223068237305
    },
    {
      "auxiliary_loss_clip": 0.0120757,
      "auxiliary_loss_mlp": 0.01034359,
      "balance_loss_clip": 1.04651737,
      "balance_loss_mlp": 1.02385688,
      "epoch": 0.24986472674803104,
      "flos": 30007279532160.0,
      "grad_norm": 2.8438708228982366,
      "language_loss": 0.6763922,
      "learning_rate": 3.514139406952014e-06,
      "loss": 0.69881141,
      "num_input_tokens_seen": 44396210,
      "step": 2078,
      "time_per_iteration": 3.74403715133667
    },
    {
      "auxiliary_loss_clip": 0.01208753,
      "auxiliary_loss_mlp": 0.01030381,
      "balance_loss_clip": 1.05797398,
      "balance_loss_mlp": 1.02123725,
      "epoch": 0.24998496963867012,
      "flos": 26613074833920.0,
      "grad_norm": 1.8937621928930766,
      "language_loss": 0.83584082,
      "learning_rate": 3.5136303633672454e-06,
      "loss": 0.85823208,
      "num_input_tokens_seen": 44416340,
      "step": 2079,
      "time_per_iteration": 2.7008824348449707
    },
    {
      "auxiliary_loss_clip": 0.01222198,
      "auxiliary_loss_mlp": 0.01208976,
      "balance_loss_clip": 1.05616784,
      "balance_loss_mlp": 1.00052035,
      "epoch": 0.25010521252930923,
      "flos": 23553989049600.0,
      "grad_norm": 1.8042167416007049,
      "language_loss": 0.74595815,
      "learning_rate": 3.5131210901659544e-06,
      "loss": 0.77026987,
      "num_input_tokens_seen": 44438095,
      "step": 2080,
      "time_per_iteration": 2.761042833328247
    },
    {
      "auxiliary_loss_clip": 0.01207168,
      "auxiliary_loss_mlp": 0.01028277,
      "balance_loss_clip": 1.05173659,
      "balance_loss_mlp": 1.01891959,
      "epoch": 0.2502254554199483,
      "flos": 23441193365760.0,
      "grad_norm": 2.490827027311509,
      "language_loss": 0.81762815,
      "learning_rate": 3.5126115874253967e-06,
      "loss": 0.83998263,
      "num_input_tokens_seen": 44457650,
      "step": 2081,
      "time_per_iteration": 2.6860992908477783
    },
    {
      "auxiliary_loss_clip": 0.01215551,
      "auxiliary_loss_mlp": 0.01032008,
      "balance_loss_clip": 1.05682755,
      "balance_loss_mlp": 1.02213728,
      "epoch": 0.2503456983105874,
      "flos": 28761681651840.0,
      "grad_norm": 2.7998653414319166,
      "language_loss": 0.80679655,
      "learning_rate": 3.5121018552228644e-06,
      "loss": 0.82927215,
      "num_input_tokens_seen": 44476155,
      "step": 2082,
      "time_per_iteration": 2.747042179107666
    },
    {
      "auxiliary_loss_clip": 0.01215195,
      "auxiliary_loss_mlp": 0.01029904,
      "balance_loss_clip": 1.05424881,
      "balance_loss_mlp": 1.02047443,
      "epoch": 0.2504659412012265,
      "flos": 18770256673920.0,
      "grad_norm": 3.808655607295656,
      "language_loss": 0.7661407,
      "learning_rate": 3.5115918936356827e-06,
      "loss": 0.78859168,
      "num_input_tokens_seen": 44492910,
      "step": 2083,
      "time_per_iteration": 2.7080540657043457
    },
    {
      "auxiliary_loss_clip": 0.01197147,
      "auxiliary_loss_mlp": 0.0103464,
      "balance_loss_clip": 1.05290401,
      "balance_loss_mlp": 1.02531791,
      "epoch": 0.25058618409186556,
      "flos": 16873383346560.0,
      "grad_norm": 1.917826842456176,
      "language_loss": 0.79100144,
      "learning_rate": 3.5110817027412123e-06,
      "loss": 0.81331933,
      "num_input_tokens_seen": 44512000,
      "step": 2084,
      "time_per_iteration": 2.750976324081421
    },
    {
      "auxiliary_loss_clip": 0.01205899,
      "auxiliary_loss_mlp": 0.0102916,
      "balance_loss_clip": 1.04949343,
      "balance_loss_mlp": 1.02023077,
      "epoch": 0.25070642698250467,
      "flos": 24425540651520.0,
      "grad_norm": 2.5600458751901694,
      "language_loss": 0.68677175,
      "learning_rate": 3.5105712826168493e-06,
      "loss": 0.70912236,
      "num_input_tokens_seen": 44531650,
      "step": 2085,
      "time_per_iteration": 2.6743810176849365
    },
    {
      "auxiliary_loss_clip": 0.01208517,
      "auxiliary_loss_mlp": 0.01207941,
      "balance_loss_clip": 1.05550873,
      "balance_loss_mlp": 1.0005461,
      "epoch": 0.2508266698731437,
      "flos": 20260944028800.0,
      "grad_norm": 5.142424909601349,
      "language_loss": 0.7047267,
      "learning_rate": 3.5100606333400235e-06,
      "loss": 0.72889125,
      "num_input_tokens_seen": 44548785,
      "step": 2086,
      "time_per_iteration": 2.664149761199951
    },
    {
      "auxiliary_loss_clip": 0.01224289,
      "auxiliary_loss_mlp": 0.01032612,
      "balance_loss_clip": 1.05758512,
      "balance_loss_mlp": 1.02217507,
      "epoch": 0.25094691276378284,
      "flos": 19245318975360.0,
      "grad_norm": 1.9618631977321386,
      "language_loss": 0.76896942,
      "learning_rate": 3.5095497549882006e-06,
      "loss": 0.79153836,
      "num_input_tokens_seen": 44567230,
      "step": 2087,
      "time_per_iteration": 2.6843173503875732
    },
    {
      "auxiliary_loss_clip": 0.01215557,
      "auxiliary_loss_mlp": 0.01033102,
      "balance_loss_clip": 1.06207454,
      "balance_loss_mlp": 1.02317786,
      "epoch": 0.25106715565442195,
      "flos": 26943237671040.0,
      "grad_norm": 2.300400658112508,
      "language_loss": 0.7312839,
      "learning_rate": 3.50903864763888e-06,
      "loss": 0.75377041,
      "num_input_tokens_seen": 44588020,
      "step": 2088,
      "time_per_iteration": 2.702531576156616
    },
    {
      "auxiliary_loss_clip": 0.01216042,
      "auxiliary_loss_mlp": 0.01030234,
      "balance_loss_clip": 1.05785036,
      "balance_loss_mlp": 1.02063215,
      "epoch": 0.251187398545061,
      "flos": 48359570572800.0,
      "grad_norm": 2.032626791898107,
      "language_loss": 0.76494831,
      "learning_rate": 3.5085273113695965e-06,
      "loss": 0.78741109,
      "num_input_tokens_seen": 44612590,
      "step": 2089,
      "time_per_iteration": 2.8843631744384766
    },
    {
      "auxiliary_loss_clip": 0.01210819,
      "auxiliary_loss_mlp": 0.01034873,
      "balance_loss_clip": 1.05998242,
      "balance_loss_mlp": 1.02495515,
      "epoch": 0.2513076414357001,
      "flos": 27016100409600.0,
      "grad_norm": 2.1048538676722943,
      "language_loss": 0.78398693,
      "learning_rate": 3.508015746257919e-06,
      "loss": 0.80644387,
      "num_input_tokens_seen": 44631630,
      "step": 2090,
      "time_per_iteration": 2.6453025341033936
    },
    {
      "auxiliary_loss_clip": 0.0121766,
      "auxiliary_loss_mlp": 0.01034655,
      "balance_loss_clip": 1.05491853,
      "balance_loss_mlp": 1.02456951,
      "epoch": 0.2514278843263392,
      "flos": 19463619882240.0,
      "grad_norm": 2.5169471326591495,
      "language_loss": 0.83334363,
      "learning_rate": 3.5075039523814518e-06,
      "loss": 0.85586679,
      "num_input_tokens_seen": 44650820,
      "step": 2091,
      "time_per_iteration": 2.716305732727051
    },
    {
      "auxiliary_loss_clip": 0.01218659,
      "auxiliary_loss_mlp": 0.01033742,
      "balance_loss_clip": 1.0586307,
      "balance_loss_mlp": 1.02324545,
      "epoch": 0.2515481272169783,
      "flos": 16866092885760.0,
      "grad_norm": 2.3878818738536807,
      "language_loss": 0.81823063,
      "learning_rate": 3.506991929817834e-06,
      "loss": 0.84075463,
      "num_input_tokens_seen": 44667540,
      "step": 2092,
      "time_per_iteration": 2.614140272140503
    },
    {
      "auxiliary_loss_clip": 0.01207262,
      "auxiliary_loss_mlp": 0.01030918,
      "balance_loss_clip": 1.06093109,
      "balance_loss_mlp": 1.02198339,
      "epoch": 0.2516683701076174,
      "flos": 23732464752000.0,
      "grad_norm": 2.0839293801677305,
      "language_loss": 0.82665008,
      "learning_rate": 3.506479678644738e-06,
      "loss": 0.84903193,
      "num_input_tokens_seen": 44687935,
      "step": 2093,
      "time_per_iteration": 2.629199743270874
    },
    {
      "auxiliary_loss_clip": 0.01211701,
      "auxiliary_loss_mlp": 0.01027681,
      "balance_loss_clip": 1.05384958,
      "balance_loss_mlp": 1.01847208,
      "epoch": 0.2517886129982565,
      "flos": 27635954434560.0,
      "grad_norm": 3.2233919336295114,
      "language_loss": 0.74186307,
      "learning_rate": 3.505967198939873e-06,
      "loss": 0.76425689,
      "num_input_tokens_seen": 44704975,
      "step": 2094,
      "time_per_iteration": 2.7259409427642822
    },
    {
      "auxiliary_loss_clip": 0.01209579,
      "auxiliary_loss_mlp": 0.01027387,
      "balance_loss_clip": 1.05344748,
      "balance_loss_mlp": 1.01780272,
      "epoch": 0.25190885588889556,
      "flos": 38104596529920.0,
      "grad_norm": 2.0724342543770575,
      "language_loss": 0.78360176,
      "learning_rate": 3.5054544907809813e-06,
      "loss": 0.80597138,
      "num_input_tokens_seen": 44725475,
      "step": 2095,
      "time_per_iteration": 2.8131649494171143
    },
    {
      "auxiliary_loss_clip": 0.01211606,
      "auxiliary_loss_mlp": 0.01209301,
      "balance_loss_clip": 1.0579102,
      "balance_loss_mlp": 1.0006597,
      "epoch": 0.25202909877953467,
      "flos": 22269894768000.0,
      "grad_norm": 2.0887592265131385,
      "language_loss": 0.80510986,
      "learning_rate": 3.50494155424584e-06,
      "loss": 0.82931894,
      "num_input_tokens_seen": 44744380,
      "step": 2096,
      "time_per_iteration": 2.6720950603485107
    },
    {
      "auxiliary_loss_clip": 0.01216089,
      "auxiliary_loss_mlp": 0.01032903,
      "balance_loss_clip": 1.05950642,
      "balance_loss_mlp": 1.02278197,
      "epoch": 0.2521493416701738,
      "flos": 21761759018880.0,
      "grad_norm": 2.7574501410909784,
      "language_loss": 0.8305465,
      "learning_rate": 3.504428389412262e-06,
      "loss": 0.8530364,
      "num_input_tokens_seen": 44765190,
      "step": 2097,
      "time_per_iteration": 3.768242835998535
    },
    {
      "auxiliary_loss_clip": 0.01207631,
      "auxiliary_loss_mlp": 0.01036107,
      "balance_loss_clip": 1.05586541,
      "balance_loss_mlp": 1.0268743,
      "epoch": 0.25226958456081283,
      "flos": 27746738956800.0,
      "grad_norm": 2.2984211571796798,
      "language_loss": 0.73145604,
      "learning_rate": 3.5039149963580927e-06,
      "loss": 0.75389338,
      "num_input_tokens_seen": 44785210,
      "step": 2098,
      "time_per_iteration": 2.7937214374542236
    },
    {
      "auxiliary_loss_clip": 0.01210409,
      "auxiliary_loss_mlp": 0.01033806,
      "balance_loss_clip": 1.05965757,
      "balance_loss_mlp": 1.02458525,
      "epoch": 0.25238982745145194,
      "flos": 30732171903360.0,
      "grad_norm": 2.82771018394731,
      "language_loss": 0.70104152,
      "learning_rate": 3.503401375161215e-06,
      "loss": 0.72348362,
      "num_input_tokens_seen": 44804955,
      "step": 2099,
      "time_per_iteration": 3.6541504859924316
    },
    {
      "auxiliary_loss_clip": 0.01205168,
      "auxiliary_loss_mlp": 0.01028669,
      "balance_loss_clip": 1.05742693,
      "balance_loss_mlp": 1.01974583,
      "epoch": 0.252510070342091,
      "flos": 20266331068800.0,
      "grad_norm": 4.487569033673038,
      "language_loss": 0.83527714,
      "learning_rate": 3.502887525899544e-06,
      "loss": 0.85761553,
      "num_input_tokens_seen": 44823935,
      "step": 2100,
      "time_per_iteration": 2.623051166534424
    },
    {
      "auxiliary_loss_clip": 0.01212739,
      "auxiliary_loss_mlp": 0.01028807,
      "balance_loss_clip": 1.05619192,
      "balance_loss_mlp": 1.01893008,
      "epoch": 0.2526303132327301,
      "flos": 22747399194240.0,
      "grad_norm": 1.7979861273553108,
      "language_loss": 0.83183354,
      "learning_rate": 3.50237344865103e-06,
      "loss": 0.85424894,
      "num_input_tokens_seen": 44844935,
      "step": 2101,
      "time_per_iteration": 2.6616060733795166
    },
    {
      "auxiliary_loss_clip": 0.01210981,
      "auxiliary_loss_mlp": 0.01037275,
      "balance_loss_clip": 1.0600276,
      "balance_loss_mlp": 1.0278697,
      "epoch": 0.2527505561233692,
      "flos": 30263466309120.0,
      "grad_norm": 2.0146694900582123,
      "language_loss": 0.76487416,
      "learning_rate": 3.501859143493658e-06,
      "loss": 0.78735673,
      "num_input_tokens_seen": 44865565,
      "step": 2102,
      "time_per_iteration": 2.734915018081665
    },
    {
      "auxiliary_loss_clip": 0.01105767,
      "auxiliary_loss_mlp": 0.01007363,
      "balance_loss_clip": 1.02248299,
      "balance_loss_mlp": 1.00581348,
      "epoch": 0.2528707990140083,
      "flos": 58492917164160.0,
      "grad_norm": 0.9185394152840876,
      "language_loss": 0.60505617,
      "learning_rate": 3.5013446105054488e-06,
      "loss": 0.62618744,
      "num_input_tokens_seen": 44918485,
      "step": 2103,
      "time_per_iteration": 3.810565710067749
    },
    {
      "auxiliary_loss_clip": 0.01197679,
      "auxiliary_loss_mlp": 0.01036012,
      "balance_loss_clip": 1.05125952,
      "balance_loss_mlp": 1.02642107,
      "epoch": 0.2529910419046474,
      "flos": 24645134448000.0,
      "grad_norm": 2.475674559765937,
      "language_loss": 0.75209183,
      "learning_rate": 3.5008298497644555e-06,
      "loss": 0.77442873,
      "num_input_tokens_seen": 44937530,
      "step": 2104,
      "time_per_iteration": 2.652611494064331
    },
    {
      "auxiliary_loss_clip": 0.01217655,
      "auxiliary_loss_mlp": 0.01034272,
      "balance_loss_clip": 1.05638671,
      "balance_loss_mlp": 1.02405596,
      "epoch": 0.2531112847952865,
      "flos": 23842135952640.0,
      "grad_norm": 1.748254629206495,
      "language_loss": 0.87910289,
      "learning_rate": 3.500314861348767e-06,
      "loss": 0.90162218,
      "num_input_tokens_seen": 44958165,
      "step": 2105,
      "time_per_iteration": 3.6515917778015137
    },
    {
      "auxiliary_loss_clip": 0.01209084,
      "auxiliary_loss_mlp": 0.01037237,
      "balance_loss_clip": 1.05678701,
      "balance_loss_mlp": 1.02808738,
      "epoch": 0.25323152768592555,
      "flos": 16143822207360.0,
      "grad_norm": 2.3549564685177673,
      "language_loss": 0.77450389,
      "learning_rate": 3.499799645336507e-06,
      "loss": 0.79696703,
      "num_input_tokens_seen": 44975060,
      "step": 2106,
      "time_per_iteration": 2.7296385765075684
    },
    {
      "auxiliary_loss_clip": 0.01216063,
      "auxiliary_loss_mlp": 0.01032237,
      "balance_loss_clip": 1.06241608,
      "balance_loss_mlp": 1.0233618,
      "epoch": 0.25335177057656466,
      "flos": 28405161210240.0,
      "grad_norm": 1.701089917653615,
      "language_loss": 0.86982697,
      "learning_rate": 3.4992842018058336e-06,
      "loss": 0.89230996,
      "num_input_tokens_seen": 44997960,
      "step": 2107,
      "time_per_iteration": 2.75319504737854
    },
    {
      "auxiliary_loss_clip": 0.01222033,
      "auxiliary_loss_mlp": 0.0102835,
      "balance_loss_clip": 1.05676281,
      "balance_loss_mlp": 1.01910508,
      "epoch": 0.25347201346720377,
      "flos": 18799666934400.0,
      "grad_norm": 2.4357006665543954,
      "language_loss": 0.8868655,
      "learning_rate": 3.4987685308349384e-06,
      "loss": 0.90936929,
      "num_input_tokens_seen": 45015690,
      "step": 2108,
      "time_per_iteration": 2.698396682739258
    },
    {
      "auxiliary_loss_clip": 0.01213052,
      "auxiliary_loss_mlp": 0.01034162,
      "balance_loss_clip": 1.05100548,
      "balance_loss_mlp": 1.02484608,
      "epoch": 0.2535922563578428,
      "flos": 15815490963840.0,
      "grad_norm": 2.051963945854046,
      "language_loss": 0.61431289,
      "learning_rate": 3.4982526325020497e-06,
      "loss": 0.63678503,
      "num_input_tokens_seen": 45032660,
      "step": 2109,
      "time_per_iteration": 2.67790150642395
    },
    {
      "auxiliary_loss_clip": 0.01215764,
      "auxiliary_loss_mlp": 0.01032666,
      "balance_loss_clip": 1.05749154,
      "balance_loss_mlp": 1.02275348,
      "epoch": 0.25371249924848194,
      "flos": 16318922031360.0,
      "grad_norm": 2.757797840459074,
      "language_loss": 0.81996763,
      "learning_rate": 3.4977365068854273e-06,
      "loss": 0.84245193,
      "num_input_tokens_seen": 45048280,
      "step": 2110,
      "time_per_iteration": 2.624563217163086
    },
    {
      "auxiliary_loss_clip": 0.01206961,
      "auxiliary_loss_mlp": 0.01033422,
      "balance_loss_clip": 1.05515742,
      "balance_loss_mlp": 1.02333105,
      "epoch": 0.25383274213912105,
      "flos": 21761615364480.0,
      "grad_norm": 2.0046759463231063,
      "language_loss": 0.7368418,
      "learning_rate": 3.4972201540633676e-06,
      "loss": 0.75924563,
      "num_input_tokens_seen": 45067635,
      "step": 2111,
      "time_per_iteration": 2.6719982624053955
    },
    {
      "auxiliary_loss_clip": 0.01205066,
      "auxiliary_loss_mlp": 0.01032074,
      "balance_loss_clip": 1.05609143,
      "balance_loss_mlp": 1.02172661,
      "epoch": 0.2539529850297601,
      "flos": 21396870708480.0,
      "grad_norm": 2.028797021535629,
      "language_loss": 0.85346258,
      "learning_rate": 3.4967035741142008e-06,
      "loss": 0.87583399,
      "num_input_tokens_seen": 45086455,
      "step": 2112,
      "time_per_iteration": 2.7250025272369385
    },
    {
      "auxiliary_loss_clip": 0.01205164,
      "auxiliary_loss_mlp": 0.01029102,
      "balance_loss_clip": 1.06151724,
      "balance_loss_mlp": 1.02034044,
      "epoch": 0.2540732279203992,
      "flos": 25228467319680.0,
      "grad_norm": 3.5351809390381326,
      "language_loss": 0.82243103,
      "learning_rate": 3.4961867671162917e-06,
      "loss": 0.84477371,
      "num_input_tokens_seen": 45106385,
      "step": 2113,
      "time_per_iteration": 2.6828949451446533
    },
    {
      "auxiliary_loss_clip": 0.0121426,
      "auxiliary_loss_mlp": 0.01028958,
      "balance_loss_clip": 1.0620476,
      "balance_loss_mlp": 1.01859272,
      "epoch": 0.2541934708110383,
      "flos": 19427386037760.0,
      "grad_norm": 3.2033878468103305,
      "language_loss": 0.77130973,
      "learning_rate": 3.4956697331480402e-06,
      "loss": 0.79374188,
      "num_input_tokens_seen": 45124955,
      "step": 2114,
      "time_per_iteration": 2.678701162338257
    },
    {
      "auxiliary_loss_clip": 0.01222207,
      "auxiliary_loss_mlp": 0.01031635,
      "balance_loss_clip": 1.05527091,
      "balance_loss_mlp": 1.02154422,
      "epoch": 0.2543137137016774,
      "flos": 23949436855680.0,
      "grad_norm": 1.602439677885593,
      "language_loss": 0.800951,
      "learning_rate": 3.495152472287879e-06,
      "loss": 0.82348949,
      "num_input_tokens_seen": 45145665,
      "step": 2115,
      "time_per_iteration": 2.6855385303497314
    },
    {
      "auxiliary_loss_clip": 0.01214843,
      "auxiliary_loss_mlp": 0.0103213,
      "balance_loss_clip": 1.0565033,
      "balance_loss_mlp": 1.02327907,
      "epoch": 0.2544339565923165,
      "flos": 25593283802880.0,
      "grad_norm": 1.9563410443131282,
      "language_loss": 0.74241537,
      "learning_rate": 3.4946349846142766e-06,
      "loss": 0.76488507,
      "num_input_tokens_seen": 45164805,
      "step": 2116,
      "time_per_iteration": 2.7307193279266357
    },
    {
      "auxiliary_loss_clip": 0.01209697,
      "auxiliary_loss_mlp": 0.01040569,
      "balance_loss_clip": 1.06059599,
      "balance_loss_mlp": 1.03124666,
      "epoch": 0.25455419948295555,
      "flos": 21689470897920.0,
      "grad_norm": 1.9816692373573013,
      "language_loss": 0.75882411,
      "learning_rate": 3.4941172702057353e-06,
      "loss": 0.78132677,
      "num_input_tokens_seen": 45184865,
      "step": 2117,
      "time_per_iteration": 2.6210033893585205
    },
    {
      "auxiliary_loss_clip": 0.01212403,
      "auxiliary_loss_mlp": 0.01033233,
      "balance_loss_clip": 1.058478,
      "balance_loss_mlp": 1.02354145,
      "epoch": 0.25467444237359466,
      "flos": 26250341339520.0,
      "grad_norm": 1.7482997462420278,
      "language_loss": 0.80443621,
      "learning_rate": 3.4935993291407924e-06,
      "loss": 0.82689261,
      "num_input_tokens_seen": 45203690,
      "step": 2118,
      "time_per_iteration": 2.7178328037261963
    },
    {
      "auxiliary_loss_clip": 0.01210096,
      "auxiliary_loss_mlp": 0.01032881,
      "balance_loss_clip": 1.05601931,
      "balance_loss_mlp": 1.02249813,
      "epoch": 0.25479468526423377,
      "flos": 26979686997120.0,
      "grad_norm": 2.423632653227952,
      "language_loss": 0.71176779,
      "learning_rate": 3.4930811614980183e-06,
      "loss": 0.73419756,
      "num_input_tokens_seen": 45225385,
      "step": 2119,
      "time_per_iteration": 2.70271372795105
    },
    {
      "auxiliary_loss_clip": 0.01205581,
      "auxiliary_loss_mlp": 0.01037133,
      "balance_loss_clip": 1.05764103,
      "balance_loss_mlp": 1.02783489,
      "epoch": 0.2549149281548728,
      "flos": 23475811098240.0,
      "grad_norm": 2.477884648123969,
      "language_loss": 0.79436034,
      "learning_rate": 3.4925627673560198e-06,
      "loss": 0.81678748,
      "num_input_tokens_seen": 45246045,
      "step": 2120,
      "time_per_iteration": 2.7244415283203125
    },
    {
      "auxiliary_loss_clip": 0.01213538,
      "auxiliary_loss_mlp": 0.01036717,
      "balance_loss_clip": 1.05496454,
      "balance_loss_mlp": 1.0282414,
      "epoch": 0.25503517104551193,
      "flos": 25812302981760.0,
      "grad_norm": 1.7050488041644467,
      "language_loss": 0.88363791,
      "learning_rate": 3.4920441467934357e-06,
      "loss": 0.90614045,
      "num_input_tokens_seen": 45266560,
      "step": 2121,
      "time_per_iteration": 2.73907470703125
    },
    {
      "auxiliary_loss_clip": 0.01207236,
      "auxiliary_loss_mlp": 0.01037647,
      "balance_loss_clip": 1.05500555,
      "balance_loss_mlp": 1.0287658,
      "epoch": 0.25515541393615104,
      "flos": 26645106787200.0,
      "grad_norm": 2.243364692782511,
      "language_loss": 0.83093739,
      "learning_rate": 3.491525299888941e-06,
      "loss": 0.85338622,
      "num_input_tokens_seen": 45285405,
      "step": 2122,
      "time_per_iteration": 2.691615104675293
    },
    {
      "auxiliary_loss_clip": 0.01111417,
      "auxiliary_loss_mlp": 0.01200404,
      "balance_loss_clip": 1.02991176,
      "balance_loss_mlp": 1.00082648,
      "epoch": 0.2552756568267901,
      "flos": 65955945847680.0,
      "grad_norm": 0.8797728691720447,
      "language_loss": 0.62685907,
      "learning_rate": 3.491006226721244e-06,
      "loss": 0.64997733,
      "num_input_tokens_seen": 45349615,
      "step": 2123,
      "time_per_iteration": 3.2517008781433105
    },
    {
      "auxiliary_loss_clip": 0.01216391,
      "auxiliary_loss_mlp": 0.01208777,
      "balance_loss_clip": 1.06020474,
      "balance_loss_mlp": 1.00071311,
      "epoch": 0.2553958997174292,
      "flos": 17931096161280.0,
      "grad_norm": 2.4246401650701808,
      "language_loss": 0.77654076,
      "learning_rate": 3.4904869273690882e-06,
      "loss": 0.8007924,
      "num_input_tokens_seen": 45367505,
      "step": 2124,
      "time_per_iteration": 3.615729331970215
    },
    {
      "auxiliary_loss_clip": 0.01213976,
      "auxiliary_loss_mlp": 0.01027506,
      "balance_loss_clip": 1.05885041,
      "balance_loss_mlp": 1.01844621,
      "epoch": 0.2555161426080683,
      "flos": 23367791923200.0,
      "grad_norm": 1.871062358721282,
      "language_loss": 0.89033782,
      "learning_rate": 3.489967401911251e-06,
      "loss": 0.91275263,
      "num_input_tokens_seen": 45386805,
      "step": 2125,
      "time_per_iteration": 2.7297167778015137
    },
    {
      "auxiliary_loss_clip": 0.01216092,
      "auxiliary_loss_mlp": 0.01033891,
      "balance_loss_clip": 1.06348848,
      "balance_loss_mlp": 1.02282238,
      "epoch": 0.2556363854987074,
      "flos": 40625130723840.0,
      "grad_norm": 1.584429038330773,
      "language_loss": 0.69395071,
      "learning_rate": 3.4894476504265428e-06,
      "loss": 0.71645057,
      "num_input_tokens_seen": 45411045,
      "step": 2126,
      "time_per_iteration": 3.730405807495117
    },
    {
      "auxiliary_loss_clip": 0.01107676,
      "auxiliary_loss_mlp": 0.01014525,
      "balance_loss_clip": 1.02332044,
      "balance_loss_mlp": 1.01283216,
      "epoch": 0.2557566283893465,
      "flos": 68019443389440.0,
      "grad_norm": 0.7342616476573911,
      "language_loss": 0.5439769,
      "learning_rate": 3.4889276729938104e-06,
      "loss": 0.5651989,
      "num_input_tokens_seen": 45469575,
      "step": 2127,
      "time_per_iteration": 3.0705983638763428
    },
    {
      "auxiliary_loss_clip": 0.01206719,
      "auxiliary_loss_mlp": 0.01027877,
      "balance_loss_clip": 1.05461836,
      "balance_loss_mlp": 1.01748848,
      "epoch": 0.2558768712799856,
      "flos": 22635645004800.0,
      "grad_norm": 2.2673241618432654,
      "language_loss": 0.80514663,
      "learning_rate": 3.488407469691934e-06,
      "loss": 0.82749259,
      "num_input_tokens_seen": 45490270,
      "step": 2128,
      "time_per_iteration": 2.696648359298706
    },
    {
      "auxiliary_loss_clip": 0.01212221,
      "auxiliary_loss_mlp": 0.01034446,
      "balance_loss_clip": 1.05534112,
      "balance_loss_mlp": 1.02459896,
      "epoch": 0.25599711417062465,
      "flos": 26396354125440.0,
      "grad_norm": 2.2374756257197483,
      "language_loss": 0.81241155,
      "learning_rate": 3.487887040599828e-06,
      "loss": 0.83487827,
      "num_input_tokens_seen": 45510070,
      "step": 2129,
      "time_per_iteration": 2.7024269104003906
    },
    {
      "auxiliary_loss_clip": 0.01213446,
      "auxiliary_loss_mlp": 0.01033198,
      "balance_loss_clip": 1.06287324,
      "balance_loss_mlp": 1.02303505,
      "epoch": 0.25611735706126376,
      "flos": 22852042490880.0,
      "grad_norm": 2.66376355833499,
      "language_loss": 0.75696927,
      "learning_rate": 3.4873663857964407e-06,
      "loss": 0.77943575,
      "num_input_tokens_seen": 45527285,
      "step": 2130,
      "time_per_iteration": 3.552058219909668
    },
    {
      "auxiliary_loss_clip": 0.01214918,
      "auxiliary_loss_mlp": 0.01037077,
      "balance_loss_clip": 1.05465877,
      "balance_loss_mlp": 1.02718306,
      "epoch": 0.2562375999519028,
      "flos": 23367863750400.0,
      "grad_norm": 2.0922908710396815,
      "language_loss": 0.66639984,
      "learning_rate": 3.4868455053607556e-06,
      "loss": 0.68891978,
      "num_input_tokens_seen": 45546900,
      "step": 2131,
      "time_per_iteration": 2.8546817302703857
    },
    {
      "auxiliary_loss_clip": 0.01215655,
      "auxiliary_loss_mlp": 0.01037157,
      "balance_loss_clip": 1.05721688,
      "balance_loss_mlp": 1.02702987,
      "epoch": 0.2563578428425419,
      "flos": 22856962654080.0,
      "grad_norm": 2.5028573134432586,
      "language_loss": 0.7177242,
      "learning_rate": 3.486324399371789e-06,
      "loss": 0.74025232,
      "num_input_tokens_seen": 45566200,
      "step": 2132,
      "time_per_iteration": 3.579486131668091
    },
    {
      "auxiliary_loss_clip": 0.01212196,
      "auxiliary_loss_mlp": 0.01033475,
      "balance_loss_clip": 1.05623472,
      "balance_loss_mlp": 1.02447486,
      "epoch": 0.25647808573318104,
      "flos": 21653883498240.0,
      "grad_norm": 2.0751299850850056,
      "language_loss": 0.78401589,
      "learning_rate": 3.485803067908593e-06,
      "loss": 0.80647266,
      "num_input_tokens_seen": 45585710,
      "step": 2133,
      "time_per_iteration": 2.645498514175415
    },
    {
      "auxiliary_loss_clip": 0.01194068,
      "auxiliary_loss_mlp": 0.0103359,
      "balance_loss_clip": 1.04373384,
      "balance_loss_mlp": 1.02398193,
      "epoch": 0.2565983286238201,
      "flos": 33730569659520.0,
      "grad_norm": 1.9460735929881967,
      "language_loss": 0.79093957,
      "learning_rate": 3.485281511050253e-06,
      "loss": 0.81321621,
      "num_input_tokens_seen": 45607845,
      "step": 2134,
      "time_per_iteration": 2.898310661315918
    },
    {
      "auxiliary_loss_clip": 0.01214482,
      "auxiliary_loss_mlp": 0.01034556,
      "balance_loss_clip": 1.05749273,
      "balance_loss_mlp": 1.0246737,
      "epoch": 0.2567185715144592,
      "flos": 16216002587520.0,
      "grad_norm": 5.877102238108793,
      "language_loss": 0.89844668,
      "learning_rate": 3.484759728875889e-06,
      "loss": 0.92093706,
      "num_input_tokens_seen": 45623210,
      "step": 2135,
      "time_per_iteration": 2.614975690841675
    },
    {
      "auxiliary_loss_clip": 0.01202604,
      "auxiliary_loss_mlp": 0.01036888,
      "balance_loss_clip": 1.05096221,
      "balance_loss_mlp": 1.02754176,
      "epoch": 0.2568388144050983,
      "flos": 17458475984640.0,
      "grad_norm": 1.706972864853393,
      "language_loss": 0.80715793,
      "learning_rate": 3.4842377214646543e-06,
      "loss": 0.82955283,
      "num_input_tokens_seen": 45641505,
      "step": 2136,
      "time_per_iteration": 2.747943639755249
    },
    {
      "auxiliary_loss_clip": 0.01208302,
      "auxiliary_loss_mlp": 0.01032909,
      "balance_loss_clip": 1.06006098,
      "balance_loss_mlp": 1.02378333,
      "epoch": 0.25695905729573737,
      "flos": 20887442069760.0,
      "grad_norm": 2.636237516284182,
      "language_loss": 0.6702013,
      "learning_rate": 3.483715488895737e-06,
      "loss": 0.69261342,
      "num_input_tokens_seen": 45661835,
      "step": 2137,
      "time_per_iteration": 2.658067464828491
    },
    {
      "auxiliary_loss_clip": 0.01213786,
      "auxiliary_loss_mlp": 0.01026919,
      "balance_loss_clip": 1.04990923,
      "balance_loss_mlp": 1.01732814,
      "epoch": 0.2570793001863765,
      "flos": 24717278914560.0,
      "grad_norm": 1.7318913616040799,
      "language_loss": 0.78096926,
      "learning_rate": 3.48319303124836e-06,
      "loss": 0.80337632,
      "num_input_tokens_seen": 45682215,
      "step": 2138,
      "time_per_iteration": 2.9243862628936768
    },
    {
      "auxiliary_loss_clip": 0.01208197,
      "auxiliary_loss_mlp": 0.01032069,
      "balance_loss_clip": 1.05730844,
      "balance_loss_mlp": 1.02271676,
      "epoch": 0.2571995430770156,
      "flos": 26906896085760.0,
      "grad_norm": 10.508583666631168,
      "language_loss": 0.67207885,
      "learning_rate": 3.4826703486017798e-06,
      "loss": 0.69448155,
      "num_input_tokens_seen": 45701840,
      "step": 2139,
      "time_per_iteration": 2.6688766479492188
    },
    {
      "auxiliary_loss_clip": 0.01210421,
      "auxiliary_loss_mlp": 0.01029719,
      "balance_loss_clip": 1.05979323,
      "balance_loss_mlp": 1.02046847,
      "epoch": 0.25731978596765465,
      "flos": 19792561656960.0,
      "grad_norm": 1.8650509613815134,
      "language_loss": 0.76439226,
      "learning_rate": 3.4821474410352867e-06,
      "loss": 0.78679371,
      "num_input_tokens_seen": 45720500,
      "step": 2140,
      "time_per_iteration": 2.6672115325927734
    },
    {
      "auxiliary_loss_clip": 0.01125274,
      "auxiliary_loss_mlp": 0.01002708,
      "balance_loss_clip": 1.02981138,
      "balance_loss_mlp": 1.00083673,
      "epoch": 0.25744002885829376,
      "flos": 70564970471040.0,
      "grad_norm": 0.9058598596535918,
      "language_loss": 0.62611324,
      "learning_rate": 3.481624308628205e-06,
      "loss": 0.64739305,
      "num_input_tokens_seen": 45781870,
      "step": 2141,
      "time_per_iteration": 3.368253707885742
    },
    {
      "auxiliary_loss_clip": 0.01210687,
      "auxiliary_loss_mlp": 0.01032947,
      "balance_loss_clip": 1.05481744,
      "balance_loss_mlp": 1.02325559,
      "epoch": 0.25756027174893287,
      "flos": 18038181582720.0,
      "grad_norm": 3.178058075795774,
      "language_loss": 1.00444531,
      "learning_rate": 3.481100951459893e-06,
      "loss": 1.02688169,
      "num_input_tokens_seen": 45794890,
      "step": 2142,
      "time_per_iteration": 2.729968786239624
    },
    {
      "auxiliary_loss_clip": 0.01207436,
      "auxiliary_loss_mlp": 0.01029241,
      "balance_loss_clip": 1.05707574,
      "balance_loss_mlp": 1.01955557,
      "epoch": 0.2576805146395719,
      "flos": 22674069578880.0,
      "grad_norm": 2.368912175007772,
      "language_loss": 0.78673786,
      "learning_rate": 3.4805773696097453e-06,
      "loss": 0.80910468,
      "num_input_tokens_seen": 45815780,
      "step": 2143,
      "time_per_iteration": 2.6850571632385254
    },
    {
      "auxiliary_loss_clip": 0.01209694,
      "auxiliary_loss_mlp": 0.01029457,
      "balance_loss_clip": 1.06009412,
      "balance_loss_mlp": 1.01998568,
      "epoch": 0.25780075753021103,
      "flos": 16472225278080.0,
      "grad_norm": 2.1576426009557963,
      "language_loss": 0.87806052,
      "learning_rate": 3.4800535631571874e-06,
      "loss": 0.90045202,
      "num_input_tokens_seen": 45831310,
      "step": 2144,
      "time_per_iteration": 2.6153008937835693
    },
    {
      "auxiliary_loss_clip": 0.01217476,
      "auxiliary_loss_mlp": 0.01038366,
      "balance_loss_clip": 1.05709076,
      "balance_loss_mlp": 1.02817369,
      "epoch": 0.25792100042085014,
      "flos": 22820297846400.0,
      "grad_norm": 2.469334128097787,
      "language_loss": 0.764624,
      "learning_rate": 3.4795295321816804e-06,
      "loss": 0.78718239,
      "num_input_tokens_seen": 45850135,
      "step": 2145,
      "time_per_iteration": 2.678330183029175
    },
    {
      "auxiliary_loss_clip": 0.01202605,
      "auxiliary_loss_mlp": 0.01039786,
      "balance_loss_clip": 1.05558157,
      "balance_loss_mlp": 1.02976036,
      "epoch": 0.2580412433114892,
      "flos": 18697286194560.0,
      "grad_norm": 2.141009395572796,
      "language_loss": 0.91115081,
      "learning_rate": 3.47900527676272e-06,
      "loss": 0.93357474,
      "num_input_tokens_seen": 45868470,
      "step": 2146,
      "time_per_iteration": 2.6117143630981445
    },
    {
      "auxiliary_loss_clip": 0.0120983,
      "auxiliary_loss_mlp": 0.01034535,
      "balance_loss_clip": 1.06097794,
      "balance_loss_mlp": 1.0250107,
      "epoch": 0.2581614862021283,
      "flos": 14283146810880.0,
      "grad_norm": 2.323256202930795,
      "language_loss": 0.88410723,
      "learning_rate": 3.478480796979835e-06,
      "loss": 0.90655094,
      "num_input_tokens_seen": 45886355,
      "step": 2147,
      "time_per_iteration": 2.6810719966888428
    },
    {
      "auxiliary_loss_clip": 0.01207987,
      "auxiliary_loss_mlp": 0.01029111,
      "balance_loss_clip": 1.05541348,
      "balance_loss_mlp": 1.01999176,
      "epoch": 0.25828172909276736,
      "flos": 29498281856640.0,
      "grad_norm": 1.6476242085908803,
      "language_loss": 0.78061652,
      "learning_rate": 3.4779560929125894e-06,
      "loss": 0.80298758,
      "num_input_tokens_seen": 45907900,
      "step": 2148,
      "time_per_iteration": 2.7263333797454834
    },
    {
      "auxiliary_loss_clip": 0.01118731,
      "auxiliary_loss_mlp": 0.01006693,
      "balance_loss_clip": 1.02397752,
      "balance_loss_mlp": 1.00486875,
      "epoch": 0.2584019719834065,
      "flos": 67114387376640.0,
      "grad_norm": 0.6688127684802576,
      "language_loss": 0.56935501,
      "learning_rate": 3.4774311646405783e-06,
      "loss": 0.59060925,
      "num_input_tokens_seen": 45977805,
      "step": 2149,
      "time_per_iteration": 3.339311122894287
    },
    {
      "auxiliary_loss_clip": 0.01205172,
      "auxiliary_loss_mlp": 0.01031909,
      "balance_loss_clip": 1.05324793,
      "balance_loss_mlp": 1.02239609,
      "epoch": 0.2585222148740456,
      "flos": 22893555634560.0,
      "grad_norm": 1.9402537220176692,
      "language_loss": 0.8344993,
      "learning_rate": 3.476906012243435e-06,
      "loss": 0.85687011,
      "num_input_tokens_seen": 45996715,
      "step": 2150,
      "time_per_iteration": 3.595694065093994
    },
    {
      "auxiliary_loss_clip": 0.01200975,
      "auxiliary_loss_mlp": 0.01029723,
      "balance_loss_clip": 1.05809045,
      "balance_loss_mlp": 1.01968002,
      "epoch": 0.25864245776468464,
      "flos": 28909202808960.0,
      "grad_norm": 2.7302103927654167,
      "language_loss": 0.81360078,
      "learning_rate": 3.476380635800824e-06,
      "loss": 0.83590782,
      "num_input_tokens_seen": 46017915,
      "step": 2151,
      "time_per_iteration": 2.6796467304229736
    },
    {
      "auxiliary_loss_clip": 0.0121173,
      "auxiliary_loss_mlp": 0.0103074,
      "balance_loss_clip": 1.05832183,
      "balance_loss_mlp": 1.02147174,
      "epoch": 0.25876270065532375,
      "flos": 14793185980800.0,
      "grad_norm": 2.4930797187347102,
      "language_loss": 0.86325574,
      "learning_rate": 3.475855035392444e-06,
      "loss": 0.88568044,
      "num_input_tokens_seen": 46033235,
      "step": 2152,
      "time_per_iteration": 2.7061541080474854
    },
    {
      "auxiliary_loss_clip": 0.01213858,
      "auxiliary_loss_mlp": 0.01028133,
      "balance_loss_clip": 1.05187201,
      "balance_loss_mlp": 1.01860189,
      "epoch": 0.25888294354596286,
      "flos": 60467821810560.0,
      "grad_norm": 1.8561734877045764,
      "language_loss": 0.71614158,
      "learning_rate": 3.475329211098029e-06,
      "loss": 0.73856145,
      "num_input_tokens_seen": 46056390,
      "step": 2153,
      "time_per_iteration": 3.9726662635803223
    },
    {
      "auxiliary_loss_clip": 0.01220999,
      "auxiliary_loss_mlp": 0.01032261,
      "balance_loss_clip": 1.05729413,
      "balance_loss_mlp": 1.02301025,
      "epoch": 0.2590031864366019,
      "flos": 27851166771840.0,
      "grad_norm": 1.8974058331403834,
      "language_loss": 0.82593566,
      "learning_rate": 3.4748031629973453e-06,
      "loss": 0.8484683,
      "num_input_tokens_seen": 46077120,
      "step": 2154,
      "time_per_iteration": 2.7611539363861084
    },
    {
      "auxiliary_loss_clip": 0.01121228,
      "auxiliary_loss_mlp": 0.01000789,
      "balance_loss_clip": 1.02241194,
      "balance_loss_mlp": 0.99902481,
      "epoch": 0.25912342932724103,
      "flos": 62422444206720.0,
      "grad_norm": 0.9150670978812632,
      "language_loss": 0.56524241,
      "learning_rate": 3.4742768911701944e-06,
      "loss": 0.58646262,
      "num_input_tokens_seen": 46139815,
      "step": 2155,
      "time_per_iteration": 3.351438283920288
    },
    {
      "auxiliary_loss_clip": 0.01215839,
      "auxiliary_loss_mlp": 0.01040708,
      "balance_loss_clip": 1.05961215,
      "balance_loss_mlp": 1.02977097,
      "epoch": 0.25924367221788014,
      "flos": 12378839368320.0,
      "grad_norm": 3.09701833809034,
      "language_loss": 0.71643162,
      "learning_rate": 3.4737503956964113e-06,
      "loss": 0.7389971,
      "num_input_tokens_seen": 46152120,
      "step": 2156,
      "time_per_iteration": 2.6113834381103516
    },
    {
      "auxiliary_loss_clip": 0.01206364,
      "auxiliary_loss_mlp": 0.0104127,
      "balance_loss_clip": 1.05557847,
      "balance_loss_mlp": 1.03052306,
      "epoch": 0.2593639151085192,
      "flos": 14575208296320.0,
      "grad_norm": 2.0205574348791773,
      "language_loss": 0.67283225,
      "learning_rate": 3.473223676655865e-06,
      "loss": 0.69530857,
      "num_input_tokens_seen": 46170120,
      "step": 2157,
      "time_per_iteration": 3.556922197341919
    },
    {
      "auxiliary_loss_clip": 0.01204895,
      "auxiliary_loss_mlp": 0.01037142,
      "balance_loss_clip": 1.05277848,
      "balance_loss_mlp": 1.02641928,
      "epoch": 0.2594841579991583,
      "flos": 15230937029760.0,
      "grad_norm": 2.76771396658168,
      "language_loss": 0.79889858,
      "learning_rate": 3.472696734128459e-06,
      "loss": 0.82131898,
      "num_input_tokens_seen": 46187985,
      "step": 2158,
      "time_per_iteration": 2.671452283859253
    },
    {
      "auxiliary_loss_clip": 0.01212389,
      "auxiliary_loss_mlp": 0.01031795,
      "balance_loss_clip": 1.05959606,
      "balance_loss_mlp": 1.02196074,
      "epoch": 0.2596044008897974,
      "flos": 23623583650560.0,
      "grad_norm": 1.9911244631949314,
      "language_loss": 0.75819802,
      "learning_rate": 3.4721695681941286e-06,
      "loss": 0.78063989,
      "num_input_tokens_seen": 46207025,
      "step": 2159,
      "time_per_iteration": 3.562920570373535
    },
    {
      "auxiliary_loss_clip": 0.01210652,
      "auxiliary_loss_mlp": 0.01209614,
      "balance_loss_clip": 1.05514407,
      "balance_loss_mlp": 1.00054359,
      "epoch": 0.25972464378043647,
      "flos": 13772281628160.0,
      "grad_norm": 2.1937753058817853,
      "language_loss": 0.82733452,
      "learning_rate": 3.471642178932845e-06,
      "loss": 0.85153711,
      "num_input_tokens_seen": 46225670,
      "step": 2160,
      "time_per_iteration": 2.700988292694092
    },
    {
      "auxiliary_loss_clip": 0.01215379,
      "auxiliary_loss_mlp": 0.01032557,
      "balance_loss_clip": 1.05620432,
      "balance_loss_mlp": 1.02226329,
      "epoch": 0.2598448866710756,
      "flos": 19573578391680.0,
      "grad_norm": 2.2940862897877032,
      "language_loss": 0.89614546,
      "learning_rate": 3.471114566424613e-06,
      "loss": 0.91862488,
      "num_input_tokens_seen": 46244130,
      "step": 2161,
      "time_per_iteration": 2.6384665966033936
    },
    {
      "auxiliary_loss_clip": 0.01213868,
      "auxiliary_loss_mlp": 0.01033424,
      "balance_loss_clip": 1.0595367,
      "balance_loss_mlp": 1.02257037,
      "epoch": 0.25996512956171464,
      "flos": 21653237053440.0,
      "grad_norm": 1.9313776987181266,
      "language_loss": 0.75668442,
      "learning_rate": 3.4705867307494715e-06,
      "loss": 0.7791574,
      "num_input_tokens_seen": 46263200,
      "step": 2162,
      "time_per_iteration": 2.710615634918213
    },
    {
      "auxiliary_loss_clip": 0.01213603,
      "auxiliary_loss_mlp": 0.01031887,
      "balance_loss_clip": 1.05839992,
      "balance_loss_mlp": 1.02240443,
      "epoch": 0.26008537245235375,
      "flos": 18223480869120.0,
      "grad_norm": 2.9385334737366904,
      "language_loss": 0.84283173,
      "learning_rate": 3.470058671987492e-06,
      "loss": 0.86528665,
      "num_input_tokens_seen": 46281465,
      "step": 2163,
      "time_per_iteration": 2.706143379211426
    },
    {
      "auxiliary_loss_clip": 0.01215396,
      "auxiliary_loss_mlp": 0.01035492,
      "balance_loss_clip": 1.05763638,
      "balance_loss_mlp": 1.02480507,
      "epoch": 0.26020561534299286,
      "flos": 24645385843200.0,
      "grad_norm": 1.9917089583949672,
      "language_loss": 0.84011316,
      "learning_rate": 3.4695303902187805e-06,
      "loss": 0.86262208,
      "num_input_tokens_seen": 46301020,
      "step": 2164,
      "time_per_iteration": 2.694570779800415
    },
    {
      "auxiliary_loss_clip": 0.01210379,
      "auxiliary_loss_mlp": 0.01037283,
      "balance_loss_clip": 1.05198431,
      "balance_loss_mlp": 1.02690613,
      "epoch": 0.2603258582336319,
      "flos": 25773662926080.0,
      "grad_norm": 1.98321346493158,
      "language_loss": 0.78715128,
      "learning_rate": 3.469001885523478e-06,
      "loss": 0.80962789,
      "num_input_tokens_seen": 46321740,
      "step": 2165,
      "time_per_iteration": 2.71506929397583
    },
    {
      "auxiliary_loss_clip": 0.01207995,
      "auxiliary_loss_mlp": 0.01035897,
      "balance_loss_clip": 1.05848324,
      "balance_loss_mlp": 1.02553725,
      "epoch": 0.260446101124271,
      "flos": 28766314506240.0,
      "grad_norm": 2.3337498625191317,
      "language_loss": 0.81220388,
      "learning_rate": 3.4684731579817568e-06,
      "loss": 0.83464283,
      "num_input_tokens_seen": 46342730,
      "step": 2166,
      "time_per_iteration": 2.653966188430786
    },
    {
      "auxiliary_loss_clip": 0.01214273,
      "auxiliary_loss_mlp": 0.01033993,
      "balance_loss_clip": 1.05199814,
      "balance_loss_mlp": 1.02433753,
      "epoch": 0.26056634401491013,
      "flos": 25666757072640.0,
      "grad_norm": 1.7153587218095714,
      "language_loss": 0.76604903,
      "learning_rate": 3.4679442076738247e-06,
      "loss": 0.78853166,
      "num_input_tokens_seen": 46362445,
      "step": 2167,
      "time_per_iteration": 2.7907586097717285
    },
    {
      "auxiliary_loss_clip": 0.01212451,
      "auxiliary_loss_mlp": 0.01034647,
      "balance_loss_clip": 1.06083965,
      "balance_loss_mlp": 1.02378678,
      "epoch": 0.2606865869055492,
      "flos": 27052765217280.0,
      "grad_norm": 2.2767468223251024,
      "language_loss": 0.83292943,
      "learning_rate": 3.4674150346799245e-06,
      "loss": 0.85540044,
      "num_input_tokens_seen": 46382145,
      "step": 2168,
      "time_per_iteration": 2.627922773361206
    },
    {
      "auxiliary_loss_clip": 0.01212097,
      "auxiliary_loss_mlp": 0.01032824,
      "balance_loss_clip": 1.05758953,
      "balance_loss_mlp": 1.02284026,
      "epoch": 0.2608068297961883,
      "flos": 17712615686400.0,
      "grad_norm": 2.6444325743715846,
      "language_loss": 0.8035289,
      "learning_rate": 3.4668856390803295e-06,
      "loss": 0.82597816,
      "num_input_tokens_seen": 46400025,
      "step": 2169,
      "time_per_iteration": 2.6478583812713623
    },
    {
      "auxiliary_loss_clip": 0.01199533,
      "auxiliary_loss_mlp": 0.01033129,
      "balance_loss_clip": 1.0565002,
      "balance_loss_mlp": 1.0234257,
      "epoch": 0.2609270726868274,
      "flos": 18551632544640.0,
      "grad_norm": 2.125413934857295,
      "language_loss": 0.89551121,
      "learning_rate": 3.4663560209553495e-06,
      "loss": 0.91783786,
      "num_input_tokens_seen": 46418090,
      "step": 2170,
      "time_per_iteration": 2.630661964416504
    },
    {
      "auxiliary_loss_clip": 0.01206981,
      "auxiliary_loss_mlp": 0.01034849,
      "balance_loss_clip": 1.05713689,
      "balance_loss_mlp": 1.02487159,
      "epoch": 0.26104731557746647,
      "flos": 21835699165440.0,
      "grad_norm": 2.2837372065580923,
      "language_loss": 0.79451096,
      "learning_rate": 3.4658261803853267e-06,
      "loss": 0.81692916,
      "num_input_tokens_seen": 46436015,
      "step": 2171,
      "time_per_iteration": 2.7030837535858154
    },
    {
      "auxiliary_loss_clip": 0.01207889,
      "auxiliary_loss_mlp": 0.01031734,
      "balance_loss_clip": 1.05635309,
      "balance_loss_mlp": 1.02117229,
      "epoch": 0.2611675584681056,
      "flos": 21689650465920.0,
      "grad_norm": 12.378606142261336,
      "language_loss": 0.8068434,
      "learning_rate": 3.4652961174506383e-06,
      "loss": 0.82923961,
      "num_input_tokens_seen": 46455885,
      "step": 2172,
      "time_per_iteration": 2.6597001552581787
    },
    {
      "auxiliary_loss_clip": 0.01103184,
      "auxiliary_loss_mlp": 0.01003505,
      "balance_loss_clip": 1.0220716,
      "balance_loss_mlp": 1.00194323,
      "epoch": 0.2612878013587447,
      "flos": 71862101389440.0,
      "grad_norm": 0.9716735959246577,
      "language_loss": 0.58100641,
      "learning_rate": 3.464765832231694e-06,
      "loss": 0.60207331,
      "num_input_tokens_seen": 46510050,
      "step": 2173,
      "time_per_iteration": 3.2227823734283447
    },
    {
      "auxiliary_loss_clip": 0.01212504,
      "auxiliary_loss_mlp": 0.01029973,
      "balance_loss_clip": 1.06028461,
      "balance_loss_mlp": 1.02001309,
      "epoch": 0.26140804424938374,
      "flos": 20227511445120.0,
      "grad_norm": 1.8889031237800282,
      "language_loss": 0.70970625,
      "learning_rate": 3.4642353248089373e-06,
      "loss": 0.732131,
      "num_input_tokens_seen": 46528810,
      "step": 2174,
      "time_per_iteration": 2.6293911933898926
    },
    {
      "auxiliary_loss_clip": 0.01207446,
      "auxiliary_loss_mlp": 0.01032348,
      "balance_loss_clip": 1.055516,
      "balance_loss_mlp": 1.02214956,
      "epoch": 0.26152828714002285,
      "flos": 25557085872000.0,
      "grad_norm": 1.8814894681825691,
      "language_loss": 0.80217648,
      "learning_rate": 3.463704595262846e-06,
      "loss": 0.82457447,
      "num_input_tokens_seen": 46549690,
      "step": 2175,
      "time_per_iteration": 2.7354748249053955
    },
    {
      "auxiliary_loss_clip": 0.01212812,
      "auxiliary_loss_mlp": 0.01036586,
      "balance_loss_clip": 1.05644059,
      "balance_loss_mlp": 1.0266856,
      "epoch": 0.26164853003066196,
      "flos": 25446516831360.0,
      "grad_norm": 3.8191216671447337,
      "language_loss": 0.71086729,
      "learning_rate": 3.463173643673931e-06,
      "loss": 0.73336124,
      "num_input_tokens_seen": 46572215,
      "step": 2176,
      "time_per_iteration": 2.7846410274505615
    },
    {
      "auxiliary_loss_clip": 0.01110558,
      "auxiliary_loss_mlp": 0.01003475,
      "balance_loss_clip": 1.02181327,
      "balance_loss_mlp": 1.00191295,
      "epoch": 0.261768772921301,
      "flos": 53944580568960.0,
      "grad_norm": 0.9068622114535219,
      "language_loss": 0.63413686,
      "learning_rate": 3.4626424701227387e-06,
      "loss": 0.65527713,
      "num_input_tokens_seen": 46627275,
      "step": 2177,
      "time_per_iteration": 4.019364833831787
    },
    {
      "auxiliary_loss_clip": 0.01104422,
      "auxiliary_loss_mlp": 0.01006303,
      "balance_loss_clip": 1.02311254,
      "balance_loss_mlp": 1.00470543,
      "epoch": 0.26188901581194013,
      "flos": 70687606481280.0,
      "grad_norm": 0.8220073337359476,
      "language_loss": 0.55759108,
      "learning_rate": 3.4621110746898452e-06,
      "loss": 0.5786984,
      "num_input_tokens_seen": 46695135,
      "step": 2178,
      "time_per_iteration": 3.2374656200408936
    },
    {
      "auxiliary_loss_clip": 0.01215958,
      "auxiliary_loss_mlp": 0.01033101,
      "balance_loss_clip": 1.0613699,
      "balance_loss_mlp": 1.02336204,
      "epoch": 0.2620092587025792,
      "flos": 21069580959360.0,
      "grad_norm": 2.369960080489022,
      "language_loss": 0.74871683,
      "learning_rate": 3.4615794574558654e-06,
      "loss": 0.77120745,
      "num_input_tokens_seen": 46714145,
      "step": 2179,
      "time_per_iteration": 2.679001808166504
    },
    {
      "auxiliary_loss_clip": 0.01215479,
      "auxiliary_loss_mlp": 0.01030064,
      "balance_loss_clip": 1.05856371,
      "balance_loss_mlp": 1.02086711,
      "epoch": 0.2621295015932183,
      "flos": 18369601395840.0,
      "grad_norm": 4.749196828479744,
      "language_loss": 0.84514713,
      "learning_rate": 3.4610476185014436e-06,
      "loss": 0.86760259,
      "num_input_tokens_seen": 46731405,
      "step": 2180,
      "time_per_iteration": 3.5872678756713867
    },
    {
      "auxiliary_loss_clip": 0.01210757,
      "auxiliary_loss_mlp": 0.01033873,
      "balance_loss_clip": 1.05933452,
      "balance_loss_mlp": 1.02307224,
      "epoch": 0.2622497444838574,
      "flos": 23659997063040.0,
      "grad_norm": 1.7053989679804447,
      "language_loss": 0.79536408,
      "learning_rate": 3.4605155579072597e-06,
      "loss": 0.81781042,
      "num_input_tokens_seen": 46751260,
      "step": 2181,
      "time_per_iteration": 2.6787493228912354
    },
    {
      "auxiliary_loss_clip": 0.01212078,
      "auxiliary_loss_mlp": 0.01029749,
      "balance_loss_clip": 1.05378032,
      "balance_loss_mlp": 1.02009916,
      "epoch": 0.26236998737449646,
      "flos": 22123810154880.0,
      "grad_norm": 2.343313964168966,
      "language_loss": 0.71413726,
      "learning_rate": 3.459983275754027e-06,
      "loss": 0.73655552,
      "num_input_tokens_seen": 46770155,
      "step": 2182,
      "time_per_iteration": 2.695990800857544
    },
    {
      "auxiliary_loss_clip": 0.01208681,
      "auxiliary_loss_mlp": 0.01031958,
      "balance_loss_clip": 1.06001925,
      "balance_loss_mlp": 1.02201629,
      "epoch": 0.26249023026513557,
      "flos": 17895185539200.0,
      "grad_norm": 2.9051915934979844,
      "language_loss": 0.79719418,
      "learning_rate": 3.4594507721224918e-06,
      "loss": 0.81960058,
      "num_input_tokens_seen": 46788805,
      "step": 2183,
      "time_per_iteration": 3.6024272441864014
    },
    {
      "auxiliary_loss_clip": 0.01214241,
      "auxiliary_loss_mlp": 0.01041198,
      "balance_loss_clip": 1.05529535,
      "balance_loss_mlp": 1.03113055,
      "epoch": 0.2626104731557747,
      "flos": 18332936588160.0,
      "grad_norm": 4.530546829695175,
      "language_loss": 0.8210237,
      "learning_rate": 3.4589180470934353e-06,
      "loss": 0.8435781,
      "num_input_tokens_seen": 46808670,
      "step": 2184,
      "time_per_iteration": 2.7107698917388916
    },
    {
      "auxiliary_loss_clip": 0.01218667,
      "auxiliary_loss_mlp": 0.01037889,
      "balance_loss_clip": 1.05731678,
      "balance_loss_mlp": 1.0272193,
      "epoch": 0.26273071604641374,
      "flos": 19317714837120.0,
      "grad_norm": 1.7893153948553588,
      "language_loss": 0.76465267,
      "learning_rate": 3.4583851007476713e-06,
      "loss": 0.78721821,
      "num_input_tokens_seen": 46827140,
      "step": 2185,
      "time_per_iteration": 2.6208384037017822
    },
    {
      "auxiliary_loss_clip": 0.01219433,
      "auxiliary_loss_mlp": 0.01036644,
      "balance_loss_clip": 1.05742168,
      "balance_loss_mlp": 1.02572489,
      "epoch": 0.26285095893705285,
      "flos": 18327477720960.0,
      "grad_norm": 2.1918032909222327,
      "language_loss": 0.68421173,
      "learning_rate": 3.4578519331660464e-06,
      "loss": 0.70677257,
      "num_input_tokens_seen": 46844135,
      "step": 2186,
      "time_per_iteration": 3.6616506576538086
    },
    {
      "auxiliary_loss_clip": 0.01210125,
      "auxiliary_loss_mlp": 0.01035109,
      "balance_loss_clip": 1.06188583,
      "balance_loss_mlp": 1.02567911,
      "epoch": 0.26297120182769196,
      "flos": 20193827466240.0,
      "grad_norm": 2.2397651360425375,
      "language_loss": 0.82149935,
      "learning_rate": 3.4573185444294426e-06,
      "loss": 0.8439517,
      "num_input_tokens_seen": 46862500,
      "step": 2187,
      "time_per_iteration": 2.663437604904175
    },
    {
      "auxiliary_loss_clip": 0.01211884,
      "auxiliary_loss_mlp": 0.01209565,
      "balance_loss_clip": 1.05688953,
      "balance_loss_mlp": 1.00057364,
      "epoch": 0.263091444718331,
      "flos": 22418421505920.0,
      "grad_norm": 2.342799560421493,
      "language_loss": 0.78907323,
      "learning_rate": 3.456784934618774e-06,
      "loss": 0.81328768,
      "num_input_tokens_seen": 46883665,
      "step": 2188,
      "time_per_iteration": 2.6878435611724854
    },
    {
      "auxiliary_loss_clip": 0.01211454,
      "auxiliary_loss_mlp": 0.01031027,
      "balance_loss_clip": 1.05629826,
      "balance_loss_mlp": 1.02181172,
      "epoch": 0.2632116876089701,
      "flos": 19024827338880.0,
      "grad_norm": 1.9493241957821434,
      "language_loss": 0.80011982,
      "learning_rate": 3.4562511038149897e-06,
      "loss": 0.82254457,
      "num_input_tokens_seen": 46899160,
      "step": 2189,
      "time_per_iteration": 2.614307403564453
    },
    {
      "auxiliary_loss_clip": 0.01119547,
      "auxiliary_loss_mlp": 0.0100092,
      "balance_loss_clip": 1.02216482,
      "balance_loss_mlp": 0.99917936,
      "epoch": 0.26333193049960923,
      "flos": 67308054531840.0,
      "grad_norm": 0.8556711228408594,
      "language_loss": 0.57680035,
      "learning_rate": 3.4557170520990705e-06,
      "loss": 0.59800506,
      "num_input_tokens_seen": 46959835,
      "step": 2190,
      "time_per_iteration": 3.3181610107421875
    },
    {
      "auxiliary_loss_clip": 0.01204785,
      "auxiliary_loss_mlp": 0.01037894,
      "balance_loss_clip": 1.0568285,
      "balance_loss_mlp": 1.02840543,
      "epoch": 0.2634521733902483,
      "flos": 25048806468480.0,
      "grad_norm": 1.5424756029824755,
      "language_loss": 0.86615431,
      "learning_rate": 3.4551827795520324e-06,
      "loss": 0.8885811,
      "num_input_tokens_seen": 46982720,
      "step": 2191,
      "time_per_iteration": 2.7063896656036377
    },
    {
      "auxiliary_loss_clip": 0.01211623,
      "auxiliary_loss_mlp": 0.0102994,
      "balance_loss_clip": 1.05740881,
      "balance_loss_mlp": 1.02104664,
      "epoch": 0.2635724162808874,
      "flos": 20594985534720.0,
      "grad_norm": 1.6524954286228222,
      "language_loss": 0.84924948,
      "learning_rate": 3.4546482862549226e-06,
      "loss": 0.87166512,
      "num_input_tokens_seen": 47003035,
      "step": 2192,
      "time_per_iteration": 2.6292688846588135
    },
    {
      "auxiliary_loss_clip": 0.01207211,
      "auxiliary_loss_mlp": 0.01040354,
      "balance_loss_clip": 1.05268228,
      "balance_loss_mlp": 1.02976179,
      "epoch": 0.2636926591715265,
      "flos": 19244636616960.0,
      "grad_norm": 2.34957893183231,
      "language_loss": 0.78636312,
      "learning_rate": 3.4541135722888253e-06,
      "loss": 0.80883873,
      "num_input_tokens_seen": 47019625,
      "step": 2193,
      "time_per_iteration": 2.7315213680267334
    },
    {
      "auxiliary_loss_clip": 0.01206123,
      "auxiliary_loss_mlp": 0.01030897,
      "balance_loss_clip": 1.05825281,
      "balance_loss_mlp": 1.02069831,
      "epoch": 0.26381290206216557,
      "flos": 28804882734720.0,
      "grad_norm": 3.1934557230961835,
      "language_loss": 0.80140233,
      "learning_rate": 3.453578637734854e-06,
      "loss": 0.82377255,
      "num_input_tokens_seen": 47040815,
      "step": 2194,
      "time_per_iteration": 2.6743950843811035
    },
    {
      "auxiliary_loss_clip": 0.01213751,
      "auxiliary_loss_mlp": 0.01034725,
      "balance_loss_clip": 1.06475496,
      "balance_loss_mlp": 1.0245508,
      "epoch": 0.2639331449528047,
      "flos": 25008909436800.0,
      "grad_norm": 1.9520402821107303,
      "language_loss": 0.78789842,
      "learning_rate": 3.4530434826741605e-06,
      "loss": 0.8103832,
      "num_input_tokens_seen": 47061755,
      "step": 2195,
      "time_per_iteration": 2.7310187816619873
    },
    {
      "auxiliary_loss_clip": 0.01207647,
      "auxiliary_loss_mlp": 0.0102997,
      "balance_loss_clip": 1.05693471,
      "balance_loss_mlp": 1.02087426,
      "epoch": 0.26405338784344373,
      "flos": 46535775465600.0,
      "grad_norm": 3.056083783085084,
      "language_loss": 0.69079781,
      "learning_rate": 3.452508107187926e-06,
      "loss": 0.71317399,
      "num_input_tokens_seen": 47085130,
      "step": 2196,
      "time_per_iteration": 2.864842414855957
    },
    {
      "auxiliary_loss_clip": 0.01216944,
      "auxiliary_loss_mlp": 0.01031681,
      "balance_loss_clip": 1.0490346,
      "balance_loss_mlp": 1.02086282,
      "epoch": 0.26417363073408284,
      "flos": 21179467641600.0,
      "grad_norm": 2.060052989930734,
      "language_loss": 0.77032,
      "learning_rate": 3.451972511357366e-06,
      "loss": 0.79280627,
      "num_input_tokens_seen": 47104675,
      "step": 2197,
      "time_per_iteration": 2.763040542602539
    },
    {
      "auxiliary_loss_clip": 0.01207642,
      "auxiliary_loss_mlp": 0.01031416,
      "balance_loss_clip": 1.05902624,
      "balance_loss_mlp": 1.02262473,
      "epoch": 0.26429387362472195,
      "flos": 22674751937280.0,
      "grad_norm": 1.7643163058181477,
      "language_loss": 0.85288179,
      "learning_rate": 3.45143669526373e-06,
      "loss": 0.87527239,
      "num_input_tokens_seen": 47124435,
      "step": 2198,
      "time_per_iteration": 2.716088056564331
    },
    {
      "auxiliary_loss_clip": 0.01115232,
      "auxiliary_loss_mlp": 0.01001894,
      "balance_loss_clip": 1.02026379,
      "balance_loss_mlp": 1.00028443,
      "epoch": 0.264414116515361,
      "flos": 67180534272000.0,
      "grad_norm": 0.7980990251777145,
      "language_loss": 0.6324302,
      "learning_rate": 3.450900658988302e-06,
      "loss": 0.65360153,
      "num_input_tokens_seen": 47185985,
      "step": 2199,
      "time_per_iteration": 3.174163818359375
    },
    {
      "auxiliary_loss_clip": 0.01202085,
      "auxiliary_loss_mlp": 0.01035401,
      "balance_loss_clip": 1.05492163,
      "balance_loss_mlp": 1.02551222,
      "epoch": 0.2645343594060001,
      "flos": 25664709997440.0,
      "grad_norm": 1.8417868715107872,
      "language_loss": 0.77808964,
      "learning_rate": 3.450364402612397e-06,
      "loss": 0.80046445,
      "num_input_tokens_seen": 47203140,
      "step": 2200,
      "time_per_iteration": 2.758070945739746
    },
    {
      "auxiliary_loss_clip": 0.01208934,
      "auxiliary_loss_mlp": 0.01031413,
      "balance_loss_clip": 1.05562425,
      "balance_loss_mlp": 1.02092314,
      "epoch": 0.26465460229663923,
      "flos": 22491822948480.0,
      "grad_norm": 2.3301392134027448,
      "language_loss": 0.84008759,
      "learning_rate": 3.449827926217366e-06,
      "loss": 0.86249107,
      "num_input_tokens_seen": 47222575,
      "step": 2201,
      "time_per_iteration": 2.6501238346099854
    },
    {
      "auxiliary_loss_clip": 0.01213684,
      "auxiliary_loss_mlp": 0.01034654,
      "balance_loss_clip": 1.05207145,
      "balance_loss_mlp": 1.02472365,
      "epoch": 0.2647748451872783,
      "flos": 29388036038400.0,
      "grad_norm": 1.8917607885518337,
      "language_loss": 0.80505359,
      "learning_rate": 3.449291229884591e-06,
      "loss": 0.82753706,
      "num_input_tokens_seen": 47243815,
      "step": 2202,
      "time_per_iteration": 2.7789480686187744
    },
    {
      "auxiliary_loss_clip": 0.01215392,
      "auxiliary_loss_mlp": 0.010291,
      "balance_loss_clip": 1.05185235,
      "balance_loss_mlp": 1.01891327,
      "epoch": 0.2648950880779174,
      "flos": 26797799502720.0,
      "grad_norm": 1.876015040304544,
      "language_loss": 0.86767566,
      "learning_rate": 3.4487543136954887e-06,
      "loss": 0.89012063,
      "num_input_tokens_seen": 47263435,
      "step": 2203,
      "time_per_iteration": 2.6921775341033936
    },
    {
      "auxiliary_loss_clip": 0.01210618,
      "auxiliary_loss_mlp": 0.01031718,
      "balance_loss_clip": 1.0530045,
      "balance_loss_mlp": 1.02182341,
      "epoch": 0.2650153309685565,
      "flos": 28841008838400.0,
      "grad_norm": 2.031323599432084,
      "language_loss": 0.91244602,
      "learning_rate": 3.448217177731509e-06,
      "loss": 0.93486941,
      "num_input_tokens_seen": 47283920,
      "step": 2204,
      "time_per_iteration": 3.6563479900360107
    },
    {
      "auxiliary_loss_clip": 0.01205615,
      "auxiliary_loss_mlp": 0.01033318,
      "balance_loss_clip": 1.05634809,
      "balance_loss_mlp": 1.02438903,
      "epoch": 0.26513557385919556,
      "flos": 20303247271680.0,
      "grad_norm": 1.9807923310955522,
      "language_loss": 0.77932882,
      "learning_rate": 3.4476798220741348e-06,
      "loss": 0.80171818,
      "num_input_tokens_seen": 47302800,
      "step": 2205,
      "time_per_iteration": 2.7239325046539307
    },
    {
      "auxiliary_loss_clip": 0.01207461,
      "auxiliary_loss_mlp": 0.01031043,
      "balance_loss_clip": 1.06074905,
      "balance_loss_mlp": 1.02217364,
      "epoch": 0.26525581674983467,
      "flos": 17676274101120.0,
      "grad_norm": 1.7982697803946994,
      "language_loss": 0.77987766,
      "learning_rate": 3.4471422468048826e-06,
      "loss": 0.80226266,
      "num_input_tokens_seen": 47321525,
      "step": 2206,
      "time_per_iteration": 3.5724716186523438
    },
    {
      "auxiliary_loss_clip": 0.01199892,
      "auxiliary_loss_mlp": 0.01031539,
      "balance_loss_clip": 1.05699098,
      "balance_loss_mlp": 1.02223492,
      "epoch": 0.2653760596404738,
      "flos": 26833746038400.0,
      "grad_norm": 3.6897176443756408,
      "language_loss": 0.72829181,
      "learning_rate": 3.4466044520053022e-06,
      "loss": 0.75060606,
      "num_input_tokens_seen": 47340530,
      "step": 2207,
      "time_per_iteration": 2.694775104522705
    },
    {
      "auxiliary_loss_clip": 0.01196248,
      "auxiliary_loss_mlp": 0.01036203,
      "balance_loss_clip": 1.05228376,
      "balance_loss_mlp": 1.02664208,
      "epoch": 0.26549630253111284,
      "flos": 22782160581120.0,
      "grad_norm": 1.8646432711881136,
      "language_loss": 0.60087544,
      "learning_rate": 3.446066437756977e-06,
      "loss": 0.62319994,
      "num_input_tokens_seen": 47359735,
      "step": 2208,
      "time_per_iteration": 2.652062177658081
    },
    {
      "auxiliary_loss_clip": 0.01208149,
      "auxiliary_loss_mlp": 0.01027956,
      "balance_loss_clip": 1.05549979,
      "balance_loss_mlp": 1.01858592,
      "epoch": 0.26561654542175195,
      "flos": 23550002640000.0,
      "grad_norm": 1.9927722412740307,
      "language_loss": 0.75382656,
      "learning_rate": 3.4455282041415224e-06,
      "loss": 0.7761876,
      "num_input_tokens_seen": 47378945,
      "step": 2209,
      "time_per_iteration": 2.668468475341797
    },
    {
      "auxiliary_loss_clip": 0.01214034,
      "auxiliary_loss_mlp": 0.01028559,
      "balance_loss_clip": 1.0546658,
      "balance_loss_mlp": 1.01911736,
      "epoch": 0.265736788312391,
      "flos": 26906680604160.0,
      "grad_norm": 3.6674706660398204,
      "language_loss": 0.87410563,
      "learning_rate": 3.4449897512405894e-06,
      "loss": 0.89653158,
      "num_input_tokens_seen": 47398095,
      "step": 2210,
      "time_per_iteration": 3.6614556312561035
    },
    {
      "auxiliary_loss_clip": 0.01209671,
      "auxiliary_loss_mlp": 0.0120882,
      "balance_loss_clip": 1.04912007,
      "balance_loss_mlp": 1.00049567,
      "epoch": 0.2658570312030301,
      "flos": 23477139901440.0,
      "grad_norm": 3.82171643286649,
      "language_loss": 0.7544449,
      "learning_rate": 3.444451079135859e-06,
      "loss": 0.77862984,
      "num_input_tokens_seen": 47417605,
      "step": 2211,
      "time_per_iteration": 2.7083778381347656
    },
    {
      "auxiliary_loss_clip": 0.01204748,
      "auxiliary_loss_mlp": 0.01209301,
      "balance_loss_clip": 1.04889488,
      "balance_loss_mlp": 1.00044847,
      "epoch": 0.2659772740936692,
      "flos": 21866402315520.0,
      "grad_norm": 2.3220065265281433,
      "language_loss": 0.74642718,
      "learning_rate": 3.4439121879090493e-06,
      "loss": 0.77056772,
      "num_input_tokens_seen": 47435385,
      "step": 2212,
      "time_per_iteration": 3.6869192123413086
    },
    {
      "auxiliary_loss_clip": 0.01212305,
      "auxiliary_loss_mlp": 0.01032502,
      "balance_loss_clip": 1.05491066,
      "balance_loss_mlp": 1.02270317,
      "epoch": 0.2660975169843083,
      "flos": 19793100360960.0,
      "grad_norm": 2.1812872504657888,
      "language_loss": 0.83229458,
      "learning_rate": 3.4433730776419082e-06,
      "loss": 0.85474265,
      "num_input_tokens_seen": 47454310,
      "step": 2213,
      "time_per_iteration": 2.6430397033691406
    },
    {
      "auxiliary_loss_clip": 0.01210885,
      "auxiliary_loss_mlp": 0.01209193,
      "balance_loss_clip": 1.05520475,
      "balance_loss_mlp": 1.0007025,
      "epoch": 0.2662177598749474,
      "flos": 29018981750400.0,
      "grad_norm": 2.326963608266322,
      "language_loss": 0.80642068,
      "learning_rate": 3.4428337484162183e-06,
      "loss": 0.83062148,
      "num_input_tokens_seen": 47475120,
      "step": 2214,
      "time_per_iteration": 2.7276670932769775
    },
    {
      "auxiliary_loss_clip": 0.01205989,
      "auxiliary_loss_mlp": 0.01032619,
      "balance_loss_clip": 1.05387878,
      "balance_loss_mlp": 1.02307034,
      "epoch": 0.2663380027655865,
      "flos": 21762549118080.0,
      "grad_norm": 1.979951739376231,
      "language_loss": 0.84194744,
      "learning_rate": 3.442294200313797e-06,
      "loss": 0.86433351,
      "num_input_tokens_seen": 47493150,
      "step": 2215,
      "time_per_iteration": 2.7321114540100098
    },
    {
      "auxiliary_loss_clip": 0.01099277,
      "auxiliary_loss_mlp": 0.01007141,
      "balance_loss_clip": 1.01899278,
      "balance_loss_mlp": 1.00553215,
      "epoch": 0.26645824565622556,
      "flos": 66980333819520.0,
      "grad_norm": 0.7777439560558945,
      "language_loss": 0.52693665,
      "learning_rate": 3.4417544334164916e-06,
      "loss": 0.54800081,
      "num_input_tokens_seen": 47557295,
      "step": 2216,
      "time_per_iteration": 3.2037813663482666
    },
    {
      "auxiliary_loss_clip": 0.0120606,
      "auxiliary_loss_mlp": 0.01030822,
      "balance_loss_clip": 1.05405223,
      "balance_loss_mlp": 1.02104735,
      "epoch": 0.26657848854686467,
      "flos": 25264198373760.0,
      "grad_norm": 1.5687294060345909,
      "language_loss": 0.77391696,
      "learning_rate": 3.4412144478061854e-06,
      "loss": 0.79628575,
      "num_input_tokens_seen": 47579705,
      "step": 2217,
      "time_per_iteration": 2.7701163291931152
    },
    {
      "auxiliary_loss_clip": 0.0121693,
      "auxiliary_loss_mlp": 0.01031917,
      "balance_loss_clip": 1.04537785,
      "balance_loss_mlp": 1.02167141,
      "epoch": 0.2666987314375038,
      "flos": 23696769611520.0,
      "grad_norm": 2.4279361979612317,
      "language_loss": 0.75347316,
      "learning_rate": 3.4406742435647925e-06,
      "loss": 0.77596176,
      "num_input_tokens_seen": 47599770,
      "step": 2218,
      "time_per_iteration": 2.773649215698242
    },
    {
      "auxiliary_loss_clip": 0.01203704,
      "auxiliary_loss_mlp": 0.01034733,
      "balance_loss_clip": 1.05777574,
      "balance_loss_mlp": 1.02559602,
      "epoch": 0.26681897432814283,
      "flos": 27048958375680.0,
      "grad_norm": 2.6008002168351725,
      "language_loss": 0.78810543,
      "learning_rate": 3.440133820774263e-06,
      "loss": 0.81048977,
      "num_input_tokens_seen": 47619580,
      "step": 2219,
      "time_per_iteration": 2.731419086456299
    },
    {
      "auxiliary_loss_clip": 0.01212269,
      "auxiliary_loss_mlp": 0.0104122,
      "balance_loss_clip": 1.05547619,
      "balance_loss_mlp": 1.03066349,
      "epoch": 0.26693921721878194,
      "flos": 28985944216320.0,
      "grad_norm": 2.2506363470434985,
      "language_loss": 0.81748807,
      "learning_rate": 3.439593179516578e-06,
      "loss": 0.84002292,
      "num_input_tokens_seen": 47639490,
      "step": 2220,
      "time_per_iteration": 2.68819522857666
    },
    {
      "auxiliary_loss_clip": 0.01214003,
      "auxiliary_loss_mlp": 0.01031336,
      "balance_loss_clip": 1.05729771,
      "balance_loss_mlp": 1.02112007,
      "epoch": 0.26705946010942105,
      "flos": 21507834798720.0,
      "grad_norm": 2.1201683445542563,
      "language_loss": 0.81008971,
      "learning_rate": 3.4390523198737524e-06,
      "loss": 0.83254313,
      "num_input_tokens_seen": 47658650,
      "step": 2221,
      "time_per_iteration": 2.715590238571167
    },
    {
      "auxiliary_loss_clip": 0.0120611,
      "auxiliary_loss_mlp": 0.01209005,
      "balance_loss_clip": 1.05827975,
      "balance_loss_mlp": 1.00055432,
      "epoch": 0.2671797030000601,
      "flos": 21471277731840.0,
      "grad_norm": 1.987882921569226,
      "language_loss": 0.73652446,
      "learning_rate": 3.4385112419278333e-06,
      "loss": 0.76067561,
      "num_input_tokens_seen": 47679875,
      "step": 2222,
      "time_per_iteration": 2.61106014251709
    },
    {
      "auxiliary_loss_clip": 0.0110766,
      "auxiliary_loss_mlp": 0.01002701,
      "balance_loss_clip": 1.02018702,
      "balance_loss_mlp": 1.00107992,
      "epoch": 0.2672999458906992,
      "flos": 64189929767040.0,
      "grad_norm": 0.7892599437410619,
      "language_loss": 0.64787799,
      "learning_rate": 3.4379699457609033e-06,
      "loss": 0.66898161,
      "num_input_tokens_seen": 47737700,
      "step": 2223,
      "time_per_iteration": 3.1376540660858154
    },
    {
      "auxiliary_loss_clip": 0.01200698,
      "auxiliary_loss_mlp": 0.01028647,
      "balance_loss_clip": 1.05273616,
      "balance_loss_mlp": 1.01881242,
      "epoch": 0.26742018878133833,
      "flos": 16909042573440.0,
      "grad_norm": 2.8328250477262444,
      "language_loss": 0.90468347,
      "learning_rate": 3.4374284314550755e-06,
      "loss": 0.92697692,
      "num_input_tokens_seen": 47756740,
      "step": 2224,
      "time_per_iteration": 2.6900622844696045
    },
    {
      "auxiliary_loss_clip": 0.01203944,
      "auxiliary_loss_mlp": 0.01027483,
      "balance_loss_clip": 1.05766606,
      "balance_loss_mlp": 1.01786876,
      "epoch": 0.2675404316719774,
      "flos": 20667560964480.0,
      "grad_norm": 2.196537159252471,
      "language_loss": 0.81003475,
      "learning_rate": 3.436886699092498e-06,
      "loss": 0.83234906,
      "num_input_tokens_seen": 47775255,
      "step": 2225,
      "time_per_iteration": 2.6765358448028564
    },
    {
      "auxiliary_loss_clip": 0.01207763,
      "auxiliary_loss_mlp": 0.01034365,
      "balance_loss_clip": 1.05887067,
      "balance_loss_mlp": 1.02444112,
      "epoch": 0.2676606745626165,
      "flos": 17485013157120.0,
      "grad_norm": 2.6862249028979917,
      "language_loss": 0.71879482,
      "learning_rate": 3.4363447487553502e-06,
      "loss": 0.74121612,
      "num_input_tokens_seen": 47788570,
      "step": 2226,
      "time_per_iteration": 2.5973825454711914
    },
    {
      "auxiliary_loss_clip": 0.01206338,
      "auxiliary_loss_mlp": 0.01031642,
      "balance_loss_clip": 1.05574834,
      "balance_loss_mlp": 1.02140141,
      "epoch": 0.26778091745325555,
      "flos": 27852675143040.0,
      "grad_norm": 1.8585211618143127,
      "language_loss": 0.78551084,
      "learning_rate": 3.4358025805258455e-06,
      "loss": 0.80789065,
      "num_input_tokens_seen": 47808275,
      "step": 2227,
      "time_per_iteration": 2.6972358226776123
    },
    {
      "auxiliary_loss_clip": 0.01218315,
      "auxiliary_loss_mlp": 0.01028488,
      "balance_loss_clip": 1.05065954,
      "balance_loss_mlp": 1.01889181,
      "epoch": 0.26790116034389466,
      "flos": 20955995176320.0,
      "grad_norm": 1.979020741465299,
      "language_loss": 0.83721632,
      "learning_rate": 3.435260194486232e-06,
      "loss": 0.85968435,
      "num_input_tokens_seen": 47826245,
      "step": 2228,
      "time_per_iteration": 2.9043164253234863
    },
    {
      "auxiliary_loss_clip": 0.01210022,
      "auxiliary_loss_mlp": 0.01032264,
      "balance_loss_clip": 1.05636454,
      "balance_loss_mlp": 1.02231562,
      "epoch": 0.2680214032345338,
      "flos": 18040659621120.0,
      "grad_norm": 2.2313590115015436,
      "language_loss": 0.82244039,
      "learning_rate": 3.4347175907187875e-06,
      "loss": 0.84486324,
      "num_input_tokens_seen": 47843235,
      "step": 2229,
      "time_per_iteration": 2.6375668048858643
    },
    {
      "auxiliary_loss_clip": 0.01208846,
      "auxiliary_loss_mlp": 0.0103453,
      "balance_loss_clip": 1.05910373,
      "balance_loss_mlp": 1.02573872,
      "epoch": 0.26814164612517283,
      "flos": 22419427086720.0,
      "grad_norm": 1.95123261233312,
      "language_loss": 0.88047135,
      "learning_rate": 3.4341747693058254e-06,
      "loss": 0.90290511,
      "num_input_tokens_seen": 47861710,
      "step": 2230,
      "time_per_iteration": 2.691683292388916
    },
    {
      "auxiliary_loss_clip": 0.01211385,
      "auxiliary_loss_mlp": 0.0103293,
      "balance_loss_clip": 1.04680681,
      "balance_loss_mlp": 1.02368522,
      "epoch": 0.26826188901581194,
      "flos": 35627371159680.0,
      "grad_norm": 2.9066678881651553,
      "language_loss": 0.77207315,
      "learning_rate": 3.4336317303296916e-06,
      "loss": 0.79451632,
      "num_input_tokens_seen": 47882685,
      "step": 2231,
      "time_per_iteration": 3.8344340324401855
    },
    {
      "auxiliary_loss_clip": 0.01202788,
      "auxiliary_loss_mlp": 0.01028918,
      "balance_loss_clip": 1.0558517,
      "balance_loss_mlp": 1.01943445,
      "epoch": 0.26838213190645105,
      "flos": 17639788861440.0,
      "grad_norm": 2.252053005350333,
      "language_loss": 0.75333107,
      "learning_rate": 3.4330884738727635e-06,
      "loss": 0.77564812,
      "num_input_tokens_seen": 47900860,
      "step": 2232,
      "time_per_iteration": 2.651583433151245
    },
    {
      "auxiliary_loss_clip": 0.01208068,
      "auxiliary_loss_mlp": 0.01031106,
      "balance_loss_clip": 1.05355966,
      "balance_loss_mlp": 1.0213306,
      "epoch": 0.2685023747970901,
      "flos": 22674823764480.0,
      "grad_norm": 2.015630378815436,
      "language_loss": 0.7091772,
      "learning_rate": 3.4325450000174535e-06,
      "loss": 0.73156893,
      "num_input_tokens_seen": 47917500,
      "step": 2233,
      "time_per_iteration": 3.6790568828582764
    },
    {
      "auxiliary_loss_clip": 0.01207197,
      "auxiliary_loss_mlp": 0.01033082,
      "balance_loss_clip": 1.05329609,
      "balance_loss_mlp": 1.02293718,
      "epoch": 0.2686226176877292,
      "flos": 20120533764480.0,
      "grad_norm": 2.0263595836156787,
      "language_loss": 0.74579006,
      "learning_rate": 3.4320013088462067e-06,
      "loss": 0.76819283,
      "num_input_tokens_seen": 47934860,
      "step": 2234,
      "time_per_iteration": 2.6656200885772705
    },
    {
      "auxiliary_loss_clip": 0.01216204,
      "auxiliary_loss_mlp": 0.01030954,
      "balance_loss_clip": 1.05327225,
      "balance_loss_mlp": 1.02135766,
      "epoch": 0.2687428605783683,
      "flos": 21872040750720.0,
      "grad_norm": 1.4866282298340654,
      "language_loss": 0.81746972,
      "learning_rate": 3.431457400441499e-06,
      "loss": 0.83994126,
      "num_input_tokens_seen": 47955255,
      "step": 2235,
      "time_per_iteration": 2.7107553482055664
    },
    {
      "auxiliary_loss_clip": 0.01119114,
      "auxiliary_loss_mlp": 0.01000442,
      "balance_loss_clip": 1.01961637,
      "balance_loss_mlp": 0.99877322,
      "epoch": 0.2688631034690074,
      "flos": 69943320766080.0,
      "grad_norm": 0.910445521338648,
      "language_loss": 0.60808194,
      "learning_rate": 3.4309132748858424e-06,
      "loss": 0.62927747,
      "num_input_tokens_seen": 48016245,
      "step": 2236,
      "time_per_iteration": 4.28736686706543
    },
    {
      "auxiliary_loss_clip": 0.01206099,
      "auxiliary_loss_mlp": 0.01034858,
      "balance_loss_clip": 1.05960226,
      "balance_loss_mlp": 1.02517259,
      "epoch": 0.2689833463596465,
      "flos": 22856639431680.0,
      "grad_norm": 1.9847047464103025,
      "language_loss": 0.83620137,
      "learning_rate": 3.430368932261779e-06,
      "loss": 0.85861093,
      "num_input_tokens_seen": 48036600,
      "step": 2237,
      "time_per_iteration": 2.9872779846191406
    },
    {
      "auxiliary_loss_clip": 0.01208138,
      "auxiliary_loss_mlp": 0.01030948,
      "balance_loss_clip": 1.05616784,
      "balance_loss_mlp": 1.02111948,
      "epoch": 0.2691035892502856,
      "flos": 17200242132480.0,
      "grad_norm": 1.8564442968861337,
      "language_loss": 0.74872977,
      "learning_rate": 3.429824372651886e-06,
      "loss": 0.77112061,
      "num_input_tokens_seen": 48054750,
      "step": 2238,
      "time_per_iteration": 2.762730121612549
    },
    {
      "auxiliary_loss_clip": 0.01223573,
      "auxiliary_loss_mlp": 0.01035655,
      "balance_loss_clip": 1.05704021,
      "balance_loss_mlp": 1.02568948,
      "epoch": 0.26922383214092466,
      "flos": 17747484814080.0,
      "grad_norm": 1.9021667019737696,
      "language_loss": 0.8311196,
      "learning_rate": 3.4292795961387732e-06,
      "loss": 0.85371184,
      "num_input_tokens_seen": 48072650,
      "step": 2239,
      "time_per_iteration": 3.674438953399658
    },
    {
      "auxiliary_loss_clip": 0.01206825,
      "auxiliary_loss_mlp": 0.01032609,
      "balance_loss_clip": 1.05898356,
      "balance_loss_mlp": 1.02361488,
      "epoch": 0.26934407503156377,
      "flos": 16173376122240.0,
      "grad_norm": 2.9453375140389078,
      "language_loss": 0.87614644,
      "learning_rate": 3.4287346028050818e-06,
      "loss": 0.89854074,
      "num_input_tokens_seen": 48088720,
      "step": 2240,
      "time_per_iteration": 2.5672171115875244
    },
    {
      "auxiliary_loss_clip": 0.01206192,
      "auxiliary_loss_mlp": 0.01030141,
      "balance_loss_clip": 1.05379283,
      "balance_loss_mlp": 1.02111101,
      "epoch": 0.2694643179222028,
      "flos": 23732895715200.0,
      "grad_norm": 2.2705844519012475,
      "language_loss": 0.79601759,
      "learning_rate": 3.4281893927334866e-06,
      "loss": 0.81838089,
      "num_input_tokens_seen": 48108630,
      "step": 2241,
      "time_per_iteration": 2.6858103275299072
    },
    {
      "auxiliary_loss_clip": 0.01208853,
      "auxiliary_loss_mlp": 0.0102996,
      "balance_loss_clip": 1.05961549,
      "balance_loss_mlp": 1.02117467,
      "epoch": 0.26958456081284193,
      "flos": 24718140840960.0,
      "grad_norm": 2.4617732202859326,
      "language_loss": 0.75180626,
      "learning_rate": 3.4276439660066963e-06,
      "loss": 0.77419448,
      "num_input_tokens_seen": 48128330,
      "step": 2242,
      "time_per_iteration": 2.657222270965576
    },
    {
      "auxiliary_loss_clip": 0.01204503,
      "auxiliary_loss_mlp": 0.0103015,
      "balance_loss_clip": 1.05917203,
      "balance_loss_mlp": 1.02030921,
      "epoch": 0.26970480370348104,
      "flos": 18112588606080.0,
      "grad_norm": 2.5857812161244293,
      "language_loss": 0.8463999,
      "learning_rate": 3.427098322707452e-06,
      "loss": 0.8687464,
      "num_input_tokens_seen": 48144295,
      "step": 2243,
      "time_per_iteration": 2.7062888145446777
    },
    {
      "auxiliary_loss_clip": 0.0120961,
      "auxiliary_loss_mlp": 0.01036326,
      "balance_loss_clip": 1.06378436,
      "balance_loss_mlp": 1.02562737,
      "epoch": 0.2698250465941201,
      "flos": 10816546250880.0,
      "grad_norm": 2.819814973205774,
      "language_loss": 0.89821172,
      "learning_rate": 3.426552462918526e-06,
      "loss": 0.92067099,
      "num_input_tokens_seen": 48162230,
      "step": 2244,
      "time_per_iteration": 2.622337818145752
    },
    {
      "auxiliary_loss_clip": 0.0120516,
      "auxiliary_loss_mlp": 0.01032644,
      "balance_loss_clip": 1.06049562,
      "balance_loss_mlp": 1.02375078,
      "epoch": 0.2699452894847592,
      "flos": 17308117653120.0,
      "grad_norm": 2.3853796815110777,
      "language_loss": 0.73446655,
      "learning_rate": 3.426006386722726e-06,
      "loss": 0.75684452,
      "num_input_tokens_seen": 48180290,
      "step": 2245,
      "time_per_iteration": 2.645542860031128
    },
    {
      "auxiliary_loss_clip": 0.01213049,
      "auxiliary_loss_mlp": 0.01037464,
      "balance_loss_clip": 1.05825782,
      "balance_loss_mlp": 1.02808225,
      "epoch": 0.2700655323753983,
      "flos": 18078150441600.0,
      "grad_norm": 1.8867916199381682,
      "language_loss": 0.92399621,
      "learning_rate": 3.4254600942028914e-06,
      "loss": 0.94650137,
      "num_input_tokens_seen": 48198165,
      "step": 2246,
      "time_per_iteration": 2.6258647441864014
    },
    {
      "auxiliary_loss_clip": 0.01208175,
      "auxiliary_loss_mlp": 0.01033144,
      "balance_loss_clip": 1.0583539,
      "balance_loss_mlp": 1.02404189,
      "epoch": 0.2701857752660374,
      "flos": 18186636493440.0,
      "grad_norm": 2.263601059370973,
      "language_loss": 0.8263495,
      "learning_rate": 3.424913585441893e-06,
      "loss": 0.84876263,
      "num_input_tokens_seen": 48216000,
      "step": 2247,
      "time_per_iteration": 2.6656646728515625
    },
    {
      "auxiliary_loss_clip": 0.01202409,
      "auxiliary_loss_mlp": 0.01029354,
      "balance_loss_clip": 1.05775762,
      "balance_loss_mlp": 1.01988244,
      "epoch": 0.2703060181566765,
      "flos": 16319496648960.0,
      "grad_norm": 2.5995535058915897,
      "language_loss": 0.87143493,
      "learning_rate": 3.4243668605226374e-06,
      "loss": 0.89375257,
      "num_input_tokens_seen": 48233025,
      "step": 2248,
      "time_per_iteration": 2.613560676574707
    },
    {
      "auxiliary_loss_clip": 0.01209363,
      "auxiliary_loss_mlp": 0.01032619,
      "balance_loss_clip": 1.05440617,
      "balance_loss_mlp": 1.02317798,
      "epoch": 0.2704262610473156,
      "flos": 19572357329280.0,
      "grad_norm": 2.358045515496185,
      "language_loss": 0.82583934,
      "learning_rate": 3.423819919528061e-06,
      "loss": 0.84825909,
      "num_input_tokens_seen": 48251110,
      "step": 2249,
      "time_per_iteration": 2.6945273876190186
    },
    {
      "auxiliary_loss_clip": 0.01217735,
      "auxiliary_loss_mlp": 0.01030508,
      "balance_loss_clip": 1.05046868,
      "balance_loss_mlp": 1.02037501,
      "epoch": 0.27054650393795465,
      "flos": 20740746925440.0,
      "grad_norm": 1.8261909308915862,
      "language_loss": 0.780707,
      "learning_rate": 3.4232727625411355e-06,
      "loss": 0.80318946,
      "num_input_tokens_seen": 48270215,
      "step": 2250,
      "time_per_iteration": 2.8075826168060303
    },
    {
      "auxiliary_loss_clip": 0.01204808,
      "auxiliary_loss_mlp": 0.01024634,
      "balance_loss_clip": 1.04656649,
      "balance_loss_mlp": 1.01618218,
      "epoch": 0.27066674682859376,
      "flos": 18658322916480.0,
      "grad_norm": 1.6052982341867916,
      "language_loss": 0.86192805,
      "learning_rate": 3.4227253896448626e-06,
      "loss": 0.88422245,
      "num_input_tokens_seen": 48288075,
      "step": 2251,
      "time_per_iteration": 2.673701286315918
    },
    {
      "auxiliary_loss_clip": 0.01205192,
      "auxiliary_loss_mlp": 0.01031107,
      "balance_loss_clip": 1.05929089,
      "balance_loss_mlp": 1.02202892,
      "epoch": 0.2707869897192329,
      "flos": 23002759958400.0,
      "grad_norm": 3.1947211981815546,
      "language_loss": 0.82009906,
      "learning_rate": 3.42217780092228e-06,
      "loss": 0.84246206,
      "num_input_tokens_seen": 48306415,
      "step": 2252,
      "time_per_iteration": 2.7869133949279785
    },
    {
      "auxiliary_loss_clip": 0.01126428,
      "auxiliary_loss_mlp": 0.0101804,
      "balance_loss_clip": 1.02345538,
      "balance_loss_mlp": 1.01649022,
      "epoch": 0.27090723260987193,
      "flos": 58323240293760.0,
      "grad_norm": 0.7956347631769961,
      "language_loss": 0.60375732,
      "learning_rate": 3.421629996456456e-06,
      "loss": 0.62520206,
      "num_input_tokens_seen": 48365035,
      "step": 2253,
      "time_per_iteration": 3.1415762901306152
    },
    {
      "auxiliary_loss_clip": 0.0120316,
      "auxiliary_loss_mlp": 0.01035438,
      "balance_loss_clip": 1.05418849,
      "balance_loss_mlp": 1.0254128,
      "epoch": 0.27102747550051104,
      "flos": 11984540797440.0,
      "grad_norm": 1.920103540925317,
      "language_loss": 0.82566881,
      "learning_rate": 3.421081976330491e-06,
      "loss": 0.84805477,
      "num_input_tokens_seen": 48383550,
      "step": 2254,
      "time_per_iteration": 2.6229162216186523
    },
    {
      "auxiliary_loss_clip": 0.01202085,
      "auxiliary_loss_mlp": 0.01033021,
      "balance_loss_clip": 1.05206561,
      "balance_loss_mlp": 1.02353811,
      "epoch": 0.27114771839115015,
      "flos": 19900401264000.0,
      "grad_norm": 2.1829578159932597,
      "language_loss": 0.87954533,
      "learning_rate": 3.4205337406275207e-06,
      "loss": 0.90189636,
      "num_input_tokens_seen": 48403670,
      "step": 2255,
      "time_per_iteration": 2.655954360961914
    },
    {
      "auxiliary_loss_clip": 0.0120205,
      "auxiliary_loss_mlp": 0.01025572,
      "balance_loss_clip": 1.05730975,
      "balance_loss_mlp": 1.01661921,
      "epoch": 0.2712679612817892,
      "flos": 18331966920960.0,
      "grad_norm": 2.9367023371744123,
      "language_loss": 0.75583118,
      "learning_rate": 3.4199852894307114e-06,
      "loss": 0.7781074,
      "num_input_tokens_seen": 48420420,
      "step": 2256,
      "time_per_iteration": 2.643710136413574
    },
    {
      "auxiliary_loss_clip": 0.01217269,
      "auxiliary_loss_mlp": 0.0103381,
      "balance_loss_clip": 1.05256188,
      "balance_loss_mlp": 1.02453566,
      "epoch": 0.2713882041724283,
      "flos": 24460302038400.0,
      "grad_norm": 1.9748010178939268,
      "language_loss": 0.78610313,
      "learning_rate": 3.419436622823262e-06,
      "loss": 0.8086139,
      "num_input_tokens_seen": 48441140,
      "step": 2257,
      "time_per_iteration": 3.7398903369903564
    },
    {
      "auxiliary_loss_clip": 0.01206683,
      "auxiliary_loss_mlp": 0.01032403,
      "balance_loss_clip": 1.0568285,
      "balance_loss_mlp": 1.02331972,
      "epoch": 0.27150844706306737,
      "flos": 23039317025280.0,
      "grad_norm": 2.7725630798308134,
      "language_loss": 0.73987854,
      "learning_rate": 3.4188877408884063e-06,
      "loss": 0.76226938,
      "num_input_tokens_seen": 48461845,
      "step": 2258,
      "time_per_iteration": 2.662987470626831
    },
    {
      "auxiliary_loss_clip": 0.0120367,
      "auxiliary_loss_mlp": 0.01037926,
      "balance_loss_clip": 1.05462468,
      "balance_loss_mlp": 1.02813911,
      "epoch": 0.2716286899537065,
      "flos": 22563644192640.0,
      "grad_norm": 6.763782753254718,
      "language_loss": 0.65384817,
      "learning_rate": 3.4183386437094088e-06,
      "loss": 0.67626417,
      "num_input_tokens_seen": 48478510,
      "step": 2259,
      "time_per_iteration": 3.6681859493255615
    },
    {
      "auxiliary_loss_clip": 0.01208379,
      "auxiliary_loss_mlp": 0.01028563,
      "balance_loss_clip": 1.05308127,
      "balance_loss_mlp": 1.01974189,
      "epoch": 0.2717489328443456,
      "flos": 13115044523520.0,
      "grad_norm": 3.277668626010711,
      "language_loss": 0.82148492,
      "learning_rate": 3.417789331369565e-06,
      "loss": 0.84385431,
      "num_input_tokens_seen": 48494300,
      "step": 2260,
      "time_per_iteration": 2.6094467639923096
    },
    {
      "auxiliary_loss_clip": 0.01207423,
      "auxiliary_loss_mlp": 0.01034071,
      "balance_loss_clip": 1.06017923,
      "balance_loss_mlp": 1.02423632,
      "epoch": 0.27186917573498465,
      "flos": 29278688060160.0,
      "grad_norm": 2.0674236638107133,
      "language_loss": 0.91437113,
      "learning_rate": 3.4172398039522088e-06,
      "loss": 0.93678606,
      "num_input_tokens_seen": 48515585,
      "step": 2261,
      "time_per_iteration": 2.6816599369049072
    },
    {
      "auxiliary_loss_clip": 0.01206012,
      "auxiliary_loss_mlp": 0.01026803,
      "balance_loss_clip": 1.05656922,
      "balance_loss_mlp": 1.0178802,
      "epoch": 0.27198941862562376,
      "flos": 26032220000640.0,
      "grad_norm": 2.645372090871193,
      "language_loss": 0.79824418,
      "learning_rate": 3.4166900615407e-06,
      "loss": 0.82057232,
      "num_input_tokens_seen": 48533500,
      "step": 2262,
      "time_per_iteration": 2.6537764072418213
    },
    {
      "auxiliary_loss_clip": 0.01206375,
      "auxiliary_loss_mlp": 0.01030284,
      "balance_loss_clip": 1.05805087,
      "balance_loss_mlp": 1.02107489,
      "epoch": 0.27210966151626287,
      "flos": 32780983760640.0,
      "grad_norm": 2.288030148454197,
      "language_loss": 0.74980694,
      "learning_rate": 3.416140104218436e-06,
      "loss": 0.77217352,
      "num_input_tokens_seen": 48552865,
      "step": 2263,
      "time_per_iteration": 3.6863489151000977
    },
    {
      "auxiliary_loss_clip": 0.01114703,
      "auxiliary_loss_mlp": 0.01200149,
      "balance_loss_clip": 1.02424073,
      "balance_loss_mlp": 1.00028598,
      "epoch": 0.2722299044069019,
      "flos": 65471043219840.0,
      "grad_norm": 0.8427850176370173,
      "language_loss": 0.69659054,
      "learning_rate": 3.4155899320688437e-06,
      "loss": 0.71973908,
      "num_input_tokens_seen": 48618940,
      "step": 2264,
      "time_per_iteration": 3.2656333446502686
    },
    {
      "auxiliary_loss_clip": 0.01215303,
      "auxiliary_loss_mlp": 0.01028455,
      "balance_loss_clip": 1.05188262,
      "balance_loss_mlp": 1.01854885,
      "epoch": 0.27235014729754103,
      "flos": 15334143782400.0,
      "grad_norm": 2.668888679729057,
      "language_loss": 0.74329054,
      "learning_rate": 3.415039545175384e-06,
      "loss": 0.76572812,
      "num_input_tokens_seen": 48634665,
      "step": 2265,
      "time_per_iteration": 2.676732063293457
    },
    {
      "auxiliary_loss_clip": 0.01207805,
      "auxiliary_loss_mlp": 0.01031223,
      "balance_loss_clip": 1.05737388,
      "balance_loss_mlp": 1.02218068,
      "epoch": 0.27247039018818014,
      "flos": 21872363973120.0,
      "grad_norm": 2.3439439691046697,
      "language_loss": 0.64925587,
      "learning_rate": 3.414488943621551e-06,
      "loss": 0.67164618,
      "num_input_tokens_seen": 48653330,
      "step": 2266,
      "time_per_iteration": 3.5333662033081055
    },
    {
      "auxiliary_loss_clip": 0.01203259,
      "auxiliary_loss_mlp": 0.01033463,
      "balance_loss_clip": 1.0570457,
      "balance_loss_mlp": 1.02411103,
      "epoch": 0.2725906330788192,
      "flos": 18695490514560.0,
      "grad_norm": 2.0653405361110733,
      "language_loss": 0.74286079,
      "learning_rate": 3.41393812749087e-06,
      "loss": 0.76522803,
      "num_input_tokens_seen": 48671375,
      "step": 2267,
      "time_per_iteration": 2.6636340618133545
    },
    {
      "auxiliary_loss_clip": 0.01206675,
      "auxiliary_loss_mlp": 0.01034733,
      "balance_loss_clip": 1.05686426,
      "balance_loss_mlp": 1.02541041,
      "epoch": 0.2727108759694583,
      "flos": 17886099398400.0,
      "grad_norm": 3.2869509449785115,
      "language_loss": 0.71562326,
      "learning_rate": 3.4133870968668984e-06,
      "loss": 0.73803735,
      "num_input_tokens_seen": 48686175,
      "step": 2268,
      "time_per_iteration": 2.6246178150177
    },
    {
      "auxiliary_loss_clip": 0.01211264,
      "auxiliary_loss_mlp": 0.01031925,
      "balance_loss_clip": 1.05670929,
      "balance_loss_mlp": 1.02295446,
      "epoch": 0.2728311188600974,
      "flos": 24461666755200.0,
      "grad_norm": 2.3402076466508386,
      "language_loss": 0.78863919,
      "learning_rate": 3.412835851833229e-06,
      "loss": 0.81107104,
      "num_input_tokens_seen": 48708370,
      "step": 2269,
      "time_per_iteration": 2.9154610633850098
    },
    {
      "auxiliary_loss_clip": 0.01205542,
      "auxiliary_loss_mlp": 0.0103073,
      "balance_loss_clip": 1.06056571,
      "balance_loss_mlp": 1.02140224,
      "epoch": 0.2729513617507365,
      "flos": 30993314757120.0,
      "grad_norm": 2.2094011011695773,
      "language_loss": 0.77938354,
      "learning_rate": 3.4122843924734834e-06,
      "loss": 0.80174631,
      "num_input_tokens_seen": 48730670,
      "step": 2270,
      "time_per_iteration": 2.6788277626037598
    },
    {
      "auxiliary_loss_clip": 0.01204581,
      "auxiliary_loss_mlp": 0.01033821,
      "balance_loss_clip": 1.05616689,
      "balance_loss_mlp": 1.02401578,
      "epoch": 0.2730716046413756,
      "flos": 19094637421440.0,
      "grad_norm": 1.8702180601453968,
      "language_loss": 0.88053465,
      "learning_rate": 3.411732718871319e-06,
      "loss": 0.9029187,
      "num_input_tokens_seen": 48746510,
      "step": 2271,
      "time_per_iteration": 2.696925640106201
    },
    {
      "auxiliary_loss_clip": 0.01202797,
      "auxiliary_loss_mlp": 0.01032991,
      "balance_loss_clip": 1.06036806,
      "balance_loss_mlp": 1.02414608,
      "epoch": 0.27319184753201464,
      "flos": 26944566474240.0,
      "grad_norm": 2.0559513618590866,
      "language_loss": 0.78735906,
      "learning_rate": 3.4111808311104227e-06,
      "loss": 0.80971694,
      "num_input_tokens_seen": 48768825,
      "step": 2272,
      "time_per_iteration": 2.659564971923828
    },
    {
      "auxiliary_loss_clip": 0.01212444,
      "auxiliary_loss_mlp": 0.01031351,
      "balance_loss_clip": 1.05314243,
      "balance_loss_mlp": 1.02134383,
      "epoch": 0.27331209042265375,
      "flos": 31759828012800.0,
      "grad_norm": 1.8385091274012835,
      "language_loss": 0.69730741,
      "learning_rate": 3.410628729274517e-06,
      "loss": 0.71974528,
      "num_input_tokens_seen": 48790345,
      "step": 2273,
      "time_per_iteration": 2.7441186904907227
    },
    {
      "auxiliary_loss_clip": 0.01203049,
      "auxiliary_loss_mlp": 0.01208717,
      "balance_loss_clip": 1.0547359,
      "balance_loss_mlp": 1.00052047,
      "epoch": 0.27343233331329286,
      "flos": 25739081107200.0,
      "grad_norm": 2.219625791832592,
      "language_loss": 0.82547325,
      "learning_rate": 3.4100764134473546e-06,
      "loss": 0.8495909,
      "num_input_tokens_seen": 48809630,
      "step": 2274,
      "time_per_iteration": 2.7034523487091064
    },
    {
      "auxiliary_loss_clip": 0.01204873,
      "auxiliary_loss_mlp": 0.01032939,
      "balance_loss_clip": 1.06062365,
      "balance_loss_mlp": 1.02388525,
      "epoch": 0.2735525762039319,
      "flos": 24389414547840.0,
      "grad_norm": 3.1568704858832093,
      "language_loss": 0.84715819,
      "learning_rate": 3.4095238837127215e-06,
      "loss": 0.8695364,
      "num_input_tokens_seen": 48828770,
      "step": 2275,
      "time_per_iteration": 2.6639511585235596
    },
    {
      "auxiliary_loss_clip": 0.01207026,
      "auxiliary_loss_mlp": 0.01025599,
      "balance_loss_clip": 1.05325985,
      "balance_loss_mlp": 1.0166049,
      "epoch": 0.27367281909457103,
      "flos": 14465357527680.0,
      "grad_norm": 2.1055555393656906,
      "language_loss": 0.79425001,
      "learning_rate": 3.4089711401544355e-06,
      "loss": 0.8165763,
      "num_input_tokens_seen": 48846365,
      "step": 2276,
      "time_per_iteration": 2.7015254497528076
    },
    {
      "auxiliary_loss_clip": 0.01202784,
      "auxiliary_loss_mlp": 0.01027948,
      "balance_loss_clip": 1.05364692,
      "balance_loss_mlp": 1.01876879,
      "epoch": 0.27379306198521014,
      "flos": 23476996247040.0,
      "grad_norm": 2.7407826755460554,
      "language_loss": 0.67466503,
      "learning_rate": 3.4084181828563486e-06,
      "loss": 0.69697237,
      "num_input_tokens_seen": 48863085,
      "step": 2277,
      "time_per_iteration": 2.660465717315674
    },
    {
      "auxiliary_loss_clip": 0.012102,
      "auxiliary_loss_mlp": 0.01029409,
      "balance_loss_clip": 1.05138683,
      "balance_loss_mlp": 1.0207963,
      "epoch": 0.2739133048758492,
      "flos": 17458152762240.0,
      "grad_norm": 1.7211382887142552,
      "language_loss": 0.70908999,
      "learning_rate": 3.4078650119023428e-06,
      "loss": 0.73148608,
      "num_input_tokens_seen": 48881400,
      "step": 2278,
      "time_per_iteration": 2.7182631492614746
    },
    {
      "auxiliary_loss_clip": 0.01214902,
      "auxiliary_loss_mlp": 0.01034032,
      "balance_loss_clip": 1.04713655,
      "balance_loss_mlp": 1.02412534,
      "epoch": 0.2740335477664883,
      "flos": 19273113123840.0,
      "grad_norm": 4.23113535375748,
      "language_loss": 0.74062872,
      "learning_rate": 3.4073116273763337e-06,
      "loss": 0.76311803,
      "num_input_tokens_seen": 48895845,
      "step": 2279,
      "time_per_iteration": 2.749884605407715
    },
    {
      "auxiliary_loss_clip": 0.01211301,
      "auxiliary_loss_mlp": 0.01034026,
      "balance_loss_clip": 1.05399394,
      "balance_loss_mlp": 1.02447128,
      "epoch": 0.2741537906571274,
      "flos": 26104723603200.0,
      "grad_norm": 1.9627500260676967,
      "language_loss": 0.8150965,
      "learning_rate": 3.40675802936227e-06,
      "loss": 0.83754969,
      "num_input_tokens_seen": 48916630,
      "step": 2280,
      "time_per_iteration": 2.7284204959869385
    },
    {
      "auxiliary_loss_clip": 0.01200726,
      "auxiliary_loss_mlp": 0.01034793,
      "balance_loss_clip": 1.05589771,
      "balance_loss_mlp": 1.02474928,
      "epoch": 0.27427403354776647,
      "flos": 34164190644480.0,
      "grad_norm": 2.5477473062375315,
      "language_loss": 0.72614592,
      "learning_rate": 3.4062042179441318e-06,
      "loss": 0.74850106,
      "num_input_tokens_seen": 48937100,
      "step": 2281,
      "time_per_iteration": 2.7308475971221924
    },
    {
      "auxiliary_loss_clip": 0.01200933,
      "auxiliary_loss_mlp": 0.01022539,
      "balance_loss_clip": 1.05693495,
      "balance_loss_mlp": 1.01412344,
      "epoch": 0.2743942764384056,
      "flos": 18766988536320.0,
      "grad_norm": 2.3582361735890496,
      "language_loss": 0.80568671,
      "learning_rate": 3.4056501932059314e-06,
      "loss": 0.82792145,
      "num_input_tokens_seen": 48955175,
      "step": 2282,
      "time_per_iteration": 2.8093881607055664
    },
    {
      "auxiliary_loss_clip": 0.01115579,
      "auxiliary_loss_mlp": 0.010057,
      "balance_loss_clip": 1.03394556,
      "balance_loss_mlp": 1.00420439,
      "epoch": 0.2745145193290447,
      "flos": 64904048058240.0,
      "grad_norm": 0.7742631533204739,
      "language_loss": 0.58122075,
      "learning_rate": 3.405095955231715e-06,
      "loss": 0.60243356,
      "num_input_tokens_seen": 49006830,
      "step": 2283,
      "time_per_iteration": 3.1113600730895996
    },
    {
      "auxiliary_loss_clip": 0.0120805,
      "auxiliary_loss_mlp": 0.01028143,
      "balance_loss_clip": 1.05696392,
      "balance_loss_mlp": 1.01904154,
      "epoch": 0.27463476221968375,
      "flos": 16136926796160.0,
      "grad_norm": 2.5122355339295988,
      "language_loss": 0.94476569,
      "learning_rate": 3.4045415041055585e-06,
      "loss": 0.96712762,
      "num_input_tokens_seen": 49022470,
      "step": 2284,
      "time_per_iteration": 3.5802080631256104
    },
    {
      "auxiliary_loss_clip": 0.01211549,
      "auxiliary_loss_mlp": 0.01032904,
      "balance_loss_clip": 1.05678868,
      "balance_loss_mlp": 1.02314043,
      "epoch": 0.27475500511032286,
      "flos": 10376712213120.0,
      "grad_norm": 2.821394446109543,
      "language_loss": 0.78363889,
      "learning_rate": 3.4039868399115728e-06,
      "loss": 0.80608344,
      "num_input_tokens_seen": 49037110,
      "step": 2285,
      "time_per_iteration": 2.7072203159332275
    },
    {
      "auxiliary_loss_clip": 0.01218917,
      "auxiliary_loss_mlp": 0.01030325,
      "balance_loss_clip": 1.053967,
      "balance_loss_mlp": 1.02127147,
      "epoch": 0.27487524800096197,
      "flos": 17311062568320.0,
      "grad_norm": 2.0277602915448454,
      "language_loss": 0.80732363,
      "learning_rate": 3.4034319627339003e-06,
      "loss": 0.82981598,
      "num_input_tokens_seen": 49053975,
      "step": 2286,
      "time_per_iteration": 3.666003704071045
    },
    {
      "auxiliary_loss_clip": 0.01207712,
      "auxiliary_loss_mlp": 0.01032067,
      "balance_loss_clip": 1.05632806,
      "balance_loss_mlp": 1.02269101,
      "epoch": 0.274995490891601,
      "flos": 27120205002240.0,
      "grad_norm": 2.4808623250912554,
      "language_loss": 0.69326937,
      "learning_rate": 3.402876872656715e-06,
      "loss": 0.71566713,
      "num_input_tokens_seen": 49072295,
      "step": 2287,
      "time_per_iteration": 2.7156219482421875
    },
    {
      "auxiliary_loss_clip": 0.0120544,
      "auxiliary_loss_mlp": 0.01032927,
      "balance_loss_clip": 1.05695462,
      "balance_loss_mlp": 1.02413547,
      "epoch": 0.27511573378224013,
      "flos": 23436093634560.0,
      "grad_norm": 2.166571616574906,
      "language_loss": 0.8987518,
      "learning_rate": 3.402321569764223e-06,
      "loss": 0.92113543,
      "num_input_tokens_seen": 49091600,
      "step": 2288,
      "time_per_iteration": 2.787581443786621
    },
    {
      "auxiliary_loss_clip": 0.01217005,
      "auxiliary_loss_mlp": 0.01209429,
      "balance_loss_clip": 1.05300641,
      "balance_loss_mlp": 1.00046337,
      "epoch": 0.2752359766728792,
      "flos": 16722019434240.0,
      "grad_norm": 5.259867277398229,
      "language_loss": 0.83760864,
      "learning_rate": 3.4017660541406635e-06,
      "loss": 0.86187291,
      "num_input_tokens_seen": 49107665,
      "step": 2289,
      "time_per_iteration": 2.666562557220459
    },
    {
      "auxiliary_loss_clip": 0.0121398,
      "auxiliary_loss_mlp": 0.01026783,
      "balance_loss_clip": 1.05429447,
      "balance_loss_mlp": 1.01784813,
      "epoch": 0.2753562195635183,
      "flos": 25297738698240.0,
      "grad_norm": 6.585304128972324,
      "language_loss": 0.74353319,
      "learning_rate": 3.4012103258703092e-06,
      "loss": 0.76594079,
      "num_input_tokens_seen": 49126420,
      "step": 2290,
      "time_per_iteration": 3.75496244430542
    },
    {
      "auxiliary_loss_clip": 0.01209951,
      "auxiliary_loss_mlp": 0.01026028,
      "balance_loss_clip": 1.05346048,
      "balance_loss_mlp": 1.01684928,
      "epoch": 0.2754764624541574,
      "flos": 27338972785920.0,
      "grad_norm": 2.265133800748376,
      "language_loss": 0.83182514,
      "learning_rate": 3.4006543850374616e-06,
      "loss": 0.85418499,
      "num_input_tokens_seen": 49141470,
      "step": 2291,
      "time_per_iteration": 2.674125909805298
    },
    {
      "auxiliary_loss_clip": 0.0120638,
      "auxiliary_loss_mlp": 0.01031578,
      "balance_loss_clip": 1.05487967,
      "balance_loss_mlp": 1.02298951,
      "epoch": 0.27559670534479647,
      "flos": 17238379397760.0,
      "grad_norm": 1.988735428177295,
      "language_loss": 0.75114733,
      "learning_rate": 3.400098231726458e-06,
      "loss": 0.77352691,
      "num_input_tokens_seen": 49158570,
      "step": 2292,
      "time_per_iteration": 3.545905590057373
    },
    {
      "auxiliary_loss_clip": 0.01213691,
      "auxiliary_loss_mlp": 0.0103269,
      "balance_loss_clip": 1.05307615,
      "balance_loss_mlp": 1.02315307,
      "epoch": 0.2757169482354356,
      "flos": 21939085486080.0,
      "grad_norm": 11.764757980584932,
      "language_loss": 0.87135518,
      "learning_rate": 3.3995418660216657e-06,
      "loss": 0.89381897,
      "num_input_tokens_seen": 49176025,
      "step": 2293,
      "time_per_iteration": 2.711564779281616
    },
    {
      "auxiliary_loss_clip": 0.01210298,
      "auxiliary_loss_mlp": 0.0103211,
      "balance_loss_clip": 1.06188059,
      "balance_loss_mlp": 1.02260888,
      "epoch": 0.2758371911260747,
      "flos": 20850669521280.0,
      "grad_norm": 1.8234603449731626,
      "language_loss": 0.8028928,
      "learning_rate": 3.3989852880074848e-06,
      "loss": 0.82531685,
      "num_input_tokens_seen": 49197455,
      "step": 2294,
      "time_per_iteration": 2.8100733757019043
    },
    {
      "auxiliary_loss_clip": 0.01119226,
      "auxiliary_loss_mlp": 0.01011438,
      "balance_loss_clip": 1.03365183,
      "balance_loss_mlp": 1.00985289,
      "epoch": 0.27595743401671374,
      "flos": 69269063592960.0,
      "grad_norm": 0.7512880349326545,
      "language_loss": 0.60563242,
      "learning_rate": 3.398428497768348e-06,
      "loss": 0.62693906,
      "num_input_tokens_seen": 49262625,
      "step": 2295,
      "time_per_iteration": 3.303314685821533
    },
    {
      "auxiliary_loss_clip": 0.01216277,
      "auxiliary_loss_mlp": 0.01026865,
      "balance_loss_clip": 1.05140829,
      "balance_loss_mlp": 1.0175488,
      "epoch": 0.27607767690735285,
      "flos": 21215019127680.0,
      "grad_norm": 1.7683194537754774,
      "language_loss": 0.72327983,
      "learning_rate": 3.3978714953887205e-06,
      "loss": 0.74571133,
      "num_input_tokens_seen": 49282380,
      "step": 2296,
      "time_per_iteration": 2.6359951496124268
    },
    {
      "auxiliary_loss_clip": 0.01197469,
      "auxiliary_loss_mlp": 0.01025533,
      "balance_loss_clip": 1.04730034,
      "balance_loss_mlp": 1.01635361,
      "epoch": 0.27619791979799196,
      "flos": 24825334003200.0,
      "grad_norm": 1.786929052759254,
      "language_loss": 0.86074567,
      "learning_rate": 3.397314280953098e-06,
      "loss": 0.8829757,
      "num_input_tokens_seen": 49303205,
      "step": 2297,
      "time_per_iteration": 2.7472195625305176
    },
    {
      "auxiliary_loss_clip": 0.01199847,
      "auxiliary_loss_mlp": 0.01023649,
      "balance_loss_clip": 1.05352616,
      "balance_loss_mlp": 1.01510167,
      "epoch": 0.276318162688631,
      "flos": 24753548672640.0,
      "grad_norm": 2.0811803398492326,
      "language_loss": 0.80600381,
      "learning_rate": 3.3967568545460108e-06,
      "loss": 0.82823873,
      "num_input_tokens_seen": 49322745,
      "step": 2298,
      "time_per_iteration": 2.692678213119507
    },
    {
      "auxiliary_loss_clip": 0.01201404,
      "auxiliary_loss_mlp": 0.01028796,
      "balance_loss_clip": 1.05739617,
      "balance_loss_mlp": 1.01946759,
      "epoch": 0.27643840557927013,
      "flos": 18150007599360.0,
      "grad_norm": 3.680917906645582,
      "language_loss": 0.80994463,
      "learning_rate": 3.3961992162520185e-06,
      "loss": 0.83224666,
      "num_input_tokens_seen": 49341370,
      "step": 2299,
      "time_per_iteration": 2.6632840633392334
    },
    {
      "auxiliary_loss_clip": 0.01205224,
      "auxiliary_loss_mlp": 0.01027232,
      "balance_loss_clip": 1.05723381,
      "balance_loss_mlp": 1.01773119,
      "epoch": 0.27655864846990924,
      "flos": 24823933372800.0,
      "grad_norm": 2.1292971661599487,
      "language_loss": 0.72135574,
      "learning_rate": 3.3956413661557156e-06,
      "loss": 0.7436803,
      "num_input_tokens_seen": 49361545,
      "step": 2300,
      "time_per_iteration": 2.656773805618286
    },
    {
      "auxiliary_loss_clip": 0.01215988,
      "auxiliary_loss_mlp": 0.01028767,
      "balance_loss_clip": 1.05363584,
      "balance_loss_mlp": 1.01983213,
      "epoch": 0.2766788913605483,
      "flos": 20266582464000.0,
      "grad_norm": 2.6956980004111752,
      "language_loss": 0.6598596,
      "learning_rate": 3.3950833043417273e-06,
      "loss": 0.68230712,
      "num_input_tokens_seen": 49379690,
      "step": 2301,
      "time_per_iteration": 2.693408966064453
    },
    {
      "auxiliary_loss_clip": 0.01208763,
      "auxiliary_loss_mlp": 0.01027188,
      "balance_loss_clip": 1.05992913,
      "balance_loss_mlp": 1.01765704,
      "epoch": 0.2767991342511874,
      "flos": 21470272151040.0,
      "grad_norm": 2.0507045492765057,
      "language_loss": 0.72830385,
      "learning_rate": 3.3945250308947105e-06,
      "loss": 0.75066334,
      "num_input_tokens_seen": 49395995,
      "step": 2302,
      "time_per_iteration": 2.6199254989624023
    },
    {
      "auxiliary_loss_clip": 0.01114042,
      "auxiliary_loss_mlp": 0.01004223,
      "balance_loss_clip": 1.02582324,
      "balance_loss_mlp": 1.00281644,
      "epoch": 0.2769193771418265,
      "flos": 66002627571840.0,
      "grad_norm": 1.2641225705365,
      "language_loss": 0.68313062,
      "learning_rate": 3.3939665458993556e-06,
      "loss": 0.70431328,
      "num_input_tokens_seen": 49450415,
      "step": 2303,
      "time_per_iteration": 3.127807378768921
    },
    {
      "auxiliary_loss_clip": 0.01212077,
      "auxiliary_loss_mlp": 0.01032049,
      "balance_loss_clip": 1.0515008,
      "balance_loss_mlp": 1.02311409,
      "epoch": 0.27703962003246557,
      "flos": 20704441253760.0,
      "grad_norm": 1.9819366383965613,
      "language_loss": 0.76718795,
      "learning_rate": 3.3934078494403843e-06,
      "loss": 0.78962928,
      "num_input_tokens_seen": 49469990,
      "step": 2304,
      "time_per_iteration": 2.745051145553589
    },
    {
      "auxiliary_loss_clip": 0.01210295,
      "auxiliary_loss_mlp": 0.01209023,
      "balance_loss_clip": 1.04695642,
      "balance_loss_mlp": 1.00037038,
      "epoch": 0.2771598629231047,
      "flos": 22929897219840.0,
      "grad_norm": 1.9945351414742056,
      "language_loss": 0.81067324,
      "learning_rate": 3.3928489416025495e-06,
      "loss": 0.83486646,
      "num_input_tokens_seen": 49490835,
      "step": 2305,
      "time_per_iteration": 2.7718443870544434
    },
    {
      "auxiliary_loss_clip": 0.01205132,
      "auxiliary_loss_mlp": 0.01039189,
      "balance_loss_clip": 1.05541265,
      "balance_loss_mlp": 1.02954507,
      "epoch": 0.27728010581374374,
      "flos": 18369457741440.0,
      "grad_norm": 2.178391732886237,
      "language_loss": 0.79120886,
      "learning_rate": 3.392289822470638e-06,
      "loss": 0.8136521,
      "num_input_tokens_seen": 49508815,
      "step": 2306,
      "time_per_iteration": 2.67230224609375
    },
    {
      "auxiliary_loss_clip": 0.01203107,
      "auxiliary_loss_mlp": 0.0102372,
      "balance_loss_clip": 1.05329895,
      "balance_loss_mlp": 1.0144093,
      "epoch": 0.27740034870438285,
      "flos": 19427637432960.0,
      "grad_norm": 2.087033556912192,
      "language_loss": 0.76074016,
      "learning_rate": 3.3917304921294674e-06,
      "loss": 0.78300846,
      "num_input_tokens_seen": 49526980,
      "step": 2307,
      "time_per_iteration": 2.6302924156188965
    },
    {
      "auxiliary_loss_clip": 0.01203784,
      "auxiliary_loss_mlp": 0.01032746,
      "balance_loss_clip": 1.05528593,
      "balance_loss_mlp": 1.02346587,
      "epoch": 0.27752059159502196,
      "flos": 21614776565760.0,
      "grad_norm": 1.7248587660833634,
      "language_loss": 0.80786592,
      "learning_rate": 3.3911709506638876e-06,
      "loss": 0.83023119,
      "num_input_tokens_seen": 49546290,
      "step": 2308,
      "time_per_iteration": 2.675165891647339
    },
    {
      "auxiliary_loss_clip": 0.01196582,
      "auxiliary_loss_mlp": 0.01208716,
      "balance_loss_clip": 1.04655707,
      "balance_loss_mlp": 1.00028038,
      "epoch": 0.277640834485661,
      "flos": 26608011016320.0,
      "grad_norm": 2.2617154875427494,
      "language_loss": 0.81598747,
      "learning_rate": 3.390611198158781e-06,
      "loss": 0.84004045,
      "num_input_tokens_seen": 49564165,
      "step": 2309,
      "time_per_iteration": 2.747323989868164
    },
    {
      "auxiliary_loss_clip": 0.01207641,
      "auxiliary_loss_mlp": 0.01032035,
      "balance_loss_clip": 1.06077015,
      "balance_loss_mlp": 1.02317774,
      "epoch": 0.2777610773763001,
      "flos": 19492814661120.0,
      "grad_norm": 2.635556093138945,
      "language_loss": 0.90175307,
      "learning_rate": 3.3900512346990612e-06,
      "loss": 0.92414975,
      "num_input_tokens_seen": 49580155,
      "step": 2310,
      "time_per_iteration": 2.7167131900787354
    },
    {
      "auxiliary_loss_clip": 0.01209766,
      "auxiliary_loss_mlp": 0.01029056,
      "balance_loss_clip": 1.04656088,
      "balance_loss_mlp": 1.01913202,
      "epoch": 0.27788132026693924,
      "flos": 38290650001920.0,
      "grad_norm": 2.454698403985487,
      "language_loss": 0.66119003,
      "learning_rate": 3.389491060369674e-06,
      "loss": 0.68357825,
      "num_input_tokens_seen": 49605830,
      "step": 2311,
      "time_per_iteration": 3.790994167327881
    },
    {
      "auxiliary_loss_clip": 0.01201297,
      "auxiliary_loss_mlp": 0.01022072,
      "balance_loss_clip": 1.04902947,
      "balance_loss_mlp": 1.01351869,
      "epoch": 0.2780015631575783,
      "flos": 22382546797440.0,
      "grad_norm": 2.0702650642547797,
      "language_loss": 0.8978821,
      "learning_rate": 3.388930675255598e-06,
      "loss": 0.92011577,
      "num_input_tokens_seen": 49625680,
      "step": 2312,
      "time_per_iteration": 3.631479263305664
    },
    {
      "auxiliary_loss_clip": 0.01210034,
      "auxiliary_loss_mlp": 0.01032704,
      "balance_loss_clip": 1.05386233,
      "balance_loss_mlp": 1.0230124,
      "epoch": 0.2781218060482174,
      "flos": 12203200840320.0,
      "grad_norm": 2.329728859320366,
      "language_loss": 0.79683918,
      "learning_rate": 3.388370079441843e-06,
      "loss": 0.81926656,
      "num_input_tokens_seen": 49641195,
      "step": 2313,
      "time_per_iteration": 2.669644355773926
    },
    {
      "auxiliary_loss_clip": 0.01210078,
      "auxiliary_loss_mlp": 0.010382,
      "balance_loss_clip": 1.0557605,
      "balance_loss_mlp": 1.02916396,
      "epoch": 0.2782420489388565,
      "flos": 18107632529280.0,
      "grad_norm": 2.02202048697752,
      "language_loss": 0.92753112,
      "learning_rate": 3.3878092730134505e-06,
      "loss": 0.95001388,
      "num_input_tokens_seen": 49659180,
      "step": 2314,
      "time_per_iteration": 2.7350497245788574
    },
    {
      "auxiliary_loss_clip": 0.01197564,
      "auxiliary_loss_mlp": 0.01031274,
      "balance_loss_clip": 1.05523515,
      "balance_loss_mlp": 1.02213097,
      "epoch": 0.27836229182949557,
      "flos": 18514752255360.0,
      "grad_norm": 2.1417429300656385,
      "language_loss": 0.8080917,
      "learning_rate": 3.3872482560554947e-06,
      "loss": 0.83038008,
      "num_input_tokens_seen": 49677955,
      "step": 2315,
      "time_per_iteration": 2.601330518722534
    },
    {
      "auxiliary_loss_clip": 0.01108645,
      "auxiliary_loss_mlp": 0.01003619,
      "balance_loss_clip": 1.02152157,
      "balance_loss_mlp": 1.00234318,
      "epoch": 0.2784825347201347,
      "flos": 67079230940160.0,
      "grad_norm": 0.8014092953871282,
      "language_loss": 0.56935525,
      "learning_rate": 3.386687028653082e-06,
      "loss": 0.59047788,
      "num_input_tokens_seen": 49740800,
      "step": 2316,
      "time_per_iteration": 3.167038679122925
    },
    {
      "auxiliary_loss_clip": 0.01212599,
      "auxiliary_loss_mlp": 0.01029964,
      "balance_loss_clip": 1.05333495,
      "balance_loss_mlp": 1.02077925,
      "epoch": 0.2786027776107738,
      "flos": 22631119891200.0,
      "grad_norm": 2.0121087664663344,
      "language_loss": 0.85105717,
      "learning_rate": 3.386125590891349e-06,
      "loss": 0.87348282,
      "num_input_tokens_seen": 49757675,
      "step": 2317,
      "time_per_iteration": 3.544185161590576
    },
    {
      "auxiliary_loss_clip": 0.01194341,
      "auxiliary_loss_mlp": 0.01027924,
      "balance_loss_clip": 1.05039644,
      "balance_loss_mlp": 1.01896524,
      "epoch": 0.27872302050141284,
      "flos": 15778826156160.0,
      "grad_norm": 2.9813996779026137,
      "language_loss": 0.8270548,
      "learning_rate": 3.3855639428554657e-06,
      "loss": 0.8492775,
      "num_input_tokens_seen": 49775205,
      "step": 2318,
      "time_per_iteration": 3.5438392162323
    },
    {
      "auxiliary_loss_clip": 0.01196015,
      "auxiliary_loss_mlp": 0.01027692,
      "balance_loss_clip": 1.05207455,
      "balance_loss_mlp": 1.01899612,
      "epoch": 0.27884326339205195,
      "flos": 22126970551680.0,
      "grad_norm": 2.1724082853220557,
      "language_loss": 0.80860806,
      "learning_rate": 3.385002084630635e-06,
      "loss": 0.83084512,
      "num_input_tokens_seen": 49794175,
      "step": 2319,
      "time_per_iteration": 2.7123520374298096
    },
    {
      "auxiliary_loss_clip": 0.0120908,
      "auxiliary_loss_mlp": 0.01031207,
      "balance_loss_clip": 1.05672073,
      "balance_loss_mlp": 1.02133083,
      "epoch": 0.278963506282691,
      "flos": 20558715776640.0,
      "grad_norm": 2.3029746309747097,
      "language_loss": 0.85224241,
      "learning_rate": 3.384440016302088e-06,
      "loss": 0.87464523,
      "num_input_tokens_seen": 49812850,
      "step": 2320,
      "time_per_iteration": 2.6306474208831787
    },
    {
      "auxiliary_loss_clip": 0.01199098,
      "auxiliary_loss_mlp": 0.01030024,
      "balance_loss_clip": 1.05425,
      "balance_loss_mlp": 1.02106571,
      "epoch": 0.2790837491733301,
      "flos": 21942928241280.0,
      "grad_norm": 2.0287198404048246,
      "language_loss": 0.62811351,
      "learning_rate": 3.3838777379550923e-06,
      "loss": 0.65040469,
      "num_input_tokens_seen": 49832295,
      "step": 2321,
      "time_per_iteration": 2.6923773288726807
    },
    {
      "auxiliary_loss_clip": 0.01211698,
      "auxiliary_loss_mlp": 0.01034273,
      "balance_loss_clip": 1.05736566,
      "balance_loss_mlp": 1.02502203,
      "epoch": 0.27920399206396923,
      "flos": 26286790665600.0,
      "grad_norm": 10.890287762091834,
      "language_loss": 0.78616691,
      "learning_rate": 3.383315249674944e-06,
      "loss": 0.80862665,
      "num_input_tokens_seen": 49850860,
      "step": 2322,
      "time_per_iteration": 2.6962130069732666
    },
    {
      "auxiliary_loss_clip": 0.01211839,
      "auxiliary_loss_mlp": 0.01034741,
      "balance_loss_clip": 1.05418611,
      "balance_loss_mlp": 1.02529979,
      "epoch": 0.2793242349546083,
      "flos": 25400981364480.0,
      "grad_norm": 3.7320819188756484,
      "language_loss": 0.85674763,
      "learning_rate": 3.3827525515469715e-06,
      "loss": 0.87921345,
      "num_input_tokens_seen": 49865765,
      "step": 2323,
      "time_per_iteration": 2.696962833404541
    },
    {
      "auxiliary_loss_clip": 0.01200824,
      "auxiliary_loss_mlp": 0.01037259,
      "balance_loss_clip": 1.04879665,
      "balance_loss_mlp": 1.02693546,
      "epoch": 0.2794444778452474,
      "flos": 20850346298880.0,
      "grad_norm": 2.211983651759763,
      "language_loss": 0.71386516,
      "learning_rate": 3.3821896436565367e-06,
      "loss": 0.73624599,
      "num_input_tokens_seen": 49885425,
      "step": 2324,
      "time_per_iteration": 2.824472427368164
    },
    {
      "auxiliary_loss_clip": 0.01209832,
      "auxiliary_loss_mlp": 0.0103669,
      "balance_loss_clip": 1.06076193,
      "balance_loss_mlp": 1.02726042,
      "epoch": 0.2795647207358865,
      "flos": 21576244250880.0,
      "grad_norm": 1.764047871720252,
      "language_loss": 0.7080822,
      "learning_rate": 3.381626526089032e-06,
      "loss": 0.73054743,
      "num_input_tokens_seen": 49904990,
      "step": 2325,
      "time_per_iteration": 2.72859525680542
    },
    {
      "auxiliary_loss_clip": 0.01205788,
      "auxiliary_loss_mlp": 0.01026692,
      "balance_loss_clip": 1.05373979,
      "balance_loss_mlp": 1.01702976,
      "epoch": 0.27968496362652556,
      "flos": 21471744608640.0,
      "grad_norm": 2.6738432441224482,
      "language_loss": 0.79280442,
      "learning_rate": 3.3810631989298815e-06,
      "loss": 0.81512916,
      "num_input_tokens_seen": 49924600,
      "step": 2326,
      "time_per_iteration": 2.689237117767334
    },
    {
      "auxiliary_loss_clip": 0.01218873,
      "auxiliary_loss_mlp": 0.01032652,
      "balance_loss_clip": 1.05518246,
      "balance_loss_mlp": 1.02203012,
      "epoch": 0.2798052065171647,
      "flos": 23258695340160.0,
      "grad_norm": 2.1163410014093818,
      "language_loss": 0.83947694,
      "learning_rate": 3.3804996622645423e-06,
      "loss": 0.86199224,
      "num_input_tokens_seen": 49942600,
      "step": 2327,
      "time_per_iteration": 2.7352654933929443
    },
    {
      "auxiliary_loss_clip": 0.01204691,
      "auxiliary_loss_mlp": 0.01027908,
      "balance_loss_clip": 1.05848646,
      "balance_loss_mlp": 1.01876426,
      "epoch": 0.2799254494078038,
      "flos": 21539328048000.0,
      "grad_norm": 2.21964305667161,
      "language_loss": 0.89626271,
      "learning_rate": 3.3799359161785015e-06,
      "loss": 0.9185887,
      "num_input_tokens_seen": 49962250,
      "step": 2328,
      "time_per_iteration": 2.6501190662384033
    },
    {
      "auxiliary_loss_clip": 0.01203493,
      "auxiliary_loss_mlp": 0.01032587,
      "balance_loss_clip": 1.05654597,
      "balance_loss_mlp": 1.02265716,
      "epoch": 0.28004569229844284,
      "flos": 26393912000640.0,
      "grad_norm": 16.151742830067434,
      "language_loss": 0.86095989,
      "learning_rate": 3.3793719607572798e-06,
      "loss": 0.88332069,
      "num_input_tokens_seen": 49983215,
      "step": 2329,
      "time_per_iteration": 2.647969961166382
    },
    {
      "auxiliary_loss_clip": 0.01193788,
      "auxiliary_loss_mlp": 0.01032642,
      "balance_loss_clip": 1.05296171,
      "balance_loss_mlp": 1.02333784,
      "epoch": 0.28016593518908195,
      "flos": 33547676584320.0,
      "grad_norm": 2.5227092775894424,
      "language_loss": 0.77093327,
      "learning_rate": 3.378807796086428e-06,
      "loss": 0.79319751,
      "num_input_tokens_seen": 50006075,
      "step": 2330,
      "time_per_iteration": 2.758424997329712
    },
    {
      "auxiliary_loss_clip": 0.0120889,
      "auxiliary_loss_mlp": 0.01025474,
      "balance_loss_clip": 1.06240487,
      "balance_loss_mlp": 1.01585364,
      "epoch": 0.28028617807972106,
      "flos": 15340823712000.0,
      "grad_norm": 2.263457048607926,
      "language_loss": 0.7691806,
      "learning_rate": 3.37824342225153e-06,
      "loss": 0.79152429,
      "num_input_tokens_seen": 50022495,
      "step": 2331,
      "time_per_iteration": 2.5652565956115723
    },
    {
      "auxiliary_loss_clip": 0.01216259,
      "auxiliary_loss_mlp": 0.01034484,
      "balance_loss_clip": 1.05539358,
      "balance_loss_mlp": 1.02483964,
      "epoch": 0.2804064209703601,
      "flos": 25520277409920.0,
      "grad_norm": 2.0424492697603056,
      "language_loss": 0.77309591,
      "learning_rate": 3.3776788393382006e-06,
      "loss": 0.79560333,
      "num_input_tokens_seen": 50041975,
      "step": 2332,
      "time_per_iteration": 2.7461299896240234
    },
    {
      "auxiliary_loss_clip": 0.01209598,
      "auxiliary_loss_mlp": 0.01027724,
      "balance_loss_clip": 1.06234908,
      "balance_loss_mlp": 1.01823449,
      "epoch": 0.2805266638609992,
      "flos": 29351766280320.0,
      "grad_norm": 4.163564860374044,
      "language_loss": 0.76512104,
      "learning_rate": 3.3771140474320872e-06,
      "loss": 0.7874943,
      "num_input_tokens_seen": 50061925,
      "step": 2333,
      "time_per_iteration": 2.6633658409118652
    },
    {
      "auxiliary_loss_clip": 0.01219839,
      "auxiliary_loss_mlp": 0.01034375,
      "balance_loss_clip": 1.05779326,
      "balance_loss_mlp": 1.02547622,
      "epoch": 0.28064690675163834,
      "flos": 21463735875840.0,
      "grad_norm": 2.039473707202899,
      "language_loss": 0.7966789,
      "learning_rate": 3.3765490466188664e-06,
      "loss": 0.81922102,
      "num_input_tokens_seen": 50079325,
      "step": 2334,
      "time_per_iteration": 2.6600341796875
    },
    {
      "auxiliary_loss_clip": 0.01210906,
      "auxiliary_loss_mlp": 0.01031229,
      "balance_loss_clip": 1.05537355,
      "balance_loss_mlp": 1.02085173,
      "epoch": 0.2807671496422774,
      "flos": 20995640812800.0,
      "grad_norm": 5.171782389528396,
      "language_loss": 0.74795973,
      "learning_rate": 3.3759838369842508e-06,
      "loss": 0.77038103,
      "num_input_tokens_seen": 50097400,
      "step": 2335,
      "time_per_iteration": 2.6450090408325195
    },
    {
      "auxiliary_loss_clip": 0.01213262,
      "auxiliary_loss_mlp": 0.01031945,
      "balance_loss_clip": 1.05606234,
      "balance_loss_mlp": 1.02267694,
      "epoch": 0.2808873925329165,
      "flos": 21506577822720.0,
      "grad_norm": 2.6227933985652,
      "language_loss": 0.73461998,
      "learning_rate": 3.375418418613981e-06,
      "loss": 0.75707197,
      "num_input_tokens_seen": 50116425,
      "step": 2336,
      "time_per_iteration": 2.728480577468872
    },
    {
      "auxiliary_loss_clip": 0.01212859,
      "auxiliary_loss_mlp": 0.0103177,
      "balance_loss_clip": 1.05938935,
      "balance_loss_mlp": 1.02125549,
      "epoch": 0.28100763542355556,
      "flos": 16070815814400.0,
      "grad_norm": 2.458024660150746,
      "language_loss": 0.8302604,
      "learning_rate": 3.374852791593831e-06,
      "loss": 0.85270667,
      "num_input_tokens_seen": 50132625,
      "step": 2337,
      "time_per_iteration": 2.571176528930664
    },
    {
      "auxiliary_loss_clip": 0.01226148,
      "auxiliary_loss_mlp": 0.01035152,
      "balance_loss_clip": 1.05312276,
      "balance_loss_mlp": 1.02497709,
      "epoch": 0.28112787831419467,
      "flos": 19062605468160.0,
      "grad_norm": 3.588699833562515,
      "language_loss": 0.54373795,
      "learning_rate": 3.374286956009605e-06,
      "loss": 0.56635094,
      "num_input_tokens_seen": 50151190,
      "step": 2338,
      "time_per_iteration": 3.709831953048706
    },
    {
      "auxiliary_loss_clip": 0.01213611,
      "auxiliary_loss_mlp": 0.01030973,
      "balance_loss_clip": 1.06471848,
      "balance_loss_mlp": 1.02117372,
      "epoch": 0.2812481212048338,
      "flos": 12823629482880.0,
      "grad_norm": 2.3246036458596677,
      "language_loss": 0.75169069,
      "learning_rate": 3.3737209119471405e-06,
      "loss": 0.77413654,
      "num_input_tokens_seen": 50167700,
      "step": 2339,
      "time_per_iteration": 3.5273373126983643
    },
    {
      "auxiliary_loss_clip": 0.01216621,
      "auxiliary_loss_mlp": 0.01031855,
      "balance_loss_clip": 1.06165338,
      "balance_loss_mlp": 1.02153766,
      "epoch": 0.28136836409547283,
      "flos": 15633064765440.0,
      "grad_norm": 3.3230685687718853,
      "language_loss": 0.64082128,
      "learning_rate": 3.373154659492306e-06,
      "loss": 0.66330606,
      "num_input_tokens_seen": 50185840,
      "step": 2340,
      "time_per_iteration": 2.6530423164367676
    },
    {
      "auxiliary_loss_clip": 0.01211902,
      "auxiliary_loss_mlp": 0.01039713,
      "balance_loss_clip": 1.05750275,
      "balance_loss_mlp": 1.03048587,
      "epoch": 0.28148860698611194,
      "flos": 19933726106880.0,
      "grad_norm": 2.0155362318416343,
      "language_loss": 0.85389292,
      "learning_rate": 3.3725881987310016e-06,
      "loss": 0.87640905,
      "num_input_tokens_seen": 50203375,
      "step": 2341,
      "time_per_iteration": 2.6654529571533203
    },
    {
      "auxiliary_loss_clip": 0.01207225,
      "auxiliary_loss_mlp": 0.01032408,
      "balance_loss_clip": 1.05547345,
      "balance_loss_mlp": 1.02350926,
      "epoch": 0.28160884987675106,
      "flos": 17457219008640.0,
      "grad_norm": 3.611461796753748,
      "language_loss": 0.87578261,
      "learning_rate": 3.372021529749159e-06,
      "loss": 0.89817894,
      "num_input_tokens_seen": 50222435,
      "step": 2342,
      "time_per_iteration": 2.586604356765747
    },
    {
      "auxiliary_loss_clip": 0.01218577,
      "auxiliary_loss_mlp": 0.01035135,
      "balance_loss_clip": 1.05426669,
      "balance_loss_mlp": 1.02589619,
      "epoch": 0.2817290927673901,
      "flos": 16834743290880.0,
      "grad_norm": 1.9327871526611726,
      "language_loss": 0.92614442,
      "learning_rate": 3.3714546526327405e-06,
      "loss": 0.94868153,
      "num_input_tokens_seen": 50240435,
      "step": 2343,
      "time_per_iteration": 3.7846529483795166
    },
    {
      "auxiliary_loss_clip": 0.01217875,
      "auxiliary_loss_mlp": 0.01031546,
      "balance_loss_clip": 1.05551267,
      "balance_loss_mlp": 1.02137756,
      "epoch": 0.2818493356580292,
      "flos": 15414081500160.0,
      "grad_norm": 3.018515656183819,
      "language_loss": 0.87884259,
      "learning_rate": 3.3708875674677423e-06,
      "loss": 0.90133679,
      "num_input_tokens_seen": 50258410,
      "step": 2344,
      "time_per_iteration": 2.755563259124756
    },
    {
      "auxiliary_loss_clip": 0.01222918,
      "auxiliary_loss_mlp": 0.01033382,
      "balance_loss_clip": 1.06343067,
      "balance_loss_mlp": 1.02301097,
      "epoch": 0.28196957854866833,
      "flos": 20412451595520.0,
      "grad_norm": 2.1246194128039066,
      "language_loss": 0.84007668,
      "learning_rate": 3.37032027434019e-06,
      "loss": 0.86263973,
      "num_input_tokens_seen": 50277930,
      "step": 2345,
      "time_per_iteration": 3.708768844604492
    },
    {
      "auxiliary_loss_clip": 0.01218866,
      "auxiliary_loss_mlp": 0.01033466,
      "balance_loss_clip": 1.059852,
      "balance_loss_mlp": 1.02202773,
      "epoch": 0.2820898214393074,
      "flos": 19973120348160.0,
      "grad_norm": 2.235041229841337,
      "language_loss": 0.83360517,
      "learning_rate": 3.369752773336141e-06,
      "loss": 0.85612845,
      "num_input_tokens_seen": 50297410,
      "step": 2346,
      "time_per_iteration": 2.6576948165893555
    },
    {
      "auxiliary_loss_clip": 0.01209002,
      "auxiliary_loss_mlp": 0.01033585,
      "balance_loss_clip": 1.05556655,
      "balance_loss_mlp": 1.02305853,
      "epoch": 0.2822100643299465,
      "flos": 22528308188160.0,
      "grad_norm": 2.1333285561384607,
      "language_loss": 0.78287446,
      "learning_rate": 3.3691850645416864e-06,
      "loss": 0.80530035,
      "num_input_tokens_seen": 50317120,
      "step": 2347,
      "time_per_iteration": 2.6667044162750244
    },
    {
      "auxiliary_loss_clip": 0.01215885,
      "auxiliary_loss_mlp": 0.01039433,
      "balance_loss_clip": 1.06023133,
      "balance_loss_mlp": 1.02956843,
      "epoch": 0.2823303072205856,
      "flos": 11546682007680.0,
      "grad_norm": 2.6720690909902567,
      "language_loss": 0.83178532,
      "learning_rate": 3.368617148042945e-06,
      "loss": 0.85433847,
      "num_input_tokens_seen": 50334790,
      "step": 2348,
      "time_per_iteration": 2.621933937072754
    },
    {
      "auxiliary_loss_clip": 0.01207828,
      "auxiliary_loss_mlp": 0.01036912,
      "balance_loss_clip": 1.05570364,
      "balance_loss_mlp": 1.02605844,
      "epoch": 0.28245055011122466,
      "flos": 18259894281600.0,
      "grad_norm": 1.7402994978402446,
      "language_loss": 0.84751683,
      "learning_rate": 3.368049023926071e-06,
      "loss": 0.86996424,
      "num_input_tokens_seen": 50353785,
      "step": 2349,
      "time_per_iteration": 2.626523971557617
    },
    {
      "auxiliary_loss_clip": 0.01209153,
      "auxiliary_loss_mlp": 0.01031504,
      "balance_loss_clip": 1.0613668,
      "balance_loss_mlp": 1.02242041,
      "epoch": 0.2825707930018638,
      "flos": 24608110504320.0,
      "grad_norm": 1.727077957740032,
      "language_loss": 0.83703804,
      "learning_rate": 3.3674806922772476e-06,
      "loss": 0.85944462,
      "num_input_tokens_seen": 50374670,
      "step": 2350,
      "time_per_iteration": 2.6622610092163086
    },
    {
      "auxiliary_loss_clip": 0.01221492,
      "auxiliary_loss_mlp": 0.01038668,
      "balance_loss_clip": 1.05716491,
      "balance_loss_mlp": 1.02848125,
      "epoch": 0.28269103589250283,
      "flos": 25226994862080.0,
      "grad_norm": 1.8713140383741431,
      "language_loss": 0.7495988,
      "learning_rate": 3.3669121531826904e-06,
      "loss": 0.77220035,
      "num_input_tokens_seen": 50395650,
      "step": 2351,
      "time_per_iteration": 2.7062361240386963
    },
    {
      "auxiliary_loss_clip": 0.0121057,
      "auxiliary_loss_mlp": 0.01034964,
      "balance_loss_clip": 1.05980849,
      "balance_loss_mlp": 1.0256424,
      "epoch": 0.28281127878314194,
      "flos": 19281552819840.0,
      "grad_norm": 2.0625632280722686,
      "language_loss": 0.83150256,
      "learning_rate": 3.366343406728647e-06,
      "loss": 0.85395789,
      "num_input_tokens_seen": 50415100,
      "step": 2352,
      "time_per_iteration": 2.721390724182129
    },
    {
      "auxiliary_loss_clip": 0.01202651,
      "auxiliary_loss_mlp": 0.01032362,
      "balance_loss_clip": 1.05618131,
      "balance_loss_mlp": 1.02309346,
      "epoch": 0.28293152167378105,
      "flos": 23878405710720.0,
      "grad_norm": 2.0139477534883032,
      "language_loss": 0.68398201,
      "learning_rate": 3.3657744530013946e-06,
      "loss": 0.70633215,
      "num_input_tokens_seen": 50434335,
      "step": 2353,
      "time_per_iteration": 2.7291810512542725
    },
    {
      "auxiliary_loss_clip": 0.01211377,
      "auxiliary_loss_mlp": 0.01033093,
      "balance_loss_clip": 1.05916822,
      "balance_loss_mlp": 1.02354407,
      "epoch": 0.2830517645644201,
      "flos": 43866965928960.0,
      "grad_norm": 1.9015416757420034,
      "language_loss": 0.71430206,
      "learning_rate": 3.3652052920872437e-06,
      "loss": 0.73674679,
      "num_input_tokens_seen": 50457200,
      "step": 2354,
      "time_per_iteration": 2.798475503921509
    },
    {
      "auxiliary_loss_clip": 0.01211311,
      "auxiliary_loss_mlp": 0.01035339,
      "balance_loss_clip": 1.05529273,
      "balance_loss_mlp": 1.02527761,
      "epoch": 0.2831720074550592,
      "flos": 26651750803200.0,
      "grad_norm": 3.0911787898099785,
      "language_loss": 0.85586762,
      "learning_rate": 3.3646359240725355e-06,
      "loss": 0.87833416,
      "num_input_tokens_seen": 50476390,
      "step": 2355,
      "time_per_iteration": 2.750699996948242
    },
    {
      "auxiliary_loss_clip": 0.01203954,
      "auxiliary_loss_mlp": 0.01209282,
      "balance_loss_clip": 1.05890167,
      "balance_loss_mlp": 1.0003705,
      "epoch": 0.2832922503456983,
      "flos": 31029979564800.0,
      "grad_norm": 2.6100918259685297,
      "language_loss": 0.67534745,
      "learning_rate": 3.364066349043643e-06,
      "loss": 0.69947976,
      "num_input_tokens_seen": 50497595,
      "step": 2356,
      "time_per_iteration": 2.6845602989196777
    },
    {
      "auxiliary_loss_clip": 0.01206956,
      "auxiliary_loss_mlp": 0.01033321,
      "balance_loss_clip": 1.0554955,
      "balance_loss_mlp": 1.02426744,
      "epoch": 0.2834124932363374,
      "flos": 20405699838720.0,
      "grad_norm": 1.9222400842730953,
      "language_loss": 0.8201859,
      "learning_rate": 3.363496567086969e-06,
      "loss": 0.84258866,
      "num_input_tokens_seen": 50514690,
      "step": 2357,
      "time_per_iteration": 2.730910539627075
    },
    {
      "auxiliary_loss_clip": 0.01207201,
      "auxiliary_loss_mlp": 0.01030862,
      "balance_loss_clip": 1.06131864,
      "balance_loss_mlp": 1.02159381,
      "epoch": 0.2835327361269765,
      "flos": 39384848056320.0,
      "grad_norm": 1.9379199852451332,
      "language_loss": 0.75524461,
      "learning_rate": 3.3629265782889506e-06,
      "loss": 0.7776252,
      "num_input_tokens_seen": 50536515,
      "step": 2358,
      "time_per_iteration": 2.7400166988372803
    },
    {
      "auxiliary_loss_clip": 0.01207294,
      "auxiliary_loss_mlp": 0.01037549,
      "balance_loss_clip": 1.05199993,
      "balance_loss_mlp": 1.02741671,
      "epoch": 0.2836529790176156,
      "flos": 30261598801920.0,
      "grad_norm": 1.9022895123849473,
      "language_loss": 0.72088093,
      "learning_rate": 3.362356382736054e-06,
      "loss": 0.74332935,
      "num_input_tokens_seen": 50557120,
      "step": 2359,
      "time_per_iteration": 2.748886823654175
    },
    {
      "auxiliary_loss_clip": 0.01211509,
      "auxiliary_loss_mlp": 0.01026071,
      "balance_loss_clip": 1.05231392,
      "balance_loss_mlp": 1.0172261,
      "epoch": 0.28377322190825466,
      "flos": 12677796264960.0,
      "grad_norm": 4.436767358116267,
      "language_loss": 0.91119552,
      "learning_rate": 3.361785980514777e-06,
      "loss": 0.93357134,
      "num_input_tokens_seen": 50573320,
      "step": 2360,
      "time_per_iteration": 2.701207160949707
    },
    {
      "auxiliary_loss_clip": 0.0120964,
      "auxiliary_loss_mlp": 0.01029719,
      "balance_loss_clip": 1.05150807,
      "balance_loss_mlp": 1.02055788,
      "epoch": 0.28389346479889377,
      "flos": 18296666830080.0,
      "grad_norm": 4.107772098317567,
      "language_loss": 0.77065587,
      "learning_rate": 3.361215371711649e-06,
      "loss": 0.79304945,
      "num_input_tokens_seen": 50592415,
      "step": 2361,
      "time_per_iteration": 2.691819667816162
    },
    {
      "auxiliary_loss_clip": 0.01208453,
      "auxiliary_loss_mlp": 0.01032206,
      "balance_loss_clip": 1.05485058,
      "balance_loss_mlp": 1.02302086,
      "epoch": 0.2840137076895329,
      "flos": 20406992728320.0,
      "grad_norm": 1.8868858451467518,
      "language_loss": 0.83290637,
      "learning_rate": 3.3606445564132326e-06,
      "loss": 0.85531294,
      "num_input_tokens_seen": 50609710,
      "step": 2362,
      "time_per_iteration": 2.7297325134277344
    },
    {
      "auxiliary_loss_clip": 0.01209435,
      "auxiliary_loss_mlp": 0.01209518,
      "balance_loss_clip": 1.06392121,
      "balance_loss_mlp": 1.00032866,
      "epoch": 0.28413395058017193,
      "flos": 20048030161920.0,
      "grad_norm": 2.489694871065626,
      "language_loss": 0.82248116,
      "learning_rate": 3.360073534706118e-06,
      "loss": 0.84667063,
      "num_input_tokens_seen": 50626865,
      "step": 2363,
      "time_per_iteration": 2.5697903633117676
    },
    {
      "auxiliary_loss_clip": 0.01212838,
      "auxiliary_loss_mlp": 0.01030802,
      "balance_loss_clip": 1.05843914,
      "balance_loss_mlp": 1.0212059,
      "epoch": 0.28425419347081105,
      "flos": 37663613256960.0,
      "grad_norm": 4.119001168776408,
      "language_loss": 0.7582773,
      "learning_rate": 3.35950230667693e-06,
      "loss": 0.7807138,
      "num_input_tokens_seen": 50648560,
      "step": 2364,
      "time_per_iteration": 3.7116053104400635
    },
    {
      "auxiliary_loss_clip": 0.01209322,
      "auxiliary_loss_mlp": 0.01024761,
      "balance_loss_clip": 1.05790234,
      "balance_loss_mlp": 1.0157547,
      "epoch": 0.28437443636145016,
      "flos": 13845072539520.0,
      "grad_norm": 3.2960192499180434,
      "language_loss": 0.86020899,
      "learning_rate": 3.358930872412323e-06,
      "loss": 0.88254976,
      "num_input_tokens_seen": 50665725,
      "step": 2365,
      "time_per_iteration": 3.56569766998291
    },
    {
      "auxiliary_loss_clip": 0.01207794,
      "auxiliary_loss_mlp": 0.01031864,
      "balance_loss_clip": 1.06016338,
      "balance_loss_mlp": 1.02211869,
      "epoch": 0.2844946792520892,
      "flos": 22747794243840.0,
      "grad_norm": 3.7230952187940165,
      "language_loss": 0.8107022,
      "learning_rate": 3.3583592319989825e-06,
      "loss": 0.83309877,
      "num_input_tokens_seen": 50685095,
      "step": 2366,
      "time_per_iteration": 2.639411211013794
    },
    {
      "auxiliary_loss_clip": 0.01216566,
      "auxiliary_loss_mlp": 0.01033734,
      "balance_loss_clip": 1.06014323,
      "balance_loss_mlp": 1.02391076,
      "epoch": 0.2846149221427283,
      "flos": 32415987709440.0,
      "grad_norm": 3.2378084403039895,
      "language_loss": 0.68962145,
      "learning_rate": 3.357787385523627e-06,
      "loss": 0.71212447,
      "num_input_tokens_seen": 50706500,
      "step": 2367,
      "time_per_iteration": 2.867439031600952
    },
    {
      "auxiliary_loss_clip": 0.0122466,
      "auxiliary_loss_mlp": 0.01027628,
      "balance_loss_clip": 1.0528717,
      "balance_loss_mlp": 1.01868129,
      "epoch": 0.2847351650333674,
      "flos": 28475976873600.0,
      "grad_norm": 2.085692713895578,
      "language_loss": 0.82421911,
      "learning_rate": 3.3572153330730048e-06,
      "loss": 0.84674203,
      "num_input_tokens_seen": 50727595,
      "step": 2368,
      "time_per_iteration": 2.7713422775268555
    },
    {
      "auxiliary_loss_clip": 0.01127695,
      "auxiliary_loss_mlp": 0.01003867,
      "balance_loss_clip": 1.02801585,
      "balance_loss_mlp": 1.00254965,
      "epoch": 0.2848554079240065,
      "flos": 55753399704960.0,
      "grad_norm": 0.8333022567218303,
      "language_loss": 0.64675725,
      "learning_rate": 3.3566430747338956e-06,
      "loss": 0.66807288,
      "num_input_tokens_seen": 50782800,
      "step": 2369,
      "time_per_iteration": 3.071648120880127
    },
    {
      "auxiliary_loss_clip": 0.01209394,
      "auxiliary_loss_mlp": 0.01033942,
      "balance_loss_clip": 1.05709875,
      "balance_loss_mlp": 1.02439916,
      "epoch": 0.2849756508146456,
      "flos": 11836875985920.0,
      "grad_norm": 8.63948113938574,
      "language_loss": 0.86762029,
      "learning_rate": 3.35607061059311e-06,
      "loss": 0.89005363,
      "num_input_tokens_seen": 50797730,
      "step": 2370,
      "time_per_iteration": 3.6678757667541504
    },
    {
      "auxiliary_loss_clip": 0.01203036,
      "auxiliary_loss_mlp": 0.01035164,
      "balance_loss_clip": 1.05961323,
      "balance_loss_mlp": 1.02600276,
      "epoch": 0.28509589370528465,
      "flos": 25155209531520.0,
      "grad_norm": 5.581063174483512,
      "language_loss": 0.75261462,
      "learning_rate": 3.3554979407374917e-06,
      "loss": 0.77499658,
      "num_input_tokens_seen": 50819840,
      "step": 2371,
      "time_per_iteration": 3.5410144329071045
    },
    {
      "auxiliary_loss_clip": 0.01207955,
      "auxiliary_loss_mlp": 0.01032442,
      "balance_loss_clip": 1.0585022,
      "balance_loss_mlp": 1.02321577,
      "epoch": 0.28521613659592376,
      "flos": 19974808287360.0,
      "grad_norm": 2.986801922004999,
      "language_loss": 0.73413134,
      "learning_rate": 3.3549250652539134e-06,
      "loss": 0.75653529,
      "num_input_tokens_seen": 50838935,
      "step": 2372,
      "time_per_iteration": 2.647395610809326
    },
    {
      "auxiliary_loss_clip": 0.01208217,
      "auxiliary_loss_mlp": 0.0103051,
      "balance_loss_clip": 1.05462945,
      "balance_loss_mlp": 1.0207175,
      "epoch": 0.2853363794865629,
      "flos": 23367971491200.0,
      "grad_norm": 1.937963231763393,
      "language_loss": 0.81942081,
      "learning_rate": 3.3543519842292794e-06,
      "loss": 0.84180808,
      "num_input_tokens_seen": 50858590,
      "step": 2373,
      "time_per_iteration": 2.6863226890563965
    },
    {
      "auxiliary_loss_clip": 0.01206751,
      "auxiliary_loss_mlp": 0.01209181,
      "balance_loss_clip": 1.06149554,
      "balance_loss_mlp": 1.00046396,
      "epoch": 0.28545662237720193,
      "flos": 19861940776320.0,
      "grad_norm": 6.598459119961093,
      "language_loss": 0.83898449,
      "learning_rate": 3.353778697750527e-06,
      "loss": 0.8631438,
      "num_input_tokens_seen": 50876995,
      "step": 2374,
      "time_per_iteration": 2.6064653396606445
    },
    {
      "auxiliary_loss_clip": 0.01201378,
      "auxiliary_loss_mlp": 0.01027922,
      "balance_loss_clip": 1.05599546,
      "balance_loss_mlp": 1.0184027,
      "epoch": 0.28557686526784104,
      "flos": 23879016241920.0,
      "grad_norm": 2.031007908425535,
      "language_loss": 0.89428633,
      "learning_rate": 3.353205205904622e-06,
      "loss": 0.91657937,
      "num_input_tokens_seen": 50896105,
      "step": 2375,
      "time_per_iteration": 2.608664035797119
    },
    {
      "auxiliary_loss_clip": 0.0121073,
      "auxiliary_loss_mlp": 0.0102939,
      "balance_loss_clip": 1.05864811,
      "balance_loss_mlp": 1.02017534,
      "epoch": 0.28569710815848015,
      "flos": 44890384233600.0,
      "grad_norm": 2.2790040448881097,
      "language_loss": 0.72002506,
      "learning_rate": 3.3526315087785637e-06,
      "loss": 0.74242622,
      "num_input_tokens_seen": 50917220,
      "step": 2376,
      "time_per_iteration": 2.8668177127838135
    },
    {
      "auxiliary_loss_clip": 0.01192988,
      "auxiliary_loss_mlp": 0.0103196,
      "balance_loss_clip": 1.05129457,
      "balance_loss_mlp": 1.02286422,
      "epoch": 0.2858173510491192,
      "flos": 26829759628800.0,
      "grad_norm": 1.653542140291424,
      "language_loss": 0.80826378,
      "learning_rate": 3.3520576064593805e-06,
      "loss": 0.83051324,
      "num_input_tokens_seen": 50937175,
      "step": 2377,
      "time_per_iteration": 2.7717883586883545
    },
    {
      "auxiliary_loss_clip": 0.01212927,
      "auxiliary_loss_mlp": 0.0102958,
      "balance_loss_clip": 1.06159151,
      "balance_loss_mlp": 1.02016795,
      "epoch": 0.2859375939397583,
      "flos": 23148916398720.0,
      "grad_norm": 2.895469037516004,
      "language_loss": 0.81848609,
      "learning_rate": 3.3514834990341337e-06,
      "loss": 0.84091115,
      "num_input_tokens_seen": 50957500,
      "step": 2378,
      "time_per_iteration": 2.705627679824829
    },
    {
      "auxiliary_loss_clip": 0.01217055,
      "auxiliary_loss_mlp": 0.01028171,
      "balance_loss_clip": 1.06056225,
      "balance_loss_mlp": 1.01958227,
      "epoch": 0.2860578368303974,
      "flos": 12129799397760.0,
      "grad_norm": 2.3552321474520483,
      "language_loss": 0.93276334,
      "learning_rate": 3.3509091865899144e-06,
      "loss": 0.95521557,
      "num_input_tokens_seen": 50972690,
      "step": 2379,
      "time_per_iteration": 2.670592784881592
    },
    {
      "auxiliary_loss_clip": 0.01208316,
      "auxiliary_loss_mlp": 0.01033097,
      "balance_loss_clip": 1.06238914,
      "balance_loss_mlp": 1.02317882,
      "epoch": 0.2861780797210365,
      "flos": 19938035738880.0,
      "grad_norm": 2.1333557144441277,
      "language_loss": 0.70552748,
      "learning_rate": 3.350334669213846e-06,
      "loss": 0.72794163,
      "num_input_tokens_seen": 50990095,
      "step": 2380,
      "time_per_iteration": 2.5858259201049805
    },
    {
      "auxiliary_loss_clip": 0.01208736,
      "auxiliary_loss_mlp": 0.01032367,
      "balance_loss_clip": 1.06278741,
      "balance_loss_mlp": 1.02395034,
      "epoch": 0.2862983226116756,
      "flos": 27563127609600.0,
      "grad_norm": 2.166701085426644,
      "language_loss": 0.75757849,
      "learning_rate": 3.3497599469930816e-06,
      "loss": 0.77998948,
      "num_input_tokens_seen": 51008305,
      "step": 2381,
      "time_per_iteration": 2.720102310180664
    },
    {
      "auxiliary_loss_clip": 0.01209358,
      "auxiliary_loss_mlp": 0.01031419,
      "balance_loss_clip": 1.06224966,
      "balance_loss_mlp": 1.02212691,
      "epoch": 0.28641856550231465,
      "flos": 22053964158720.0,
      "grad_norm": 2.16931572429577,
      "language_loss": 0.83281553,
      "learning_rate": 3.349185020014807e-06,
      "loss": 0.8552233,
      "num_input_tokens_seen": 51025570,
      "step": 2382,
      "time_per_iteration": 2.642691135406494
    },
    {
      "auxiliary_loss_clip": 0.01212315,
      "auxiliary_loss_mlp": 0.01026203,
      "balance_loss_clip": 1.06014419,
      "balance_loss_mlp": 1.0172503,
      "epoch": 0.28653880839295376,
      "flos": 22378775869440.0,
      "grad_norm": 1.832586695917006,
      "language_loss": 0.74616635,
      "learning_rate": 3.348609888366237e-06,
      "loss": 0.76855153,
      "num_input_tokens_seen": 51044585,
      "step": 2383,
      "time_per_iteration": 2.720743179321289
    },
    {
      "auxiliary_loss_clip": 0.01208916,
      "auxiliary_loss_mlp": 0.0102789,
      "balance_loss_clip": 1.04874575,
      "balance_loss_mlp": 1.01795411,
      "epoch": 0.28665905128359287,
      "flos": 23367971491200.0,
      "grad_norm": 2.10800562166961,
      "language_loss": 0.63153446,
      "learning_rate": 3.348034552134619e-06,
      "loss": 0.65390247,
      "num_input_tokens_seen": 51063990,
      "step": 2384,
      "time_per_iteration": 2.7155091762542725
    },
    {
      "auxiliary_loss_clip": 0.01205539,
      "auxiliary_loss_mlp": 0.01036749,
      "balance_loss_clip": 1.05327201,
      "balance_loss_mlp": 1.02777231,
      "epoch": 0.2867792941742319,
      "flos": 20881695893760.0,
      "grad_norm": 1.9228908269749867,
      "language_loss": 0.84349054,
      "learning_rate": 3.3474590114072316e-06,
      "loss": 0.86591339,
      "num_input_tokens_seen": 51081990,
      "step": 2385,
      "time_per_iteration": 2.7108559608459473
    },
    {
      "auxiliary_loss_clip": 0.01214262,
      "auxiliary_loss_mlp": 0.01034181,
      "balance_loss_clip": 1.06249022,
      "balance_loss_mlp": 1.02444196,
      "epoch": 0.28689953706487104,
      "flos": 20664005518080.0,
      "grad_norm": 1.932330955940801,
      "language_loss": 0.83468837,
      "learning_rate": 3.3468832662713836e-06,
      "loss": 0.85717285,
      "num_input_tokens_seen": 51100235,
      "step": 2386,
      "time_per_iteration": 2.6515884399414062
    },
    {
      "auxiliary_loss_clip": 0.0121022,
      "auxiliary_loss_mlp": 0.01036306,
      "balance_loss_clip": 1.05842566,
      "balance_loss_mlp": 1.02712131,
      "epoch": 0.28701977995551015,
      "flos": 12675533708160.0,
      "grad_norm": 3.353762620419113,
      "language_loss": 0.83759463,
      "learning_rate": 3.346307316814415e-06,
      "loss": 0.86005986,
      "num_input_tokens_seen": 51115405,
      "step": 2387,
      "time_per_iteration": 2.601555585861206
    },
    {
      "auxiliary_loss_clip": 0.01209589,
      "auxiliary_loss_mlp": 0.01030164,
      "balance_loss_clip": 1.06175518,
      "balance_loss_mlp": 1.0205617,
      "epoch": 0.2871400228461492,
      "flos": 21252366293760.0,
      "grad_norm": 2.0041411663196587,
      "language_loss": 0.75752807,
      "learning_rate": 3.3457311631236965e-06,
      "loss": 0.77992553,
      "num_input_tokens_seen": 51136390,
      "step": 2388,
      "time_per_iteration": 2.5779261589050293
    },
    {
      "auxiliary_loss_clip": 0.01199443,
      "auxiliary_loss_mlp": 0.01032216,
      "balance_loss_clip": 1.05726993,
      "balance_loss_mlp": 1.02313793,
      "epoch": 0.2872602657367883,
      "flos": 25119262995840.0,
      "grad_norm": 4.660780913456702,
      "language_loss": 0.84337676,
      "learning_rate": 3.345154805286631e-06,
      "loss": 0.86569333,
      "num_input_tokens_seen": 51156650,
      "step": 2389,
      "time_per_iteration": 2.731658458709717
    },
    {
      "auxiliary_loss_clip": 0.01204223,
      "auxiliary_loss_mlp": 0.01035143,
      "balance_loss_clip": 1.05948734,
      "balance_loss_mlp": 1.02581477,
      "epoch": 0.2873805086274274,
      "flos": 16646606830080.0,
      "grad_norm": 2.2295453552094564,
      "language_loss": 0.7627027,
      "learning_rate": 3.344578243390651e-06,
      "loss": 0.78509641,
      "num_input_tokens_seen": 51172210,
      "step": 2390,
      "time_per_iteration": 2.7406651973724365
    },
    {
      "auxiliary_loss_clip": 0.01208827,
      "auxiliary_loss_mlp": 0.01031367,
      "balance_loss_clip": 1.0607698,
      "balance_loss_mlp": 1.02176452,
      "epoch": 0.2875007515180665,
      "flos": 17420123237760.0,
      "grad_norm": 2.5887874428301356,
      "language_loss": 0.78616369,
      "learning_rate": 3.3440014775232206e-06,
      "loss": 0.80856562,
      "num_input_tokens_seen": 51190265,
      "step": 2391,
      "time_per_iteration": 3.5838539600372314
    },
    {
      "auxiliary_loss_clip": 0.01214895,
      "auxiliary_loss_mlp": 0.01032172,
      "balance_loss_clip": 1.0585866,
      "balance_loss_mlp": 1.02358854,
      "epoch": 0.2876209944087056,
      "flos": 23434190213760.0,
      "grad_norm": 2.864188280281474,
      "language_loss": 0.70704734,
      "learning_rate": 3.343424507771834e-06,
      "loss": 0.729518,
      "num_input_tokens_seen": 51208475,
      "step": 2392,
      "time_per_iteration": 3.5756914615631104
    },
    {
      "auxiliary_loss_clip": 0.0121176,
      "auxiliary_loss_mlp": 0.01027647,
      "balance_loss_clip": 1.05918503,
      "balance_loss_mlp": 1.01880145,
      "epoch": 0.2877412372993447,
      "flos": 13735509079680.0,
      "grad_norm": 5.582624505241904,
      "language_loss": 0.86771244,
      "learning_rate": 3.342847334224018e-06,
      "loss": 0.89010656,
      "num_input_tokens_seen": 51225875,
      "step": 2393,
      "time_per_iteration": 2.7228293418884277
    },
    {
      "auxiliary_loss_clip": 0.0112325,
      "auxiliary_loss_mlp": 0.0100187,
      "balance_loss_clip": 1.03771496,
      "balance_loss_mlp": 1.00062454,
      "epoch": 0.28786148018998375,
      "flos": 58079695104000.0,
      "grad_norm": 1.1391892015057372,
      "language_loss": 0.62434912,
      "learning_rate": 3.342269956967329e-06,
      "loss": 0.64560032,
      "num_input_tokens_seen": 51287780,
      "step": 2394,
      "time_per_iteration": 3.224294662475586
    },
    {
      "auxiliary_loss_clip": 0.01215769,
      "auxiliary_loss_mlp": 0.01030579,
      "balance_loss_clip": 1.06447029,
      "balance_loss_mlp": 1.01990366,
      "epoch": 0.28798172308062286,
      "flos": 23435052140160.0,
      "grad_norm": 2.5264120091265827,
      "language_loss": 0.71925718,
      "learning_rate": 3.341692376089355e-06,
      "loss": 0.74172068,
      "num_input_tokens_seen": 51303335,
      "step": 2395,
      "time_per_iteration": 2.692016124725342
    },
    {
      "auxiliary_loss_clip": 0.01208225,
      "auxiliary_loss_mlp": 0.01032254,
      "balance_loss_clip": 1.06202805,
      "balance_loss_mlp": 1.02276516,
      "epoch": 0.288101965971262,
      "flos": 25110033200640.0,
      "grad_norm": 5.287075111514826,
      "language_loss": 0.83829677,
      "learning_rate": 3.3411145916777146e-06,
      "loss": 0.86070156,
      "num_input_tokens_seen": 51317495,
      "step": 2396,
      "time_per_iteration": 3.573467969894409
    },
    {
      "auxiliary_loss_clip": 0.01204666,
      "auxiliary_loss_mlp": 0.01028604,
      "balance_loss_clip": 1.05756569,
      "balance_loss_mlp": 1.01890051,
      "epoch": 0.28822220886190103,
      "flos": 16252559654400.0,
      "grad_norm": 2.7576904294360354,
      "language_loss": 0.91320229,
      "learning_rate": 3.3405366038200566e-06,
      "loss": 0.93553501,
      "num_input_tokens_seen": 51336430,
      "step": 2397,
      "time_per_iteration": 2.7175064086914062
    },
    {
      "auxiliary_loss_clip": 0.01218167,
      "auxiliary_loss_mlp": 0.01033836,
      "balance_loss_clip": 1.06630957,
      "balance_loss_mlp": 1.02406096,
      "epoch": 0.28834245175254014,
      "flos": 24535642815360.0,
      "grad_norm": 3.1504689214789265,
      "language_loss": 0.85106194,
      "learning_rate": 3.3399584126040617e-06,
      "loss": 0.87358201,
      "num_input_tokens_seen": 51355930,
      "step": 2398,
      "time_per_iteration": 3.623267650604248
    },
    {
      "auxiliary_loss_clip": 0.01206876,
      "auxiliary_loss_mlp": 0.01208291,
      "balance_loss_clip": 1.06254077,
      "balance_loss_mlp": 1.00045705,
      "epoch": 0.2884626946431792,
      "flos": 24571445696640.0,
      "grad_norm": 2.076316768756297,
      "language_loss": 0.91318047,
      "learning_rate": 3.339380018117441e-06,
      "loss": 0.93733215,
      "num_input_tokens_seen": 51376765,
      "step": 2399,
      "time_per_iteration": 2.782167673110962
    },
    {
      "auxiliary_loss_clip": 0.01207081,
      "auxiliary_loss_mlp": 0.01030664,
      "balance_loss_clip": 1.06320167,
      "balance_loss_mlp": 1.02156854,
      "epoch": 0.2885829375338183,
      "flos": 16544657053440.0,
      "grad_norm": 2.7627624778412985,
      "language_loss": 0.78478682,
      "learning_rate": 3.3388014204479366e-06,
      "loss": 0.80716431,
      "num_input_tokens_seen": 51394570,
      "step": 2400,
      "time_per_iteration": 2.606701135635376
    },
    {
      "auxiliary_loss_clip": 0.01211436,
      "auxiliary_loss_mlp": 0.01027923,
      "balance_loss_clip": 1.06436265,
      "balance_loss_mlp": 1.0188756,
      "epoch": 0.2887031804244574,
      "flos": 24061226958720.0,
      "grad_norm": 2.608543040079281,
      "language_loss": 0.91628695,
      "learning_rate": 3.338222619683321e-06,
      "loss": 0.93868059,
      "num_input_tokens_seen": 51414535,
      "step": 2401,
      "time_per_iteration": 2.5974538326263428
    },
    {
      "auxiliary_loss_clip": 0.01215,
      "auxiliary_loss_mlp": 0.01030301,
      "balance_loss_clip": 1.05960846,
      "balance_loss_mlp": 1.02091897,
      "epoch": 0.2888234233150965,
      "flos": 23330696152320.0,
      "grad_norm": 3.9898761868418986,
      "language_loss": 0.73840463,
      "learning_rate": 3.337643615911398e-06,
      "loss": 0.76085758,
      "num_input_tokens_seen": 51434160,
      "step": 2402,
      "time_per_iteration": 2.6478898525238037
    },
    {
      "auxiliary_loss_clip": 0.01210575,
      "auxiliary_loss_mlp": 0.01028266,
      "balance_loss_clip": 1.06082082,
      "balance_loss_mlp": 1.01848507,
      "epoch": 0.2889436662057356,
      "flos": 22272767856000.0,
      "grad_norm": 2.662561218152992,
      "language_loss": 0.7899425,
      "learning_rate": 3.3370644092200026e-06,
      "loss": 0.8123309,
      "num_input_tokens_seen": 51451435,
      "step": 2403,
      "time_per_iteration": 2.600285768508911
    },
    {
      "auxiliary_loss_clip": 0.01198535,
      "auxiliary_loss_mlp": 0.01029648,
      "balance_loss_clip": 1.05215275,
      "balance_loss_mlp": 1.02011693,
      "epoch": 0.2890639090963747,
      "flos": 21616931381760.0,
      "grad_norm": 2.2775785741964745,
      "language_loss": 0.7834394,
      "learning_rate": 3.3364849996969985e-06,
      "loss": 0.80572122,
      "num_input_tokens_seen": 51471455,
      "step": 2404,
      "time_per_iteration": 2.7333648204803467
    },
    {
      "auxiliary_loss_clip": 0.01208649,
      "auxiliary_loss_mlp": 0.0102766,
      "balance_loss_clip": 1.0623405,
      "balance_loss_mlp": 1.01868331,
      "epoch": 0.28918415198701375,
      "flos": 28585540333440.0,
      "grad_norm": 2.573636874506399,
      "language_loss": 0.85348469,
      "learning_rate": 3.335905387430283e-06,
      "loss": 0.87584782,
      "num_input_tokens_seen": 51492890,
      "step": 2405,
      "time_per_iteration": 2.6568691730499268
    },
    {
      "auxiliary_loss_clip": 0.01211912,
      "auxiliary_loss_mlp": 0.01030884,
      "balance_loss_clip": 1.05668449,
      "balance_loss_mlp": 1.02153254,
      "epoch": 0.28930439487765286,
      "flos": 21944688007680.0,
      "grad_norm": 1.9731060983297184,
      "language_loss": 0.8305794,
      "learning_rate": 3.335325572507782e-06,
      "loss": 0.85300738,
      "num_input_tokens_seen": 51513390,
      "step": 2406,
      "time_per_iteration": 2.709174156188965
    },
    {
      "auxiliary_loss_clip": 0.01210694,
      "auxiliary_loss_mlp": 0.01208896,
      "balance_loss_clip": 1.06729722,
      "balance_loss_mlp": 1.00058627,
      "epoch": 0.28942463776829197,
      "flos": 19281911955840.0,
      "grad_norm": 1.8558402222833714,
      "language_loss": 0.73829901,
      "learning_rate": 3.3347455550174537e-06,
      "loss": 0.76249492,
      "num_input_tokens_seen": 51532730,
      "step": 2407,
      "time_per_iteration": 2.592559814453125
    },
    {
      "auxiliary_loss_clip": 0.01207374,
      "auxiliary_loss_mlp": 0.01031299,
      "balance_loss_clip": 1.05384219,
      "balance_loss_mlp": 1.02121437,
      "epoch": 0.289544880658931,
      "flos": 14645700737280.0,
      "grad_norm": 2.5703509345395674,
      "language_loss": 0.68206143,
      "learning_rate": 3.3341653350472864e-06,
      "loss": 0.70444816,
      "num_input_tokens_seen": 51549560,
      "step": 2408,
      "time_per_iteration": 2.69356632232666
    },
    {
      "auxiliary_loss_clip": 0.01214589,
      "auxiliary_loss_mlp": 0.01031067,
      "balance_loss_clip": 1.06261098,
      "balance_loss_mlp": 1.02074909,
      "epoch": 0.28966512354957014,
      "flos": 28621881918720.0,
      "grad_norm": 14.60061122247828,
      "language_loss": 0.69260919,
      "learning_rate": 3.333584912685298e-06,
      "loss": 0.71506572,
      "num_input_tokens_seen": 51568180,
      "step": 2409,
      "time_per_iteration": 2.664466381072998
    },
    {
      "auxiliary_loss_clip": 0.01126267,
      "auxiliary_loss_mlp": 0.01000725,
      "balance_loss_clip": 1.03206992,
      "balance_loss_mlp": 0.99945527,
      "epoch": 0.28978536644020925,
      "flos": 64711784511360.0,
      "grad_norm": 0.8552838503133449,
      "language_loss": 0.55528396,
      "learning_rate": 3.3330042880195385e-06,
      "loss": 0.57655388,
      "num_input_tokens_seen": 51622530,
      "step": 2410,
      "time_per_iteration": 3.177133321762085
    },
    {
      "auxiliary_loss_clip": 0.01208896,
      "auxiliary_loss_mlp": 0.01029847,
      "balance_loss_clip": 1.0577116,
      "balance_loss_mlp": 1.02070355,
      "epoch": 0.2899056093308483,
      "flos": 18624638937600.0,
      "grad_norm": 1.7337865329255133,
      "language_loss": 0.78787863,
      "learning_rate": 3.3324234611380888e-06,
      "loss": 0.81026614,
      "num_input_tokens_seen": 51641260,
      "step": 2411,
      "time_per_iteration": 2.7644078731536865
    },
    {
      "auxiliary_loss_clip": 0.01201907,
      "auxiliary_loss_mlp": 0.01030311,
      "balance_loss_clip": 1.05439472,
      "balance_loss_mlp": 1.02124536,
      "epoch": 0.2900258522214874,
      "flos": 22893735202560.0,
      "grad_norm": 1.6990941582812613,
      "language_loss": 0.81579685,
      "learning_rate": 3.3318424321290596e-06,
      "loss": 0.83811909,
      "num_input_tokens_seen": 51660975,
      "step": 2412,
      "time_per_iteration": 2.6668171882629395
    },
    {
      "auxiliary_loss_clip": 0.01123769,
      "auxiliary_loss_mlp": 0.01001297,
      "balance_loss_clip": 1.03090024,
      "balance_loss_mlp": 1.00003898,
      "epoch": 0.2901460951121265,
      "flos": 71106036013440.0,
      "grad_norm": 0.8204159031595429,
      "language_loss": 0.5985108,
      "learning_rate": 3.3312612010805917e-06,
      "loss": 0.61976141,
      "num_input_tokens_seen": 51720550,
      "step": 2413,
      "time_per_iteration": 3.2608885765075684
    },
    {
      "auxiliary_loss_clip": 0.01195386,
      "auxiliary_loss_mlp": 0.0103298,
      "balance_loss_clip": 1.05477083,
      "balance_loss_mlp": 1.02392578,
      "epoch": 0.2902663380027656,
      "flos": 32160986081280.0,
      "grad_norm": 1.8321509196556687,
      "language_loss": 0.69940901,
      "learning_rate": 3.330679768080858e-06,
      "loss": 0.72169268,
      "num_input_tokens_seen": 51744435,
      "step": 2414,
      "time_per_iteration": 2.759040594100952
    },
    {
      "auxiliary_loss_clip": 0.01205874,
      "auxiliary_loss_mlp": 0.01033575,
      "balance_loss_clip": 1.06105781,
      "balance_loss_mlp": 1.02407384,
      "epoch": 0.2903865808934047,
      "flos": 29351658539520.0,
      "grad_norm": 2.2362666174916996,
      "language_loss": 0.83307803,
      "learning_rate": 3.3300981332180627e-06,
      "loss": 0.85547256,
      "num_input_tokens_seen": 51763640,
      "step": 2415,
      "time_per_iteration": 2.8976175785064697
    },
    {
      "auxiliary_loss_clip": 0.01216881,
      "auxiliary_loss_mlp": 0.01026033,
      "balance_loss_clip": 1.05641222,
      "balance_loss_mlp": 1.01726508,
      "epoch": 0.29050682378404374,
      "flos": 17089026647040.0,
      "grad_norm": 3.1103307118282784,
      "language_loss": 0.79539466,
      "learning_rate": 3.3295162965804373e-06,
      "loss": 0.81782383,
      "num_input_tokens_seen": 51782135,
      "step": 2416,
      "time_per_iteration": 2.8169729709625244
    },
    {
      "auxiliary_loss_clip": 0.01206524,
      "auxiliary_loss_mlp": 0.01024781,
      "balance_loss_clip": 1.05608881,
      "balance_loss_mlp": 1.01582801,
      "epoch": 0.29062706667468285,
      "flos": 17858233422720.0,
      "grad_norm": 2.4300008839748584,
      "language_loss": 0.78262293,
      "learning_rate": 3.328934258256247e-06,
      "loss": 0.80493593,
      "num_input_tokens_seen": 51800200,
      "step": 2417,
      "time_per_iteration": 3.602560520172119
    },
    {
      "auxiliary_loss_clip": 0.01205359,
      "auxiliary_loss_mlp": 0.01026353,
      "balance_loss_clip": 1.05644083,
      "balance_loss_mlp": 1.01591039,
      "epoch": 0.29074730956532197,
      "flos": 24279815174400.0,
      "grad_norm": 2.869472134580752,
      "language_loss": 0.6739434,
      "learning_rate": 3.3283520183337856e-06,
      "loss": 0.69626057,
      "num_input_tokens_seen": 51819905,
      "step": 2418,
      "time_per_iteration": 3.5962276458740234
    },
    {
      "auxiliary_loss_clip": 0.01203372,
      "auxiliary_loss_mlp": 0.01028711,
      "balance_loss_clip": 1.05554748,
      "balance_loss_mlp": 1.01946032,
      "epoch": 0.290867552455961,
      "flos": 22340961826560.0,
      "grad_norm": 2.0853466864064805,
      "language_loss": 0.69261885,
      "learning_rate": 3.3277695769013797e-06,
      "loss": 0.71493971,
      "num_input_tokens_seen": 51839350,
      "step": 2419,
      "time_per_iteration": 2.7689738273620605
    },
    {
      "auxiliary_loss_clip": 0.01207656,
      "auxiliary_loss_mlp": 0.01028995,
      "balance_loss_clip": 1.05974889,
      "balance_loss_mlp": 1.01994157,
      "epoch": 0.29098779534660013,
      "flos": 23186155824000.0,
      "grad_norm": 2.4147106765558912,
      "language_loss": 0.774984,
      "learning_rate": 3.327186934047385e-06,
      "loss": 0.79735053,
      "num_input_tokens_seen": 51858045,
      "step": 2420,
      "time_per_iteration": 2.6555535793304443
    },
    {
      "auxiliary_loss_clip": 0.01194975,
      "auxiliary_loss_mlp": 0.01032396,
      "balance_loss_clip": 1.0505619,
      "balance_loss_mlp": 1.02319336,
      "epoch": 0.29110803823723924,
      "flos": 15304194817920.0,
      "grad_norm": 1.7577624021291944,
      "language_loss": 0.65893221,
      "learning_rate": 3.3266040898601877e-06,
      "loss": 0.68120593,
      "num_input_tokens_seen": 51875880,
      "step": 2421,
      "time_per_iteration": 2.691610097885132
    },
    {
      "auxiliary_loss_clip": 0.01203563,
      "auxiliary_loss_mlp": 0.01029378,
      "balance_loss_clip": 1.04927361,
      "balance_loss_mlp": 1.02003169,
      "epoch": 0.2912282811278783,
      "flos": 22595352923520.0,
      "grad_norm": 2.2126935956346006,
      "language_loss": 0.78219962,
      "learning_rate": 3.3260210444282045e-06,
      "loss": 0.80452901,
      "num_input_tokens_seen": 51893835,
      "step": 2422,
      "time_per_iteration": 2.683065414428711
    },
    {
      "auxiliary_loss_clip": 0.01196095,
      "auxiliary_loss_mlp": 0.01032909,
      "balance_loss_clip": 1.05582857,
      "balance_loss_mlp": 1.02390265,
      "epoch": 0.2913485240185174,
      "flos": 24497900599680.0,
      "grad_norm": 2.1655326650704025,
      "language_loss": 0.73357916,
      "learning_rate": 3.325437797839883e-06,
      "loss": 0.75586921,
      "num_input_tokens_seen": 51912205,
      "step": 2423,
      "time_per_iteration": 3.6424505710601807
    },
    {
      "auxiliary_loss_clip": 0.01202989,
      "auxiliary_loss_mlp": 0.0103133,
      "balance_loss_clip": 1.0588522,
      "balance_loss_mlp": 1.02207899,
      "epoch": 0.2914687669091565,
      "flos": 17931024334080.0,
      "grad_norm": 2.380570398315337,
      "language_loss": 0.75088519,
      "learning_rate": 3.3248543501837015e-06,
      "loss": 0.77322841,
      "num_input_tokens_seen": 51929410,
      "step": 2424,
      "time_per_iteration": 3.5220565795898438
    },
    {
      "auxiliary_loss_clip": 0.01214151,
      "auxiliary_loss_mlp": 0.0103006,
      "balance_loss_clip": 1.05434489,
      "balance_loss_mlp": 1.02092206,
      "epoch": 0.2915890097997956,
      "flos": 22529313768960.0,
      "grad_norm": 1.8193098501399085,
      "language_loss": 0.77356708,
      "learning_rate": 3.3242707015481684e-06,
      "loss": 0.79600918,
      "num_input_tokens_seen": 51949345,
      "step": 2425,
      "time_per_iteration": 2.7089643478393555
    },
    {
      "auxiliary_loss_clip": 0.01203916,
      "auxiliary_loss_mlp": 0.01030551,
      "balance_loss_clip": 1.05240548,
      "balance_loss_mlp": 1.02234936,
      "epoch": 0.2917092526904347,
      "flos": 13845216193920.0,
      "grad_norm": 2.2049483827428564,
      "language_loss": 0.80961591,
      "learning_rate": 3.323686852021823e-06,
      "loss": 0.83196056,
      "num_input_tokens_seen": 51966855,
      "step": 2426,
      "time_per_iteration": 2.8004419803619385
    },
    {
      "auxiliary_loss_clip": 0.0120982,
      "auxiliary_loss_mlp": 0.01029555,
      "balance_loss_clip": 1.05081177,
      "balance_loss_mlp": 1.02051854,
      "epoch": 0.2918294955810738,
      "flos": 22674859678080.0,
      "grad_norm": 2.1888601443141815,
      "language_loss": 0.79788166,
      "learning_rate": 3.323102801693235e-06,
      "loss": 0.82027543,
      "num_input_tokens_seen": 51985620,
      "step": 2427,
      "time_per_iteration": 2.7995760440826416
    },
    {
      "auxiliary_loss_clip": 0.0119811,
      "auxiliary_loss_mlp": 0.01027421,
      "balance_loss_clip": 1.05583799,
      "balance_loss_mlp": 1.01845658,
      "epoch": 0.29194973847171285,
      "flos": 23438284364160.0,
      "grad_norm": 3.2250395188356764,
      "language_loss": 0.80604476,
      "learning_rate": 3.322518550651003e-06,
      "loss": 0.8283,
      "num_input_tokens_seen": 52004930,
      "step": 2428,
      "time_per_iteration": 2.8883180618286133
    },
    {
      "auxiliary_loss_clip": 0.01210931,
      "auxiliary_loss_mlp": 0.01032871,
      "balance_loss_clip": 1.05320561,
      "balance_loss_mlp": 1.02395964,
      "epoch": 0.29206998136235196,
      "flos": 21909064694400.0,
      "grad_norm": 1.771492162948512,
      "language_loss": 0.81310797,
      "learning_rate": 3.3219340989837586e-06,
      "loss": 0.8355459,
      "num_input_tokens_seen": 52024920,
      "step": 2429,
      "time_per_iteration": 2.6568803787231445
    },
    {
      "auxiliary_loss_clip": 0.01207295,
      "auxiliary_loss_mlp": 0.01029946,
      "balance_loss_clip": 1.05730927,
      "balance_loss_mlp": 1.02142859,
      "epoch": 0.292190224252991,
      "flos": 23215925220480.0,
      "grad_norm": 2.0486668803376764,
      "language_loss": 0.80827248,
      "learning_rate": 3.3213494467801625e-06,
      "loss": 0.83064485,
      "num_input_tokens_seen": 52044095,
      "step": 2430,
      "time_per_iteration": 2.686243772506714
    },
    {
      "auxiliary_loss_clip": 0.01213475,
      "auxiliary_loss_mlp": 0.01028627,
      "balance_loss_clip": 1.04491079,
      "balance_loss_mlp": 1.0196749,
      "epoch": 0.2923104671436301,
      "flos": 20740818752640.0,
      "grad_norm": 1.975535472995426,
      "language_loss": 0.71520901,
      "learning_rate": 3.3207645941289063e-06,
      "loss": 0.73763001,
      "num_input_tokens_seen": 52062440,
      "step": 2431,
      "time_per_iteration": 2.7805142402648926
    },
    {
      "auxiliary_loss_clip": 0.01203943,
      "auxiliary_loss_mlp": 0.01208475,
      "balance_loss_clip": 1.05906773,
      "balance_loss_mlp": 1.00054872,
      "epoch": 0.29243071003426924,
      "flos": 35809114999680.0,
      "grad_norm": 2.139480122522967,
      "language_loss": 0.80270839,
      "learning_rate": 3.320179541118711e-06,
      "loss": 0.82683253,
      "num_input_tokens_seen": 52084940,
      "step": 2432,
      "time_per_iteration": 2.7580997943878174
    },
    {
      "auxiliary_loss_clip": 0.01117175,
      "auxiliary_loss_mlp": 0.0101559,
      "balance_loss_clip": 1.03315711,
      "balance_loss_mlp": 1.01427889,
      "epoch": 0.2925509529249083,
      "flos": 58081598524800.0,
      "grad_norm": 1.0209069117481167,
      "language_loss": 0.60269248,
      "learning_rate": 3.3195942878383293e-06,
      "loss": 0.6240201,
      "num_input_tokens_seen": 52141040,
      "step": 2433,
      "time_per_iteration": 3.19176983833313
    },
    {
      "auxiliary_loss_clip": 0.01207219,
      "auxiliary_loss_mlp": 0.01028411,
      "balance_loss_clip": 1.05880821,
      "balance_loss_mlp": 1.01855206,
      "epoch": 0.2926711958155474,
      "flos": 21397122103680.0,
      "grad_norm": 4.556460999656518,
      "language_loss": 0.78398001,
      "learning_rate": 3.319008834376543e-06,
      "loss": 0.80633634,
      "num_input_tokens_seen": 52160730,
      "step": 2434,
      "time_per_iteration": 2.660897970199585
    },
    {
      "auxiliary_loss_clip": 0.01211163,
      "auxiliary_loss_mlp": 0.01025534,
      "balance_loss_clip": 1.04937506,
      "balance_loss_mlp": 1.01682019,
      "epoch": 0.2927914387061865,
      "flos": 23185796688000.0,
      "grad_norm": 2.3283110082217875,
      "language_loss": 0.88455045,
      "learning_rate": 3.3184231808221654e-06,
      "loss": 0.90691733,
      "num_input_tokens_seen": 52175055,
      "step": 2435,
      "time_per_iteration": 2.720714569091797
    },
    {
      "auxiliary_loss_clip": 0.01208191,
      "auxiliary_loss_mlp": 0.01031335,
      "balance_loss_clip": 1.05482197,
      "balance_loss_mlp": 1.02203071,
      "epoch": 0.29291168159682557,
      "flos": 22455553190400.0,
      "grad_norm": 3.3260020778933157,
      "language_loss": 0.62815118,
      "learning_rate": 3.3178373272640394e-06,
      "loss": 0.65054643,
      "num_input_tokens_seen": 52194150,
      "step": 2436,
      "time_per_iteration": 2.710026741027832
    },
    {
      "auxiliary_loss_clip": 0.01201954,
      "auxiliary_loss_mlp": 0.01033846,
      "balance_loss_clip": 1.06039298,
      "balance_loss_mlp": 1.02525687,
      "epoch": 0.2930319244874647,
      "flos": 21170632896000.0,
      "grad_norm": 3.01755541259153,
      "language_loss": 0.85428047,
      "learning_rate": 3.3172512737910387e-06,
      "loss": 0.87663853,
      "num_input_tokens_seen": 52211660,
      "step": 2437,
      "time_per_iteration": 2.7375829219818115
    },
    {
      "auxiliary_loss_clip": 0.01203821,
      "auxiliary_loss_mlp": 0.01032923,
      "balance_loss_clip": 1.05516946,
      "balance_loss_mlp": 1.02366638,
      "epoch": 0.2931521673781038,
      "flos": 31357843931520.0,
      "grad_norm": 2.280111620914061,
      "language_loss": 0.88757217,
      "learning_rate": 3.3166650204920674e-06,
      "loss": 0.90993965,
      "num_input_tokens_seen": 52232830,
      "step": 2438,
      "time_per_iteration": 2.6703829765319824
    },
    {
      "auxiliary_loss_clip": 0.01202396,
      "auxiliary_loss_mlp": 0.01029188,
      "balance_loss_clip": 1.05775881,
      "balance_loss_mlp": 1.01971114,
      "epoch": 0.29327241026874284,
      "flos": 24200990778240.0,
      "grad_norm": 2.2674288975558476,
      "language_loss": 0.81728506,
      "learning_rate": 3.316078567456059e-06,
      "loss": 0.83960086,
      "num_input_tokens_seen": 52250670,
      "step": 2439,
      "time_per_iteration": 2.7818968296051025
    },
    {
      "auxiliary_loss_clip": 0.01216847,
      "auxiliary_loss_mlp": 0.0102651,
      "balance_loss_clip": 1.04998231,
      "balance_loss_mlp": 1.01810551,
      "epoch": 0.29339265315938196,
      "flos": 24242611662720.0,
      "grad_norm": 1.6810794029487421,
      "language_loss": 0.75709271,
      "learning_rate": 3.3154919147719786e-06,
      "loss": 0.77952629,
      "num_input_tokens_seen": 52271685,
      "step": 2440,
      "time_per_iteration": 2.7549397945404053
    },
    {
      "auxiliary_loss_clip": 0.01202111,
      "auxiliary_loss_mlp": 0.01028506,
      "balance_loss_clip": 1.05630994,
      "balance_loss_mlp": 1.01940393,
      "epoch": 0.29351289605002107,
      "flos": 16946641134720.0,
      "grad_norm": 2.4574099917840724,
      "language_loss": 0.86786997,
      "learning_rate": 3.31490506252882e-06,
      "loss": 0.89017618,
      "num_input_tokens_seen": 52291065,
      "step": 2441,
      "time_per_iteration": 2.644594669342041
    },
    {
      "auxiliary_loss_clip": 0.01197028,
      "auxiliary_loss_mlp": 0.01029085,
      "balance_loss_clip": 1.04889548,
      "balance_loss_mlp": 1.020854,
      "epoch": 0.2936331389406601,
      "flos": 19829082810240.0,
      "grad_norm": 1.7367365916528272,
      "language_loss": 0.84462988,
      "learning_rate": 3.31431801081561e-06,
      "loss": 0.86689103,
      "num_input_tokens_seen": 52310000,
      "step": 2442,
      "time_per_iteration": 2.690420150756836
    },
    {
      "auxiliary_loss_clip": 0.01117032,
      "auxiliary_loss_mlp": 0.01003167,
      "balance_loss_clip": 1.03299844,
      "balance_loss_mlp": 1.00215387,
      "epoch": 0.29375338183129923,
      "flos": 71416844398080.0,
      "grad_norm": 0.8929821134313681,
      "language_loss": 0.67843765,
      "learning_rate": 3.313730759721402e-06,
      "loss": 0.69963968,
      "num_input_tokens_seen": 52372930,
      "step": 2443,
      "time_per_iteration": 3.2751517295837402
    },
    {
      "auxiliary_loss_clip": 0.01200068,
      "auxiliary_loss_mlp": 0.01030895,
      "balance_loss_clip": 1.05378771,
      "balance_loss_mlp": 1.02153754,
      "epoch": 0.29387362472193834,
      "flos": 22054502862720.0,
      "grad_norm": 2.023468057380383,
      "language_loss": 0.86134291,
      "learning_rate": 3.313143309335282e-06,
      "loss": 0.88365257,
      "num_input_tokens_seen": 52391420,
      "step": 2444,
      "time_per_iteration": 3.6639180183410645
    },
    {
      "auxiliary_loss_clip": 0.01205303,
      "auxiliary_loss_mlp": 0.01034119,
      "balance_loss_clip": 1.05408287,
      "balance_loss_mlp": 1.02519,
      "epoch": 0.2939938676125774,
      "flos": 22966418373120.0,
      "grad_norm": 1.8640397211393385,
      "language_loss": 0.84722912,
      "learning_rate": 3.3125556597463665e-06,
      "loss": 0.86962336,
      "num_input_tokens_seen": 52410725,
      "step": 2445,
      "time_per_iteration": 3.668156623840332
    },
    {
      "auxiliary_loss_clip": 0.01200447,
      "auxiliary_loss_mlp": 0.01030767,
      "balance_loss_clip": 1.05739474,
      "balance_loss_mlp": 1.02224326,
      "epoch": 0.2941141105032165,
      "flos": 31358705857920.0,
      "grad_norm": 1.5769704461972431,
      "language_loss": 0.66229957,
      "learning_rate": 3.311967811043801e-06,
      "loss": 0.68461174,
      "num_input_tokens_seen": 52432645,
      "step": 2446,
      "time_per_iteration": 2.7244434356689453
    },
    {
      "auxiliary_loss_clip": 0.0120344,
      "auxiliary_loss_mlp": 0.01029463,
      "balance_loss_clip": 1.05826783,
      "balance_loss_mlp": 1.0203495,
      "epoch": 0.29423435339385556,
      "flos": 23222138273280.0,
      "grad_norm": 3.088654768712617,
      "language_loss": 0.82280374,
      "learning_rate": 3.3113797633167617e-06,
      "loss": 0.84513271,
      "num_input_tokens_seen": 52450940,
      "step": 2447,
      "time_per_iteration": 2.8256821632385254
    },
    {
      "auxiliary_loss_clip": 0.0119977,
      "auxiliary_loss_mlp": 0.01029014,
      "balance_loss_clip": 1.05677247,
      "balance_loss_mlp": 1.01971567,
      "epoch": 0.2943545962844947,
      "flos": 26864054138880.0,
      "grad_norm": 2.980250073253398,
      "language_loss": 0.69431448,
      "learning_rate": 3.310791516654455e-06,
      "loss": 0.71660233,
      "num_input_tokens_seen": 52468000,
      "step": 2448,
      "time_per_iteration": 2.711585283279419
    },
    {
      "auxiliary_loss_clip": 0.01213825,
      "auxiliary_loss_mlp": 0.01039882,
      "balance_loss_clip": 1.05231786,
      "balance_loss_mlp": 1.02949309,
      "epoch": 0.2944748391751338,
      "flos": 20231677422720.0,
      "grad_norm": 3.089630500171285,
      "language_loss": 0.80229449,
      "learning_rate": 3.3102030711461177e-06,
      "loss": 0.8248316,
      "num_input_tokens_seen": 52487575,
      "step": 2449,
      "time_per_iteration": 3.6695809364318848
    },
    {
      "auxiliary_loss_clip": 0.01210061,
      "auxiliary_loss_mlp": 0.01026195,
      "balance_loss_clip": 1.05354309,
      "balance_loss_mlp": 1.01705217,
      "epoch": 0.29459508206577284,
      "flos": 15960965045760.0,
      "grad_norm": 2.0088856082480486,
      "language_loss": 0.68142611,
      "learning_rate": 3.3096144268810156e-06,
      "loss": 0.70378864,
      "num_input_tokens_seen": 52506335,
      "step": 2450,
      "time_per_iteration": 3.5908102989196777
    },
    {
      "auxiliary_loss_clip": 0.01194071,
      "auxiliary_loss_mlp": 0.01029067,
      "balance_loss_clip": 1.05509293,
      "balance_loss_mlp": 1.0199002,
      "epoch": 0.29471532495641195,
      "flos": 20412882558720.0,
      "grad_norm": 2.438574404727516,
      "language_loss": 0.73062479,
      "learning_rate": 3.3090255839484462e-06,
      "loss": 0.75285614,
      "num_input_tokens_seen": 52524330,
      "step": 2451,
      "time_per_iteration": 2.640294313430786
    },
    {
      "auxiliary_loss_clip": 0.01206344,
      "auxiliary_loss_mlp": 0.01027992,
      "balance_loss_clip": 1.05309653,
      "balance_loss_mlp": 1.01871717,
      "epoch": 0.29483556784705106,
      "flos": 20376576887040.0,
      "grad_norm": 4.908331825394441,
      "language_loss": 0.85568428,
      "learning_rate": 3.3084365424377366e-06,
      "loss": 0.87802768,
      "num_input_tokens_seen": 52543095,
      "step": 2452,
      "time_per_iteration": 2.751685857772827
    },
    {
      "auxiliary_loss_clip": 0.01130596,
      "auxiliary_loss_mlp": 0.01004604,
      "balance_loss_clip": 1.03314197,
      "balance_loss_mlp": 1.00311399,
      "epoch": 0.2949558107376901,
      "flos": 68555660595840.0,
      "grad_norm": 0.7278552892562937,
      "language_loss": 0.55975068,
      "learning_rate": 3.307847302438245e-06,
      "loss": 0.58110273,
      "num_input_tokens_seen": 52597075,
      "step": 2453,
      "time_per_iteration": 3.133702278137207
    },
    {
      "auxiliary_loss_clip": 0.0119394,
      "auxiliary_loss_mlp": 0.01027936,
      "balance_loss_clip": 1.04630017,
      "balance_loss_mlp": 1.01780891,
      "epoch": 0.2950760536283292,
      "flos": 16107085572480.0,
      "grad_norm": 1.995397190923732,
      "language_loss": 0.7787441,
      "learning_rate": 3.3072578640393562e-06,
      "loss": 0.80096292,
      "num_input_tokens_seen": 52614410,
      "step": 2454,
      "time_per_iteration": 2.690471887588501
    },
    {
      "auxiliary_loss_clip": 0.01206357,
      "auxiliary_loss_mlp": 0.01029935,
      "balance_loss_clip": 1.05607367,
      "balance_loss_mlp": 1.02048802,
      "epoch": 0.29519629651896834,
      "flos": 20483626394880.0,
      "grad_norm": 1.9122317169765306,
      "language_loss": 0.79618454,
      "learning_rate": 3.3066682273304886e-06,
      "loss": 0.81854749,
      "num_input_tokens_seen": 52632055,
      "step": 2455,
      "time_per_iteration": 2.668100595474243
    },
    {
      "auxiliary_loss_clip": 0.01208333,
      "auxiliary_loss_mlp": 0.01208747,
      "balance_loss_clip": 1.05877578,
      "balance_loss_mlp": 1.00069809,
      "epoch": 0.2953165394096074,
      "flos": 18916484941440.0,
      "grad_norm": 2.531336190590645,
      "language_loss": 0.79685932,
      "learning_rate": 3.3060783924010904e-06,
      "loss": 0.82103014,
      "num_input_tokens_seen": 52649980,
      "step": 2456,
      "time_per_iteration": 2.6190707683563232
    },
    {
      "auxiliary_loss_clip": 0.01210027,
      "auxiliary_loss_mlp": 0.01032808,
      "balance_loss_clip": 1.05573964,
      "balance_loss_mlp": 1.02341998,
      "epoch": 0.2954367823002465,
      "flos": 20624467622400.0,
      "grad_norm": 2.375690920945258,
      "language_loss": 0.85071069,
      "learning_rate": 3.3054883593406387e-06,
      "loss": 0.87313902,
      "num_input_tokens_seen": 52664730,
      "step": 2457,
      "time_per_iteration": 2.6834542751312256
    },
    {
      "auxiliary_loss_clip": 0.01207807,
      "auxiliary_loss_mlp": 0.01029061,
      "balance_loss_clip": 1.05621481,
      "balance_loss_mlp": 1.02018011,
      "epoch": 0.2955570251908856,
      "flos": 31175525473920.0,
      "grad_norm": 2.13528221381753,
      "language_loss": 0.64770591,
      "learning_rate": 3.3048981282386404e-06,
      "loss": 0.67007458,
      "num_input_tokens_seen": 52686040,
      "step": 2458,
      "time_per_iteration": 2.7262587547302246
    },
    {
      "auxiliary_loss_clip": 0.01196115,
      "auxiliary_loss_mlp": 0.01026733,
      "balance_loss_clip": 1.05236101,
      "balance_loss_mlp": 1.01823926,
      "epoch": 0.29567726808152467,
      "flos": 21650328051840.0,
      "grad_norm": 2.1553861093078157,
      "language_loss": 0.82638389,
      "learning_rate": 3.304307699184634e-06,
      "loss": 0.84861231,
      "num_input_tokens_seen": 52704630,
      "step": 2459,
      "time_per_iteration": 2.707519292831421
    },
    {
      "auxiliary_loss_clip": 0.01208418,
      "auxiliary_loss_mlp": 0.01030552,
      "balance_loss_clip": 1.05882692,
      "balance_loss_mlp": 1.0221771,
      "epoch": 0.2957975109721638,
      "flos": 24243868638720.0,
      "grad_norm": 1.890392225631424,
      "language_loss": 0.78733653,
      "learning_rate": 3.3037170722681866e-06,
      "loss": 0.80972624,
      "num_input_tokens_seen": 52725465,
      "step": 2460,
      "time_per_iteration": 2.6867644786834717
    },
    {
      "auxiliary_loss_clip": 0.01204379,
      "auxiliary_loss_mlp": 0.01030684,
      "balance_loss_clip": 1.05661678,
      "balance_loss_mlp": 1.02139759,
      "epoch": 0.29591775386280283,
      "flos": 13479717352320.0,
      "grad_norm": 2.108197474079222,
      "language_loss": 0.68377256,
      "learning_rate": 3.3031262475788956e-06,
      "loss": 0.70612317,
      "num_input_tokens_seen": 52742405,
      "step": 2461,
      "time_per_iteration": 2.8368897438049316
    },
    {
      "auxiliary_loss_clip": 0.01205984,
      "auxiliary_loss_mlp": 0.01030508,
      "balance_loss_clip": 1.05737293,
      "balance_loss_mlp": 1.02160335,
      "epoch": 0.29603799675344195,
      "flos": 17749783284480.0,
      "grad_norm": 2.6002588940566955,
      "language_loss": 0.7347123,
      "learning_rate": 3.3025352252063897e-06,
      "loss": 0.75707722,
      "num_input_tokens_seen": 52761100,
      "step": 2462,
      "time_per_iteration": 2.6301302909851074
    },
    {
      "auxiliary_loss_clip": 0.01206229,
      "auxiliary_loss_mlp": 0.01037204,
      "balance_loss_clip": 1.0629611,
      "balance_loss_mlp": 1.02789927,
      "epoch": 0.29615823964408106,
      "flos": 22783920347520.0,
      "grad_norm": 2.2111934393788495,
      "language_loss": 0.75035053,
      "learning_rate": 3.3019440052403252e-06,
      "loss": 0.77278489,
      "num_input_tokens_seen": 52780965,
      "step": 2463,
      "time_per_iteration": 2.6625616550445557
    },
    {
      "auxiliary_loss_clip": 0.01208845,
      "auxiliary_loss_mlp": 0.01027156,
      "balance_loss_clip": 1.05700874,
      "balance_loss_mlp": 1.01816773,
      "epoch": 0.2962784825347201,
      "flos": 23514199758720.0,
      "grad_norm": 2.917663179131767,
      "language_loss": 0.70850718,
      "learning_rate": 3.30135258777039e-06,
      "loss": 0.73086715,
      "num_input_tokens_seen": 52800335,
      "step": 2464,
      "time_per_iteration": 2.6561038494110107
    },
    {
      "auxiliary_loss_clip": 0.01210209,
      "auxiliary_loss_mlp": 0.01208784,
      "balance_loss_clip": 1.0583396,
      "balance_loss_mlp": 1.00070059,
      "epoch": 0.2963987254253592,
      "flos": 16362769559040.0,
      "grad_norm": 1.9930320774846442,
      "language_loss": 0.70387632,
      "learning_rate": 3.3007609728863024e-06,
      "loss": 0.72806621,
      "num_input_tokens_seen": 52818425,
      "step": 2465,
      "time_per_iteration": 2.7678396701812744
    },
    {
      "auxiliary_loss_clip": 0.01207178,
      "auxiliary_loss_mlp": 0.01028297,
      "balance_loss_clip": 1.05451667,
      "balance_loss_mlp": 1.01885545,
      "epoch": 0.29651896831599833,
      "flos": 33472263980160.0,
      "grad_norm": 2.250196785883734,
      "language_loss": 0.73031187,
      "learning_rate": 3.300169160677809e-06,
      "loss": 0.75266665,
      "num_input_tokens_seen": 52842340,
      "step": 2466,
      "time_per_iteration": 2.8706672191619873
    },
    {
      "auxiliary_loss_clip": 0.01217183,
      "auxiliary_loss_mlp": 0.01026933,
      "balance_loss_clip": 1.05728495,
      "balance_loss_mlp": 1.01746786,
      "epoch": 0.2966392112066374,
      "flos": 23805363404160.0,
      "grad_norm": 2.4633784500105547,
      "language_loss": 0.78078175,
      "learning_rate": 3.2995771512346878e-06,
      "loss": 0.80322289,
      "num_input_tokens_seen": 52860690,
      "step": 2467,
      "time_per_iteration": 2.6689562797546387
    },
    {
      "auxiliary_loss_clip": 0.01208175,
      "auxiliary_loss_mlp": 0.01209335,
      "balance_loss_clip": 1.06295729,
      "balance_loss_mlp": 1.00083125,
      "epoch": 0.2967594540972765,
      "flos": 19938466702080.0,
      "grad_norm": 2.710548320575444,
      "language_loss": 0.72376078,
      "learning_rate": 3.298984944646746e-06,
      "loss": 0.74793583,
      "num_input_tokens_seen": 52879370,
      "step": 2468,
      "time_per_iteration": 2.674994945526123
    },
    {
      "auxiliary_loss_clip": 0.01209679,
      "auxiliary_loss_mlp": 0.01208198,
      "balance_loss_clip": 1.06155849,
      "balance_loss_mlp": 1.00075626,
      "epoch": 0.2968796969879156,
      "flos": 23732823888000.0,
      "grad_norm": 2.529230348108541,
      "language_loss": 0.81444573,
      "learning_rate": 3.298392541003822e-06,
      "loss": 0.83862448,
      "num_input_tokens_seen": 52898775,
      "step": 2469,
      "time_per_iteration": 2.649656295776367
    },
    {
      "auxiliary_loss_clip": 0.01205119,
      "auxiliary_loss_mlp": 0.01029716,
      "balance_loss_clip": 1.05778074,
      "balance_loss_mlp": 1.02113843,
      "epoch": 0.29699993987855466,
      "flos": 22893699288960.0,
      "grad_norm": 4.544780659870459,
      "language_loss": 0.89624614,
      "learning_rate": 3.2977999403957806e-06,
      "loss": 0.91859448,
      "num_input_tokens_seen": 52917535,
      "step": 2470,
      "time_per_iteration": 3.6459977626800537
    },
    {
      "auxiliary_loss_clip": 0.01205301,
      "auxiliary_loss_mlp": 0.01030352,
      "balance_loss_clip": 1.06301928,
      "balance_loss_mlp": 1.02062464,
      "epoch": 0.2971201827691938,
      "flos": 33832555349760.0,
      "grad_norm": 1.8243636149606324,
      "language_loss": 0.67168367,
      "learning_rate": 3.2972071429125207e-06,
      "loss": 0.69404018,
      "num_input_tokens_seen": 52938755,
      "step": 2471,
      "time_per_iteration": 2.729766607284546
    },
    {
      "auxiliary_loss_clip": 0.01204687,
      "auxiliary_loss_mlp": 0.01031351,
      "balance_loss_clip": 1.05462706,
      "balance_loss_mlp": 1.02164745,
      "epoch": 0.2972404256598329,
      "flos": 22054359208320.0,
      "grad_norm": 2.163262822039383,
      "language_loss": 0.88534617,
      "learning_rate": 3.2966141486439682e-06,
      "loss": 0.90770662,
      "num_input_tokens_seen": 52957945,
      "step": 2472,
      "time_per_iteration": 3.6947667598724365
    },
    {
      "auxiliary_loss_clip": 0.01215644,
      "auxiliary_loss_mlp": 0.01028177,
      "balance_loss_clip": 1.05004573,
      "balance_loss_mlp": 1.01813984,
      "epoch": 0.29736066855047194,
      "flos": 31978595796480.0,
      "grad_norm": 2.5560610727936184,
      "language_loss": 0.64541769,
      "learning_rate": 3.29602095768008e-06,
      "loss": 0.66785592,
      "num_input_tokens_seen": 52978460,
      "step": 2473,
      "time_per_iteration": 2.905461072921753
    },
    {
      "auxiliary_loss_clip": 0.0120111,
      "auxiliary_loss_mlp": 0.01027148,
      "balance_loss_clip": 1.05821371,
      "balance_loss_mlp": 1.01861298,
      "epoch": 0.29748091144111105,
      "flos": 33510401245440.0,
      "grad_norm": 7.191520284788893,
      "language_loss": 0.64004529,
      "learning_rate": 3.2954275701108437e-06,
      "loss": 0.66232783,
      "num_input_tokens_seen": 52999640,
      "step": 2474,
      "time_per_iteration": 2.7439029216766357
    },
    {
      "auxiliary_loss_clip": 0.01209293,
      "auxiliary_loss_mlp": 0.01025367,
      "balance_loss_clip": 1.05338383,
      "balance_loss_mlp": 1.01618171,
      "epoch": 0.29760115433175016,
      "flos": 41283373409280.0,
      "grad_norm": 2.657830963229873,
      "language_loss": 0.68697834,
      "learning_rate": 3.294833986026275e-06,
      "loss": 0.70932496,
      "num_input_tokens_seen": 53022880,
      "step": 2475,
      "time_per_iteration": 2.93692946434021
    },
    {
      "auxiliary_loss_clip": 0.01202565,
      "auxiliary_loss_mlp": 0.01023755,
      "balance_loss_clip": 1.05521226,
      "balance_loss_mlp": 1.01490998,
      "epoch": 0.2977213972223892,
      "flos": 24493339572480.0,
      "grad_norm": 2.489870460544596,
      "language_loss": 0.8528049,
      "learning_rate": 3.29424020551642e-06,
      "loss": 0.87506807,
      "num_input_tokens_seen": 53041515,
      "step": 2476,
      "time_per_iteration": 3.6623518466949463
    },
    {
      "auxiliary_loss_clip": 0.01208803,
      "auxiliary_loss_mlp": 0.01032861,
      "balance_loss_clip": 1.061746,
      "balance_loss_mlp": 1.02262676,
      "epoch": 0.2978416401130283,
      "flos": 21285116519040.0,
      "grad_norm": 2.062910070041738,
      "language_loss": 0.72335893,
      "learning_rate": 3.2936462286713546e-06,
      "loss": 0.74577558,
      "num_input_tokens_seen": 53059865,
      "step": 2477,
      "time_per_iteration": 3.544877767562866
    },
    {
      "auxiliary_loss_clip": 0.01206426,
      "auxiliary_loss_mlp": 0.01029274,
      "balance_loss_clip": 1.05915332,
      "balance_loss_mlp": 1.01982069,
      "epoch": 0.2979618830036674,
      "flos": 25772154554880.0,
      "grad_norm": 2.512514241362412,
      "language_loss": 0.77514887,
      "learning_rate": 3.2930520555811846e-06,
      "loss": 0.7975058,
      "num_input_tokens_seen": 53079490,
      "step": 2478,
      "time_per_iteration": 2.696648597717285
    },
    {
      "auxiliary_loss_clip": 0.01192535,
      "auxiliary_loss_mlp": 0.01209434,
      "balance_loss_clip": 1.04662108,
      "balance_loss_mlp": 1.00091839,
      "epoch": 0.2980821258943065,
      "flos": 23476996247040.0,
      "grad_norm": 2.4184539223936707,
      "language_loss": 0.80183685,
      "learning_rate": 3.292457686336046e-06,
      "loss": 0.82585657,
      "num_input_tokens_seen": 53098810,
      "step": 2479,
      "time_per_iteration": 2.739962577819824
    },
    {
      "auxiliary_loss_clip": 0.01127368,
      "auxiliary_loss_mlp": 0.01005682,
      "balance_loss_clip": 1.03598785,
      "balance_loss_mlp": 1.00451934,
      "epoch": 0.2982023687849456,
      "flos": 69752314195200.0,
      "grad_norm": 0.8524493815607672,
      "language_loss": 0.61197412,
      "learning_rate": 3.291863121026105e-06,
      "loss": 0.6333046,
      "num_input_tokens_seen": 53162590,
      "step": 2480,
      "time_per_iteration": 3.3398332595825195
    },
    {
      "auxiliary_loss_clip": 0.01206379,
      "auxiliary_loss_mlp": 0.01031938,
      "balance_loss_clip": 1.05942702,
      "balance_loss_mlp": 1.02308083,
      "epoch": 0.29832261167558466,
      "flos": 29825930741760.0,
      "grad_norm": 3.1828931214178153,
      "language_loss": 0.76987845,
      "learning_rate": 3.2912683597415547e-06,
      "loss": 0.7922616,
      "num_input_tokens_seen": 53186675,
      "step": 2481,
      "time_per_iteration": 2.769357442855835
    },
    {
      "auxiliary_loss_clip": 0.01212447,
      "auxiliary_loss_mlp": 0.01034247,
      "balance_loss_clip": 1.05562925,
      "balance_loss_mlp": 1.02515697,
      "epoch": 0.29844285456622377,
      "flos": 33910158683520.0,
      "grad_norm": 2.315454063640958,
      "language_loss": 0.78534132,
      "learning_rate": 3.2906734025726213e-06,
      "loss": 0.80780828,
      "num_input_tokens_seen": 53205940,
      "step": 2482,
      "time_per_iteration": 2.7857441902160645
    },
    {
      "auxiliary_loss_clip": 0.01212928,
      "auxiliary_loss_mlp": 0.01034562,
      "balance_loss_clip": 1.06135416,
      "balance_loss_mlp": 1.02528095,
      "epoch": 0.2985630974568629,
      "flos": 23876933253120.0,
      "grad_norm": 1.8839772817802245,
      "language_loss": 0.87926531,
      "learning_rate": 3.290078249609559e-06,
      "loss": 0.90174019,
      "num_input_tokens_seen": 53225360,
      "step": 2483,
      "time_per_iteration": 2.743197202682495
    },
    {
      "auxiliary_loss_clip": 0.01204054,
      "auxiliary_loss_mlp": 0.01030529,
      "balance_loss_clip": 1.06249273,
      "balance_loss_mlp": 1.02118278,
      "epoch": 0.29868334034750194,
      "flos": 21799106184960.0,
      "grad_norm": 2.5350772641574495,
      "language_loss": 0.88404715,
      "learning_rate": 3.2894829009426514e-06,
      "loss": 0.90639293,
      "num_input_tokens_seen": 53243195,
      "step": 2484,
      "time_per_iteration": 2.6349165439605713
    },
    {
      "auxiliary_loss_clip": 0.01203178,
      "auxiliary_loss_mlp": 0.01029699,
      "balance_loss_clip": 1.05830002,
      "balance_loss_mlp": 1.02041841,
      "epoch": 0.29880358323814105,
      "flos": 25666649331840.0,
      "grad_norm": 2.0499527752197464,
      "language_loss": 0.77992815,
      "learning_rate": 3.288887356662213e-06,
      "loss": 0.80225694,
      "num_input_tokens_seen": 53264530,
      "step": 2485,
      "time_per_iteration": 2.78562068939209
    },
    {
      "auxiliary_loss_clip": 0.01108501,
      "auxiliary_loss_mlp": 0.01005433,
      "balance_loss_clip": 1.03074169,
      "balance_loss_mlp": 1.00416374,
      "epoch": 0.29892382612878016,
      "flos": 71005846003200.0,
      "grad_norm": 0.8048692650126206,
      "language_loss": 0.59653437,
      "learning_rate": 3.288291616858588e-06,
      "loss": 0.61767375,
      "num_input_tokens_seen": 53319920,
      "step": 2486,
      "time_per_iteration": 3.0661089420318604
    },
    {
      "auxiliary_loss_clip": 0.0120604,
      "auxiliary_loss_mlp": 0.01028025,
      "balance_loss_clip": 1.05648386,
      "balance_loss_mlp": 1.01906073,
      "epoch": 0.2990440690194192,
      "flos": 25481134563840.0,
      "grad_norm": 9.467929942398401,
      "language_loss": 0.76822054,
      "learning_rate": 3.287695681622149e-06,
      "loss": 0.79056114,
      "num_input_tokens_seen": 53339270,
      "step": 2487,
      "time_per_iteration": 2.8314149379730225
    },
    {
      "auxiliary_loss_clip": 0.0121235,
      "auxiliary_loss_mlp": 0.01024655,
      "balance_loss_clip": 1.05704439,
      "balance_loss_mlp": 1.01604784,
      "epoch": 0.2991643119100583,
      "flos": 23732357011200.0,
      "grad_norm": 3.573339731306314,
      "language_loss": 0.81652945,
      "learning_rate": 3.2870995510432982e-06,
      "loss": 0.83889949,
      "num_input_tokens_seen": 53357750,
      "step": 2488,
      "time_per_iteration": 2.766568899154663
    },
    {
      "auxiliary_loss_clip": 0.01197779,
      "auxiliary_loss_mlp": 0.01029525,
      "balance_loss_clip": 1.05707085,
      "balance_loss_mlp": 1.02102518,
      "epoch": 0.29928455480069743,
      "flos": 27417545786880.0,
      "grad_norm": 4.6630414794120485,
      "language_loss": 0.77564108,
      "learning_rate": 3.2865032252124697e-06,
      "loss": 0.79791409,
      "num_input_tokens_seen": 53378265,
      "step": 2489,
      "time_per_iteration": 2.7218949794769287
    },
    {
      "auxiliary_loss_clip": 0.01206335,
      "auxiliary_loss_mlp": 0.01030615,
      "balance_loss_clip": 1.05551267,
      "balance_loss_mlp": 1.02168655,
      "epoch": 0.2994047976913365,
      "flos": 33692935184640.0,
      "grad_norm": 1.9152574444828374,
      "language_loss": 0.77360308,
      "learning_rate": 3.2859067042201243e-06,
      "loss": 0.79597259,
      "num_input_tokens_seen": 53400305,
      "step": 2490,
      "time_per_iteration": 2.781489133834839
    },
    {
      "auxiliary_loss_clip": 0.0119172,
      "auxiliary_loss_mlp": 0.01028025,
      "balance_loss_clip": 1.04820967,
      "balance_loss_mlp": 1.01898289,
      "epoch": 0.2995250405819756,
      "flos": 16763963541120.0,
      "grad_norm": 2.002783609934027,
      "language_loss": 0.78209484,
      "learning_rate": 3.2853099881567544e-06,
      "loss": 0.80429232,
      "num_input_tokens_seen": 53418705,
      "step": 2491,
      "time_per_iteration": 2.8726062774658203
    },
    {
      "auxiliary_loss_clip": 0.01199509,
      "auxiliary_loss_mlp": 0.01026815,
      "balance_loss_clip": 1.06015992,
      "balance_loss_mlp": 1.01887584,
      "epoch": 0.29964528347261465,
      "flos": 22963976248320.0,
      "grad_norm": 2.671410546038333,
      "language_loss": 0.79376656,
      "learning_rate": 3.284713077112881e-06,
      "loss": 0.81602979,
      "num_input_tokens_seen": 53438135,
      "step": 2492,
      "time_per_iteration": 2.625316619873047
    },
    {
      "auxiliary_loss_clip": 0.0121382,
      "auxiliary_loss_mlp": 0.01030468,
      "balance_loss_clip": 1.05712104,
      "balance_loss_mlp": 1.02105665,
      "epoch": 0.29976552636325376,
      "flos": 16938021870720.0,
      "grad_norm": 2.84810462405527,
      "language_loss": 0.86251521,
      "learning_rate": 3.284115971179056e-06,
      "loss": 0.88495815,
      "num_input_tokens_seen": 53452165,
      "step": 2493,
      "time_per_iteration": 2.5929722785949707
    },
    {
      "auxiliary_loss_clip": 0.01214068,
      "auxiliary_loss_mlp": 0.01028614,
      "balance_loss_clip": 1.05149484,
      "balance_loss_mlp": 1.01939917,
      "epoch": 0.2998857692538929,
      "flos": 17056455989760.0,
      "grad_norm": 1.995482645066308,
      "language_loss": 0.78798914,
      "learning_rate": 3.283518670445859e-06,
      "loss": 0.81041598,
      "num_input_tokens_seen": 53470075,
      "step": 2494,
      "time_per_iteration": 2.7158079147338867
    },
    {
      "auxiliary_loss_clip": 0.0111448,
      "auxiliary_loss_mlp": 0.01200405,
      "balance_loss_clip": 1.03414488,
      "balance_loss_mlp": 1.00018215,
      "epoch": 0.30000601214453193,
      "flos": 68831528025600.0,
      "grad_norm": 0.6932638055097045,
      "language_loss": 0.54311919,
      "learning_rate": 3.2829211750038995e-06,
      "loss": 0.56626809,
      "num_input_tokens_seen": 53538705,
      "step": 2495,
      "time_per_iteration": 3.230621099472046
    },
    {
      "auxiliary_loss_clip": 0.01204675,
      "auxiliary_loss_mlp": 0.01031424,
      "balance_loss_clip": 1.05287981,
      "balance_loss_mlp": 1.02228653,
      "epoch": 0.30012625503517104,
      "flos": 17603267708160.0,
      "grad_norm": 2.179387820087003,
      "language_loss": 0.89304352,
      "learning_rate": 3.2823234849438183e-06,
      "loss": 0.91540456,
      "num_input_tokens_seen": 53556740,
      "step": 2496,
      "time_per_iteration": 3.5750980377197266
    },
    {
      "auxiliary_loss_clip": 0.01207502,
      "auxiliary_loss_mlp": 0.01029472,
      "balance_loss_clip": 1.05671072,
      "balance_loss_mlp": 1.02069783,
      "epoch": 0.30024649792581015,
      "flos": 21252581775360.0,
      "grad_norm": 4.066772269321353,
      "language_loss": 0.75389415,
      "learning_rate": 3.2817256003562836e-06,
      "loss": 0.77626395,
      "num_input_tokens_seen": 53577115,
      "step": 2497,
      "time_per_iteration": 2.6556951999664307
    },
    {
      "auxiliary_loss_clip": 0.01213852,
      "auxiliary_loss_mlp": 0.01035947,
      "balance_loss_clip": 1.05361533,
      "balance_loss_mlp": 1.02646422,
      "epoch": 0.3003667408164492,
      "flos": 23003262748800.0,
      "grad_norm": 1.8345376120440213,
      "language_loss": 0.65822291,
      "learning_rate": 3.281127521331995e-06,
      "loss": 0.68072093,
      "num_input_tokens_seen": 53598295,
      "step": 2498,
      "time_per_iteration": 2.7477850914001465
    },
    {
      "auxiliary_loss_clip": 0.01107061,
      "auxiliary_loss_mlp": 0.01002899,
      "balance_loss_clip": 1.03152299,
      "balance_loss_mlp": 1.00160563,
      "epoch": 0.3004869837070883,
      "flos": 64232340750720.0,
      "grad_norm": 1.181866887323206,
      "language_loss": 0.60616881,
      "learning_rate": 3.2805292479616798e-06,
      "loss": 0.62726843,
      "num_input_tokens_seen": 53657160,
      "step": 2499,
      "time_per_iteration": 4.007415056228638
    },
    {
      "auxiliary_loss_clip": 0.01209624,
      "auxiliary_loss_mlp": 0.01029826,
      "balance_loss_clip": 1.0557915,
      "balance_loss_mlp": 1.02048624,
      "epoch": 0.30060722659772743,
      "flos": 26248653400320.0,
      "grad_norm": 3.71012513420842,
      "language_loss": 0.92143035,
      "learning_rate": 3.2799307803360955e-06,
      "loss": 0.94382489,
      "num_input_tokens_seen": 53673090,
      "step": 2500,
      "time_per_iteration": 2.747631072998047
    },
    {
      "auxiliary_loss_clip": 0.0119852,
      "auxiliary_loss_mlp": 0.01029154,
      "balance_loss_clip": 1.05831778,
      "balance_loss_mlp": 1.01997519,
      "epoch": 0.3007274694883665,
      "flos": 24970879912320.0,
      "grad_norm": 1.6055023915372926,
      "language_loss": 0.8172968,
      "learning_rate": 3.27933211854603e-06,
      "loss": 0.8395735,
      "num_input_tokens_seen": 53692145,
      "step": 2501,
      "time_per_iteration": 2.665766477584839
    },
    {
      "auxiliary_loss_clip": 0.01204655,
      "auxiliary_loss_mlp": 0.01025784,
      "balance_loss_clip": 1.05616653,
      "balance_loss_mlp": 1.0168196,
      "epoch": 0.3008477123790056,
      "flos": 17055845458560.0,
      "grad_norm": 4.095784551499895,
      "language_loss": 0.86927581,
      "learning_rate": 3.278733262682299e-06,
      "loss": 0.89158016,
      "num_input_tokens_seen": 53710000,
      "step": 2502,
      "time_per_iteration": 3.819514274597168
    },
    {
      "auxiliary_loss_clip": 0.01203026,
      "auxiliary_loss_mlp": 0.01027377,
      "balance_loss_clip": 1.05961895,
      "balance_loss_mlp": 1.01839423,
      "epoch": 0.3009679552696447,
      "flos": 21506398254720.0,
      "grad_norm": 2.2490829691212637,
      "language_loss": 0.82898462,
      "learning_rate": 3.2781342128357484e-06,
      "loss": 0.85128856,
      "num_input_tokens_seen": 53729355,
      "step": 2503,
      "time_per_iteration": 2.8181910514831543
    },
    {
      "auxiliary_loss_clip": 0.0120803,
      "auxiliary_loss_mlp": 0.01027852,
      "balance_loss_clip": 1.05332267,
      "balance_loss_mlp": 1.01856613,
      "epoch": 0.30108819816028376,
      "flos": 21134004001920.0,
      "grad_norm": 4.068118184507147,
      "language_loss": 0.80137944,
      "learning_rate": 3.2775349690972547e-06,
      "loss": 0.82373822,
      "num_input_tokens_seen": 53743505,
      "step": 2504,
      "time_per_iteration": 3.5495223999023438
    },
    {
      "auxiliary_loss_clip": 0.01111485,
      "auxiliary_loss_mlp": 0.01002035,
      "balance_loss_clip": 1.03469861,
      "balance_loss_mlp": 1.00093842,
      "epoch": 0.30120844105092287,
      "flos": 71126434938240.0,
      "grad_norm": 0.7707006704779209,
      "language_loss": 0.51845765,
      "learning_rate": 3.276935531557722e-06,
      "loss": 0.53959286,
      "num_input_tokens_seen": 53808725,
      "step": 2505,
      "time_per_iteration": 3.4029014110565186
    },
    {
      "auxiliary_loss_clip": 0.01215287,
      "auxiliary_loss_mlp": 0.01030998,
      "balance_loss_clip": 1.05201209,
      "balance_loss_mlp": 1.02196229,
      "epoch": 0.301328683941562,
      "flos": 20264571302400.0,
      "grad_norm": 3.426670357400372,
      "language_loss": 0.79892534,
      "learning_rate": 3.2763359003080837e-06,
      "loss": 0.82138813,
      "num_input_tokens_seen": 53825680,
      "step": 2506,
      "time_per_iteration": 2.735412120819092
    },
    {
      "auxiliary_loss_clip": 0.01115928,
      "auxiliary_loss_mlp": 0.01006456,
      "balance_loss_clip": 1.02920711,
      "balance_loss_mlp": 1.00517416,
      "epoch": 0.30144892683220104,
      "flos": 70648212240000.0,
      "grad_norm": 0.8221907072361323,
      "language_loss": 0.62446988,
      "learning_rate": 3.2757360754393047e-06,
      "loss": 0.64569372,
      "num_input_tokens_seen": 53889750,
      "step": 2507,
      "time_per_iteration": 3.2741971015930176
    },
    {
      "auxiliary_loss_clip": 0.01203114,
      "auxiliary_loss_mlp": 0.01026557,
      "balance_loss_clip": 1.05652261,
      "balance_loss_mlp": 1.01714587,
      "epoch": 0.30156916972284015,
      "flos": 22820549241600.0,
      "grad_norm": 2.895132829434318,
      "language_loss": 0.64425087,
      "learning_rate": 3.2751360570423767e-06,
      "loss": 0.66654766,
      "num_input_tokens_seen": 53908135,
      "step": 2508,
      "time_per_iteration": 2.6599743366241455
    },
    {
      "auxiliary_loss_clip": 0.01206201,
      "auxiliary_loss_mlp": 0.01031826,
      "balance_loss_clip": 1.05592716,
      "balance_loss_mlp": 1.02272415,
      "epoch": 0.3016894126134792,
      "flos": 29899188529920.0,
      "grad_norm": 3.648524921487948,
      "language_loss": 0.75792724,
      "learning_rate": 3.2745358452083236e-06,
      "loss": 0.78030747,
      "num_input_tokens_seen": 53931035,
      "step": 2509,
      "time_per_iteration": 2.7334470748901367
    },
    {
      "auxiliary_loss_clip": 0.01204579,
      "auxiliary_loss_mlp": 0.01029362,
      "balance_loss_clip": 1.05886245,
      "balance_loss_mlp": 1.02110028,
      "epoch": 0.3018096555041183,
      "flos": 21546331200000.0,
      "grad_norm": 1.4690122807218018,
      "language_loss": 0.82437235,
      "learning_rate": 3.2739354400281955e-06,
      "loss": 0.84671175,
      "num_input_tokens_seen": 53952255,
      "step": 2510,
      "time_per_iteration": 2.709650754928589
    },
    {
      "auxiliary_loss_clip": 0.01117453,
      "auxiliary_loss_mlp": 0.01200392,
      "balance_loss_clip": 1.02381349,
      "balance_loss_mlp": 1.00018871,
      "epoch": 0.3019298983947574,
      "flos": 59136294597120.0,
      "grad_norm": 0.8708424066033542,
      "language_loss": 0.63707352,
      "learning_rate": 3.2733348415930744e-06,
      "loss": 0.66025198,
      "num_input_tokens_seen": 54014125,
      "step": 2511,
      "time_per_iteration": 3.2700717449188232
    },
    {
      "auxiliary_loss_clip": 0.01203612,
      "auxiliary_loss_mlp": 0.01028326,
      "balance_loss_clip": 1.05372477,
      "balance_loss_mlp": 1.01910543,
      "epoch": 0.3020501412853965,
      "flos": 34423070941440.0,
      "grad_norm": 1.9409142129126593,
      "language_loss": 0.80643475,
      "learning_rate": 3.27273404999407e-06,
      "loss": 0.82875413,
      "num_input_tokens_seen": 54036345,
      "step": 2512,
      "time_per_iteration": 2.8539535999298096
    },
    {
      "auxiliary_loss_clip": 0.01114472,
      "auxiliary_loss_mlp": 0.01002953,
      "balance_loss_clip": 1.02699232,
      "balance_loss_mlp": 1.00159442,
      "epoch": 0.3021703841760356,
      "flos": 71008288128000.0,
      "grad_norm": 0.8050796244207308,
      "language_loss": 0.60486925,
      "learning_rate": 3.272133065322322e-06,
      "loss": 0.62604356,
      "num_input_tokens_seen": 54094615,
      "step": 2513,
      "time_per_iteration": 3.266441583633423
    },
    {
      "auxiliary_loss_clip": 0.01198312,
      "auxiliary_loss_mlp": 0.01031157,
      "balance_loss_clip": 1.05670667,
      "balance_loss_mlp": 1.02213299,
      "epoch": 0.3022906270666747,
      "flos": 21510528318720.0,
      "grad_norm": 2.1966172497752083,
      "language_loss": 0.79355884,
      "learning_rate": 3.271531887669e-06,
      "loss": 0.8158536,
      "num_input_tokens_seen": 54114675,
      "step": 2514,
      "time_per_iteration": 2.65201735496521
    },
    {
      "auxiliary_loss_clip": 0.01213098,
      "auxiliary_loss_mlp": 0.01029002,
      "balance_loss_clip": 1.05064893,
      "balance_loss_mlp": 1.01926303,
      "epoch": 0.30241086995731375,
      "flos": 31132001168640.0,
      "grad_norm": 19.11646412768085,
      "language_loss": 0.63579619,
      "learning_rate": 3.2709305171253015e-06,
      "loss": 0.65821725,
      "num_input_tokens_seen": 54134795,
      "step": 2515,
      "time_per_iteration": 2.7651000022888184
    },
    {
      "auxiliary_loss_clip": 0.01202109,
      "auxiliary_loss_mlp": 0.0103024,
      "balance_loss_clip": 1.05697811,
      "balance_loss_mlp": 1.02129972,
      "epoch": 0.30253111284795287,
      "flos": 23511542152320.0,
      "grad_norm": 2.5354857686254433,
      "language_loss": 0.779796,
      "learning_rate": 3.2703289537824536e-06,
      "loss": 0.80211943,
      "num_input_tokens_seen": 54154595,
      "step": 2516,
      "time_per_iteration": 2.8152225017547607
    },
    {
      "auxiliary_loss_clip": 0.01212367,
      "auxiliary_loss_mlp": 0.01036589,
      "balance_loss_clip": 1.0518012,
      "balance_loss_mlp": 1.02713025,
      "epoch": 0.302651355738592,
      "flos": 18725367651840.0,
      "grad_norm": 3.3162274752033527,
      "language_loss": 0.78386295,
      "learning_rate": 3.269727197731714e-06,
      "loss": 0.8063525,
      "num_input_tokens_seen": 54167360,
      "step": 2517,
      "time_per_iteration": 2.693233013153076
    },
    {
      "auxiliary_loss_clip": 0.01206685,
      "auxiliary_loss_mlp": 0.01033619,
      "balance_loss_clip": 1.05306113,
      "balance_loss_mlp": 1.02452922,
      "epoch": 0.30277159862923103,
      "flos": 22418888382720.0,
      "grad_norm": 2.8007022610247434,
      "language_loss": 0.77946174,
      "learning_rate": 3.269125249064367e-06,
      "loss": 0.80186474,
      "num_input_tokens_seen": 54187055,
      "step": 2518,
      "time_per_iteration": 2.660629987716675
    },
    {
      "auxiliary_loss_clip": 0.01204134,
      "auxiliary_loss_mlp": 0.01031624,
      "balance_loss_clip": 1.05944192,
      "balance_loss_mlp": 1.02256393,
      "epoch": 0.30289184151987014,
      "flos": 22273126992000.0,
      "grad_norm": 2.098283671070733,
      "language_loss": 0.83272552,
      "learning_rate": 3.2685231078717297e-06,
      "loss": 0.85508311,
      "num_input_tokens_seen": 54207245,
      "step": 2519,
      "time_per_iteration": 2.6318464279174805
    },
    {
      "auxiliary_loss_clip": 0.01200246,
      "auxiliary_loss_mlp": 0.01209357,
      "balance_loss_clip": 1.0543499,
      "balance_loss_mlp": 1.00090706,
      "epoch": 0.30301208441050925,
      "flos": 25225594231680.0,
      "grad_norm": 2.074756639723109,
      "language_loss": 0.75325525,
      "learning_rate": 3.267920774245145e-06,
      "loss": 0.77735126,
      "num_input_tokens_seen": 54226650,
      "step": 2520,
      "time_per_iteration": 2.7203197479248047
    },
    {
      "auxiliary_loss_clip": 0.01206547,
      "auxiliary_loss_mlp": 0.01034339,
      "balance_loss_clip": 1.05814123,
      "balance_loss_mlp": 1.02407527,
      "epoch": 0.3031323273011483,
      "flos": 23039245198080.0,
      "grad_norm": 1.7287986284176846,
      "language_loss": 0.84552938,
      "learning_rate": 3.2673182482759876e-06,
      "loss": 0.86793822,
      "num_input_tokens_seen": 54245765,
      "step": 2521,
      "time_per_iteration": 2.754467725753784
    },
    {
      "auxiliary_loss_clip": 0.01203935,
      "auxiliary_loss_mlp": 0.01031128,
      "balance_loss_clip": 1.05786085,
      "balance_loss_mlp": 1.02152562,
      "epoch": 0.3032525701917874,
      "flos": 18876695650560.0,
      "grad_norm": 2.2063860344606803,
      "language_loss": 0.66665983,
      "learning_rate": 3.266715530055659e-06,
      "loss": 0.6890105,
      "num_input_tokens_seen": 54263915,
      "step": 2522,
      "time_per_iteration": 2.7308311462402344
    },
    {
      "auxiliary_loss_clip": 0.01193456,
      "auxiliary_loss_mlp": 0.01030212,
      "balance_loss_clip": 1.0533042,
      "balance_loss_mlp": 1.02053773,
      "epoch": 0.30337281308242653,
      "flos": 17782641250560.0,
      "grad_norm": 2.327716617095613,
      "language_loss": 0.80449712,
      "learning_rate": 3.2661126196755927e-06,
      "loss": 0.82673377,
      "num_input_tokens_seen": 54283025,
      "step": 2523,
      "time_per_iteration": 3.6546425819396973
    },
    {
      "auxiliary_loss_clip": 0.01100612,
      "auxiliary_loss_mlp": 0.01000925,
      "balance_loss_clip": 1.02651978,
      "balance_loss_mlp": 0.99960822,
      "epoch": 0.3034930559730656,
      "flos": 57824298426240.0,
      "grad_norm": 0.7792986887952508,
      "language_loss": 0.5592376,
      "learning_rate": 3.265509517227248e-06,
      "loss": 0.58025295,
      "num_input_tokens_seen": 54339840,
      "step": 2524,
      "time_per_iteration": 3.2553610801696777
    },
    {
      "auxiliary_loss_clip": 0.01205862,
      "auxiliary_loss_mlp": 0.0102671,
      "balance_loss_clip": 1.05330014,
      "balance_loss_mlp": 1.01771533,
      "epoch": 0.3036132988637047,
      "flos": 14755587419520.0,
      "grad_norm": 1.893366555508359,
      "language_loss": 0.81017053,
      "learning_rate": 3.264906222802115e-06,
      "loss": 0.83249617,
      "num_input_tokens_seen": 54357690,
      "step": 2525,
      "time_per_iteration": 3.7012577056884766
    },
    {
      "auxiliary_loss_clip": 0.01206545,
      "auxiliary_loss_mlp": 0.01029071,
      "balance_loss_clip": 1.05987597,
      "balance_loss_mlp": 1.01884842,
      "epoch": 0.30373354175434375,
      "flos": 21033203460480.0,
      "grad_norm": 4.642104102042747,
      "language_loss": 0.77827108,
      "learning_rate": 3.264302736491715e-06,
      "loss": 0.80062723,
      "num_input_tokens_seen": 54377810,
      "step": 2526,
      "time_per_iteration": 2.5553109645843506
    },
    {
      "auxiliary_loss_clip": 0.01201226,
      "auxiliary_loss_mlp": 0.0102735,
      "balance_loss_clip": 1.05935204,
      "balance_loss_mlp": 1.01798046,
      "epoch": 0.30385378464498286,
      "flos": 21143233797120.0,
      "grad_norm": 2.158019963052893,
      "language_loss": 0.87615681,
      "learning_rate": 3.263699058387594e-06,
      "loss": 0.89844257,
      "num_input_tokens_seen": 54395245,
      "step": 2527,
      "time_per_iteration": 2.6613903045654297
    },
    {
      "auxiliary_loss_clip": 0.01203659,
      "auxiliary_loss_mlp": 0.01035093,
      "balance_loss_clip": 1.0498209,
      "balance_loss_mlp": 1.02528203,
      "epoch": 0.30397402753562197,
      "flos": 20629244131200.0,
      "grad_norm": 2.6128580308747313,
      "language_loss": 0.90479779,
      "learning_rate": 3.2630951885813315e-06,
      "loss": 0.92718536,
      "num_input_tokens_seen": 54412640,
      "step": 2528,
      "time_per_iteration": 2.8281803131103516
    },
    {
      "auxiliary_loss_clip": 0.01204656,
      "auxiliary_loss_mlp": 0.01030927,
      "balance_loss_clip": 1.0527389,
      "balance_loss_mlp": 1.02153325,
      "epoch": 0.304094270426261,
      "flos": 15085678429440.0,
      "grad_norm": 2.411735834668841,
      "language_loss": 0.77870357,
      "learning_rate": 3.262491127164533e-06,
      "loss": 0.80105937,
      "num_input_tokens_seen": 54431455,
      "step": 2529,
      "time_per_iteration": 3.63581919670105
    },
    {
      "auxiliary_loss_clip": 0.01212708,
      "auxiliary_loss_mlp": 0.01209906,
      "balance_loss_clip": 1.05631495,
      "balance_loss_mlp": 1.00094676,
      "epoch": 0.30421451331690014,
      "flos": 13845216193920.0,
      "grad_norm": 2.5585666512196994,
      "language_loss": 0.80125296,
      "learning_rate": 3.2618868742288337e-06,
      "loss": 0.82547903,
      "num_input_tokens_seen": 54448380,
      "step": 2530,
      "time_per_iteration": 2.617100477218628
    },
    {
      "auxiliary_loss_clip": 0.01203454,
      "auxiliary_loss_mlp": 0.01033843,
      "balance_loss_clip": 1.05784786,
      "balance_loss_mlp": 1.02458644,
      "epoch": 0.30433475620753925,
      "flos": 17384212615680.0,
      "grad_norm": 1.8710880892259658,
      "language_loss": 0.72662008,
      "learning_rate": 3.261282429865899e-06,
      "loss": 0.74899304,
      "num_input_tokens_seen": 54466385,
      "step": 2531,
      "time_per_iteration": 3.6419570446014404
    },
    {
      "auxiliary_loss_clip": 0.01213177,
      "auxiliary_loss_mlp": 0.01208827,
      "balance_loss_clip": 1.05823588,
      "balance_loss_mlp": 1.00093377,
      "epoch": 0.3044549990981783,
      "flos": 18916951818240.0,
      "grad_norm": 2.2743108500686215,
      "language_loss": 0.72535127,
      "learning_rate": 3.2606777941674225e-06,
      "loss": 0.74957132,
      "num_input_tokens_seen": 54485040,
      "step": 2532,
      "time_per_iteration": 2.642155408859253
    },
    {
      "auxiliary_loss_clip": 0.012035,
      "auxiliary_loss_mlp": 0.01032181,
      "balance_loss_clip": 1.05204213,
      "balance_loss_mlp": 1.02232265,
      "epoch": 0.3045752419888174,
      "flos": 21068431724160.0,
      "grad_norm": 2.126787695528822,
      "language_loss": 0.84839272,
      "learning_rate": 3.2600729672251276e-06,
      "loss": 0.87074959,
      "num_input_tokens_seen": 54502755,
      "step": 2533,
      "time_per_iteration": 2.8021132946014404
    },
    {
      "auxiliary_loss_clip": 0.01203657,
      "auxiliary_loss_mlp": 0.01209591,
      "balance_loss_clip": 1.06087637,
      "balance_loss_mlp": 1.00099313,
      "epoch": 0.3046954848794565,
      "flos": 29096405516160.0,
      "grad_norm": 2.4026127508230832,
      "language_loss": 0.65724897,
      "learning_rate": 3.259467949130765e-06,
      "loss": 0.6813814,
      "num_input_tokens_seen": 54524165,
      "step": 2534,
      "time_per_iteration": 2.7157862186431885
    },
    {
      "auxiliary_loss_clip": 0.01211323,
      "auxiliary_loss_mlp": 0.01026514,
      "balance_loss_clip": 1.0595057,
      "balance_loss_mlp": 1.01748943,
      "epoch": 0.3048157277700956,
      "flos": 20295346279680.0,
      "grad_norm": 2.2049490328321912,
      "language_loss": 0.82644784,
      "learning_rate": 3.2588627399761164e-06,
      "loss": 0.84882629,
      "num_input_tokens_seen": 54540160,
      "step": 2535,
      "time_per_iteration": 2.6153836250305176
    },
    {
      "auxiliary_loss_clip": 0.01206332,
      "auxiliary_loss_mlp": 0.01026439,
      "balance_loss_clip": 1.05634236,
      "balance_loss_mlp": 1.01788545,
      "epoch": 0.3049359706607347,
      "flos": 22739929165440.0,
      "grad_norm": 2.0614685306289364,
      "language_loss": 0.70959938,
      "learning_rate": 3.2582573398529903e-06,
      "loss": 0.73192704,
      "num_input_tokens_seen": 54557515,
      "step": 2536,
      "time_per_iteration": 2.669936180114746
    },
    {
      "auxiliary_loss_clip": 0.01208637,
      "auxiliary_loss_mlp": 0.01027737,
      "balance_loss_clip": 1.05300093,
      "balance_loss_mlp": 1.01740181,
      "epoch": 0.3050562135513738,
      "flos": 18434634969600.0,
      "grad_norm": 6.176898781067255,
      "language_loss": 0.73642766,
      "learning_rate": 3.2576517488532265e-06,
      "loss": 0.75879139,
      "num_input_tokens_seen": 54573865,
      "step": 2537,
      "time_per_iteration": 2.6261353492736816
    },
    {
      "auxiliary_loss_clip": 0.0120184,
      "auxiliary_loss_mlp": 0.01031809,
      "balance_loss_clip": 1.05461764,
      "balance_loss_mlp": 1.02330375,
      "epoch": 0.30517645644201286,
      "flos": 20370327920640.0,
      "grad_norm": 2.522856424859873,
      "language_loss": 0.8757754,
      "learning_rate": 3.257045967068692e-06,
      "loss": 0.89811188,
      "num_input_tokens_seen": 54593120,
      "step": 2538,
      "time_per_iteration": 2.6280198097229004
    },
    {
      "auxiliary_loss_clip": 0.01203664,
      "auxiliary_loss_mlp": 0.01033971,
      "balance_loss_clip": 1.0586648,
      "balance_loss_mlp": 1.02399898,
      "epoch": 0.30529669933265197,
      "flos": 21945118970880.0,
      "grad_norm": 1.5854673304431868,
      "language_loss": 0.82187808,
      "learning_rate": 3.2564399945912848e-06,
      "loss": 0.84425437,
      "num_input_tokens_seen": 54612910,
      "step": 2539,
      "time_per_iteration": 2.6445376873016357
    },
    {
      "auxiliary_loss_clip": 0.01214374,
      "auxiliary_loss_mlp": 0.010304,
      "balance_loss_clip": 1.05183935,
      "balance_loss_mlp": 1.02165008,
      "epoch": 0.305416942223291,
      "flos": 21835411856640.0,
      "grad_norm": 2.2858588299439657,
      "language_loss": 0.82051253,
      "learning_rate": 3.2558338315129287e-06,
      "loss": 0.84296036,
      "num_input_tokens_seen": 54631055,
      "step": 2540,
      "time_per_iteration": 2.8580069541931152
    },
    {
      "auxiliary_loss_clip": 0.01197591,
      "auxiliary_loss_mlp": 0.01028099,
      "balance_loss_clip": 1.05567288,
      "balance_loss_mlp": 1.01820481,
      "epoch": 0.30553718511393013,
      "flos": 33911810709120.0,
      "grad_norm": 2.287749965953964,
      "language_loss": 0.75906688,
      "learning_rate": 3.2552274779255785e-06,
      "loss": 0.78132379,
      "num_input_tokens_seen": 54651985,
      "step": 2541,
      "time_per_iteration": 2.822484016418457
    },
    {
      "auxiliary_loss_clip": 0.0120331,
      "auxiliary_loss_mlp": 0.0103399,
      "balance_loss_clip": 1.05756009,
      "balance_loss_mlp": 1.02445948,
      "epoch": 0.30565742800456924,
      "flos": 22268530051200.0,
      "grad_norm": 3.421561114645871,
      "language_loss": 0.76501262,
      "learning_rate": 3.2546209339212184e-06,
      "loss": 0.78738564,
      "num_input_tokens_seen": 54671005,
      "step": 2542,
      "time_per_iteration": 2.6553051471710205
    },
    {
      "auxiliary_loss_clip": 0.01207008,
      "auxiliary_loss_mlp": 0.01031917,
      "balance_loss_clip": 1.05394673,
      "balance_loss_mlp": 1.02199912,
      "epoch": 0.3057776708952083,
      "flos": 22565044823040.0,
      "grad_norm": 1.8910769933127805,
      "language_loss": 0.77797294,
      "learning_rate": 3.25401419959186e-06,
      "loss": 0.80036223,
      "num_input_tokens_seen": 54691615,
      "step": 2543,
      "time_per_iteration": 2.692765712738037
    },
    {
      "auxiliary_loss_clip": 0.01217838,
      "auxiliary_loss_mlp": 0.01034769,
      "balance_loss_clip": 1.0605942,
      "balance_loss_mlp": 1.02520251,
      "epoch": 0.3058979137858474,
      "flos": 21799213925760.0,
      "grad_norm": 2.513117193680059,
      "language_loss": 0.76352251,
      "learning_rate": 3.253407275029545e-06,
      "loss": 0.78604865,
      "num_input_tokens_seen": 54710520,
      "step": 2544,
      "time_per_iteration": 2.630985975265503
    },
    {
      "auxiliary_loss_clip": 0.01212479,
      "auxiliary_loss_mlp": 0.01031455,
      "balance_loss_clip": 1.05729973,
      "balance_loss_mlp": 1.02109551,
      "epoch": 0.3060181566764865,
      "flos": 26979435601920.0,
      "grad_norm": 1.8383817305145116,
      "language_loss": 0.80492544,
      "learning_rate": 3.2528001603263425e-06,
      "loss": 0.8273648,
      "num_input_tokens_seen": 54732590,
      "step": 2545,
      "time_per_iteration": 2.7911159992218018
    },
    {
      "auxiliary_loss_clip": 0.01205875,
      "auxiliary_loss_mlp": 0.01025249,
      "balance_loss_clip": 1.05990744,
      "balance_loss_mlp": 1.01593268,
      "epoch": 0.3061383995671256,
      "flos": 19865101173120.0,
      "grad_norm": 3.256880315892878,
      "language_loss": 0.81446064,
      "learning_rate": 3.2521928555743514e-06,
      "loss": 0.83677185,
      "num_input_tokens_seen": 54749935,
      "step": 2546,
      "time_per_iteration": 2.6262896060943604
    },
    {
      "auxiliary_loss_clip": 0.01198907,
      "auxiliary_loss_mlp": 0.01209413,
      "balance_loss_clip": 1.05183768,
      "balance_loss_mlp": 1.00080431,
      "epoch": 0.3062586424577647,
      "flos": 22127509255680.0,
      "grad_norm": 1.8043049761540746,
      "language_loss": 0.6743682,
      "learning_rate": 3.2515853608657e-06,
      "loss": 0.6984514,
      "num_input_tokens_seen": 54767935,
      "step": 2547,
      "time_per_iteration": 2.6626296043395996
    },
    {
      "auxiliary_loss_clip": 0.01201382,
      "auxiliary_loss_mlp": 0.01027402,
      "balance_loss_clip": 1.05722427,
      "balance_loss_mlp": 1.01791859,
      "epoch": 0.3063788853484038,
      "flos": 20845497962880.0,
      "grad_norm": 2.2552775175320834,
      "language_loss": 0.75375444,
      "learning_rate": 3.250977676292545e-06,
      "loss": 0.77604234,
      "num_input_tokens_seen": 54786175,
      "step": 2548,
      "time_per_iteration": 2.5987155437469482
    },
    {
      "auxiliary_loss_clip": 0.01211987,
      "auxiliary_loss_mlp": 0.01027885,
      "balance_loss_clip": 1.05703366,
      "balance_loss_mlp": 1.01858652,
      "epoch": 0.30649912823904285,
      "flos": 16209717707520.0,
      "grad_norm": 2.7257740173025566,
      "language_loss": 0.79195154,
      "learning_rate": 3.2503698019470712e-06,
      "loss": 0.81435025,
      "num_input_tokens_seen": 54801945,
      "step": 2549,
      "time_per_iteration": 2.6666176319122314
    },
    {
      "auxiliary_loss_clip": 0.01204449,
      "auxiliary_loss_mlp": 0.01035457,
      "balance_loss_clip": 1.05645537,
      "balance_loss_mlp": 1.02537155,
      "epoch": 0.30661937112968196,
      "flos": 18617815353600.0,
      "grad_norm": 4.329327443882448,
      "language_loss": 0.77940398,
      "learning_rate": 3.249761737921492e-06,
      "loss": 0.80180299,
      "num_input_tokens_seen": 54818475,
      "step": 2550,
      "time_per_iteration": 3.759666919708252
    },
    {
      "auxiliary_loss_clip": 0.01203665,
      "auxiliary_loss_mlp": 0.01032814,
      "balance_loss_clip": 1.05666876,
      "balance_loss_mlp": 1.02347398,
      "epoch": 0.30673961402032107,
      "flos": 31390809638400.0,
      "grad_norm": 1.9930078052156504,
      "language_loss": 0.74284768,
      "learning_rate": 3.249153484308051e-06,
      "loss": 0.76521254,
      "num_input_tokens_seen": 54837090,
      "step": 2551,
      "time_per_iteration": 3.7059836387634277
    },
    {
      "auxiliary_loss_clip": 0.01201727,
      "auxiliary_loss_mlp": 0.01031095,
      "balance_loss_clip": 1.05079317,
      "balance_loss_mlp": 1.02146244,
      "epoch": 0.3068598569109601,
      "flos": 20229809915520.0,
      "grad_norm": 2.425546005827505,
      "language_loss": 0.77782243,
      "learning_rate": 3.2485450411990194e-06,
      "loss": 0.80015063,
      "num_input_tokens_seen": 54856445,
      "step": 2552,
      "time_per_iteration": 2.7285709381103516
    },
    {
      "auxiliary_loss_clip": 0.01206787,
      "auxiliary_loss_mlp": 0.01031956,
      "balance_loss_clip": 1.05984974,
      "balance_loss_mlp": 1.02221012,
      "epoch": 0.30698009980159924,
      "flos": 29601991399680.0,
      "grad_norm": 3.550794232953364,
      "language_loss": 0.82320327,
      "learning_rate": 3.2479364086866983e-06,
      "loss": 0.84559071,
      "num_input_tokens_seen": 54876700,
      "step": 2553,
      "time_per_iteration": 2.639155149459839
    },
    {
      "auxiliary_loss_clip": 0.01211379,
      "auxiliary_loss_mlp": 0.01210167,
      "balance_loss_clip": 1.06037259,
      "balance_loss_mlp": 1.00091147,
      "epoch": 0.30710034269223835,
      "flos": 23842423261440.0,
      "grad_norm": 1.8573257477582055,
      "language_loss": 0.81541252,
      "learning_rate": 3.247327586863416e-06,
      "loss": 0.83962804,
      "num_input_tokens_seen": 54897580,
      "step": 2554,
      "time_per_iteration": 2.6972405910491943
    },
    {
      "auxiliary_loss_clip": 0.01217306,
      "auxiliary_loss_mlp": 0.01030198,
      "balance_loss_clip": 1.05675471,
      "balance_loss_mlp": 1.02024364,
      "epoch": 0.3072205855828774,
      "flos": 25884986152320.0,
      "grad_norm": 2.169222339297295,
      "language_loss": 0.77073956,
      "learning_rate": 3.2467185758215304e-06,
      "loss": 0.79321456,
      "num_input_tokens_seen": 54917320,
      "step": 2555,
      "time_per_iteration": 2.7466728687286377
    },
    {
      "auxiliary_loss_clip": 0.01213772,
      "auxiliary_loss_mlp": 0.01209992,
      "balance_loss_clip": 1.05773163,
      "balance_loss_mlp": 1.00093079,
      "epoch": 0.3073408284735165,
      "flos": 22236390357120.0,
      "grad_norm": 2.47988718632357,
      "language_loss": 0.85912883,
      "learning_rate": 3.246109375653428e-06,
      "loss": 0.88336647,
      "num_input_tokens_seen": 54934085,
      "step": 2556,
      "time_per_iteration": 3.7039854526519775
    },
    {
      "auxiliary_loss_clip": 0.01204044,
      "auxiliary_loss_mlp": 0.01033673,
      "balance_loss_clip": 1.06062472,
      "balance_loss_mlp": 1.02390361,
      "epoch": 0.30746107136415557,
      "flos": 19500284689920.0,
      "grad_norm": 1.9111740720370562,
      "language_loss": 0.78220332,
      "learning_rate": 3.2454999864515243e-06,
      "loss": 0.80458051,
      "num_input_tokens_seen": 54953460,
      "step": 2557,
      "time_per_iteration": 3.7194020748138428
    },
    {
      "auxiliary_loss_clip": 0.012045,
      "auxiliary_loss_mlp": 0.01210341,
      "balance_loss_clip": 1.05683124,
      "balance_loss_mlp": 1.00081515,
      "epoch": 0.3075813142547947,
      "flos": 21724806902400.0,
      "grad_norm": 2.8803492531879535,
      "language_loss": 0.69557649,
      "learning_rate": 3.244890408308263e-06,
      "loss": 0.71972489,
      "num_input_tokens_seen": 54974165,
      "step": 2558,
      "time_per_iteration": 2.719001531600952
    },
    {
      "auxiliary_loss_clip": 0.01213076,
      "auxiliary_loss_mlp": 0.01024647,
      "balance_loss_clip": 1.05301058,
      "balance_loss_mlp": 1.01529491,
      "epoch": 0.3077015571454338,
      "flos": 24097963593600.0,
      "grad_norm": 2.363227691896973,
      "language_loss": 0.60865712,
      "learning_rate": 3.2442806413161165e-06,
      "loss": 0.63103437,
      "num_input_tokens_seen": 54993810,
      "step": 2559,
      "time_per_iteration": 2.699474334716797
    },
    {
      "auxiliary_loss_clip": 0.01214566,
      "auxiliary_loss_mlp": 0.01029769,
      "balance_loss_clip": 1.05305338,
      "balance_loss_mlp": 1.01947522,
      "epoch": 0.30782180003607285,
      "flos": 18405476104320.0,
      "grad_norm": 2.127895075298037,
      "language_loss": 0.7552014,
      "learning_rate": 3.243670685567586e-06,
      "loss": 0.77764475,
      "num_input_tokens_seen": 55011210,
      "step": 2560,
      "time_per_iteration": 2.632827043533325
    },
    {
      "auxiliary_loss_clip": 0.01206284,
      "auxiliary_loss_mlp": 0.01209387,
      "balance_loss_clip": 1.05698025,
      "balance_loss_mlp": 1.00092888,
      "epoch": 0.30794204292671196,
      "flos": 23878549365120.0,
      "grad_norm": 2.736929654984928,
      "language_loss": 0.80408561,
      "learning_rate": 3.2430605411552012e-06,
      "loss": 0.8282423,
      "num_input_tokens_seen": 55031325,
      "step": 2561,
      "time_per_iteration": 2.624176025390625
    },
    {
      "auxiliary_loss_clip": 0.01124578,
      "auxiliary_loss_mlp": 0.01001849,
      "balance_loss_clip": 1.02792442,
      "balance_loss_mlp": 1.0005734,
      "epoch": 0.30806228581735107,
      "flos": 67927800816000.0,
      "grad_norm": 0.8850570832264298,
      "language_loss": 0.70577526,
      "learning_rate": 3.2424502081715205e-06,
      "loss": 0.72703946,
      "num_input_tokens_seen": 55094440,
      "step": 2562,
      "time_per_iteration": 3.232440710067749
    },
    {
      "auxiliary_loss_clip": 0.01211034,
      "auxiliary_loss_mlp": 0.01032253,
      "balance_loss_clip": 1.05871797,
      "balance_loss_mlp": 1.02244782,
      "epoch": 0.3081825287079901,
      "flos": 23843213360640.0,
      "grad_norm": 2.7506009562073435,
      "language_loss": 0.78171933,
      "learning_rate": 3.241839686709132e-06,
      "loss": 0.80415213,
      "num_input_tokens_seen": 55115375,
      "step": 2563,
      "time_per_iteration": 2.6604857444763184
    },
    {
      "auxiliary_loss_clip": 0.01202833,
      "auxiliary_loss_mlp": 0.01036616,
      "balance_loss_clip": 1.05519903,
      "balance_loss_mlp": 1.02682316,
      "epoch": 0.30830277159862923,
      "flos": 16209969102720.0,
      "grad_norm": 6.410479920507969,
      "language_loss": 0.82439733,
      "learning_rate": 3.2412289768606495e-06,
      "loss": 0.84679186,
      "num_input_tokens_seen": 55131945,
      "step": 2564,
      "time_per_iteration": 2.6023428440093994
    },
    {
      "auxiliary_loss_clip": 0.01209162,
      "auxiliary_loss_mlp": 0.01033553,
      "balance_loss_clip": 1.05973577,
      "balance_loss_mlp": 1.02424896,
      "epoch": 0.30842301448926834,
      "flos": 29349503723520.0,
      "grad_norm": 1.662017721068644,
      "language_loss": 0.82730669,
      "learning_rate": 3.240618078718718e-06,
      "loss": 0.84973383,
      "num_input_tokens_seen": 55153405,
      "step": 2565,
      "time_per_iteration": 2.711141586303711
    },
    {
      "auxiliary_loss_clip": 0.01207025,
      "auxiliary_loss_mlp": 0.01034205,
      "balance_loss_clip": 1.05275512,
      "balance_loss_mlp": 1.02441192,
      "epoch": 0.3085432573799074,
      "flos": 21945190798080.0,
      "grad_norm": 1.9382628103274337,
      "language_loss": 0.74139047,
      "learning_rate": 3.240006992376011e-06,
      "loss": 0.76380277,
      "num_input_tokens_seen": 55173030,
      "step": 2566,
      "time_per_iteration": 2.73457932472229
    },
    {
      "auxiliary_loss_clip": 0.01214335,
      "auxiliary_loss_mlp": 0.01035876,
      "balance_loss_clip": 1.05983138,
      "balance_loss_mlp": 1.0262146,
      "epoch": 0.3086635002705465,
      "flos": 22054718344320.0,
      "grad_norm": 2.3479625774938184,
      "language_loss": 0.76235962,
      "learning_rate": 3.2393957179252284e-06,
      "loss": 0.78486168,
      "num_input_tokens_seen": 55189565,
      "step": 2567,
      "time_per_iteration": 2.687265157699585
    },
    {
      "auxiliary_loss_clip": 0.01206777,
      "auxiliary_loss_mlp": 0.01029014,
      "balance_loss_clip": 1.0629971,
      "balance_loss_mlp": 1.02000737,
      "epoch": 0.3087837431611856,
      "flos": 32665925520000.0,
      "grad_norm": 2.7541042598102403,
      "language_loss": 0.80900466,
      "learning_rate": 3.2387842554591016e-06,
      "loss": 0.83136261,
      "num_input_tokens_seen": 55210380,
      "step": 2568,
      "time_per_iteration": 2.7254788875579834
    },
    {
      "auxiliary_loss_clip": 0.01204858,
      "auxiliary_loss_mlp": 0.0103632,
      "balance_loss_clip": 1.06165934,
      "balance_loss_mlp": 1.02663422,
      "epoch": 0.3089039860518247,
      "flos": 17599245384960.0,
      "grad_norm": 3.1896306352335597,
      "language_loss": 0.87350285,
      "learning_rate": 3.238172605070388e-06,
      "loss": 0.89591467,
      "num_input_tokens_seen": 55225795,
      "step": 2569,
      "time_per_iteration": 2.601527214050293
    },
    {
      "auxiliary_loss_clip": 0.01206493,
      "auxiliary_loss_mlp": 0.0121038,
      "balance_loss_clip": 1.05950093,
      "balance_loss_mlp": 1.00073242,
      "epoch": 0.3090242289424638,
      "flos": 14383839611520.0,
      "grad_norm": 2.349373751932223,
      "language_loss": 0.78677481,
      "learning_rate": 3.2375607668518745e-06,
      "loss": 0.81094348,
      "num_input_tokens_seen": 55238830,
      "step": 2570,
      "time_per_iteration": 2.6505086421966553
    },
    {
      "auxiliary_loss_clip": 0.01199724,
      "auxiliary_loss_mlp": 0.01031086,
      "balance_loss_clip": 1.05587137,
      "balance_loss_mlp": 1.02176392,
      "epoch": 0.30914447183310284,
      "flos": 16068625084800.0,
      "grad_norm": 2.192815308943118,
      "language_loss": 0.89846635,
      "learning_rate": 3.236948740896377e-06,
      "loss": 0.92077452,
      "num_input_tokens_seen": 55253630,
      "step": 2571,
      "time_per_iteration": 2.699666976928711
    },
    {
      "auxiliary_loss_clip": 0.01206095,
      "auxiliary_loss_mlp": 0.01031905,
      "balance_loss_clip": 1.05942297,
      "balance_loss_mlp": 1.02237391,
      "epoch": 0.30926471472374195,
      "flos": 32230221546240.0,
      "grad_norm": 9.596518960989227,
      "language_loss": 0.84335071,
      "learning_rate": 3.2363365272967384e-06,
      "loss": 0.8657307,
      "num_input_tokens_seen": 55276200,
      "step": 2572,
      "time_per_iteration": 2.6978604793548584
    },
    {
      "auxiliary_loss_clip": 0.01206221,
      "auxiliary_loss_mlp": 0.0103435,
      "balance_loss_clip": 1.06299686,
      "balance_loss_mlp": 1.02331161,
      "epoch": 0.30938495761438106,
      "flos": 20370722970240.0,
      "grad_norm": 1.9138165901596864,
      "language_loss": 0.81382942,
      "learning_rate": 3.235724126145832e-06,
      "loss": 0.83623517,
      "num_input_tokens_seen": 55292235,
      "step": 2573,
      "time_per_iteration": 2.6511058807373047
    },
    {
      "auxiliary_loss_clip": 0.01196261,
      "auxiliary_loss_mlp": 0.0103538,
      "balance_loss_clip": 1.05638671,
      "balance_loss_mlp": 1.02532434,
      "epoch": 0.3095052005050201,
      "flos": 24061155131520.0,
      "grad_norm": 1.4764433667906816,
      "language_loss": 0.77587563,
      "learning_rate": 3.235111537536558e-06,
      "loss": 0.79819202,
      "num_input_tokens_seen": 55313050,
      "step": 2574,
      "time_per_iteration": 2.7070682048797607
    },
    {
      "auxiliary_loss_clip": 0.01207196,
      "auxiliary_loss_mlp": 0.01025428,
      "balance_loss_clip": 1.05986667,
      "balance_loss_mlp": 1.01657653,
      "epoch": 0.30962544339565923,
      "flos": 23401547729280.0,
      "grad_norm": 1.8614772383819365,
      "language_loss": 0.82589656,
      "learning_rate": 3.2344987615618456e-06,
      "loss": 0.84822279,
      "num_input_tokens_seen": 55332885,
      "step": 2575,
      "time_per_iteration": 2.71440052986145
    },
    {
      "auxiliary_loss_clip": 0.01207555,
      "auxiliary_loss_mlp": 0.01033167,
      "balance_loss_clip": 1.05659938,
      "balance_loss_mlp": 1.02382147,
      "epoch": 0.30974568628629834,
      "flos": 33799984692480.0,
      "grad_norm": 1.6816040772546579,
      "language_loss": 0.78665781,
      "learning_rate": 3.2338857983146533e-06,
      "loss": 0.80906498,
      "num_input_tokens_seen": 55354385,
      "step": 2576,
      "time_per_iteration": 3.729282855987549
    },
    {
      "auxiliary_loss_clip": 0.01198877,
      "auxiliary_loss_mlp": 0.01027274,
      "balance_loss_clip": 1.05710912,
      "balance_loss_mlp": 1.01700401,
      "epoch": 0.3098659291769374,
      "flos": 20229594433920.0,
      "grad_norm": 2.19829777423965,
      "language_loss": 0.76320726,
      "learning_rate": 3.233272647887966e-06,
      "loss": 0.78546876,
      "num_input_tokens_seen": 55373275,
      "step": 2577,
      "time_per_iteration": 2.648834705352783
    },
    {
      "auxiliary_loss_clip": 0.01207943,
      "auxiliary_loss_mlp": 0.01033194,
      "balance_loss_clip": 1.06333745,
      "balance_loss_mlp": 1.02396083,
      "epoch": 0.3099861720675765,
      "flos": 24748556682240.0,
      "grad_norm": 3.6793646807272276,
      "language_loss": 0.90204263,
      "learning_rate": 3.2326593103747985e-06,
      "loss": 0.92445397,
      "num_input_tokens_seen": 55392290,
      "step": 2578,
      "time_per_iteration": 3.540999412536621
    },
    {
      "auxiliary_loss_clip": 0.01207609,
      "auxiliary_loss_mlp": 0.01034071,
      "balance_loss_clip": 1.06293368,
      "balance_loss_mlp": 1.02459407,
      "epoch": 0.3101064149582156,
      "flos": 11765485704960.0,
      "grad_norm": 24.74432956581761,
      "language_loss": 0.850173,
      "learning_rate": 3.2320457858681936e-06,
      "loss": 0.87258977,
      "num_input_tokens_seen": 55410680,
      "step": 2579,
      "time_per_iteration": 2.62394380569458
    },
    {
      "auxiliary_loss_clip": 0.01210193,
      "auxiliary_loss_mlp": 0.01028705,
      "balance_loss_clip": 1.05756307,
      "balance_loss_mlp": 1.01925755,
      "epoch": 0.31022665784885467,
      "flos": 23033247626880.0,
      "grad_norm": 2.5214677001194383,
      "language_loss": 0.86133039,
      "learning_rate": 3.2314320744612228e-06,
      "loss": 0.88371933,
      "num_input_tokens_seen": 55425980,
      "step": 2580,
      "time_per_iteration": 2.712684392929077
    },
    {
      "auxiliary_loss_clip": 0.01203992,
      "auxiliary_loss_mlp": 0.01030199,
      "balance_loss_clip": 1.05887997,
      "balance_loss_mlp": 1.02101409,
      "epoch": 0.3103469007394938,
      "flos": 16289188548480.0,
      "grad_norm": 1.957609968879964,
      "language_loss": 0.7648201,
      "learning_rate": 3.2308181762469854e-06,
      "loss": 0.78716201,
      "num_input_tokens_seen": 55443925,
      "step": 2581,
      "time_per_iteration": 2.6016433238983154
    },
    {
      "auxiliary_loss_clip": 0.01209238,
      "auxiliary_loss_mlp": 0.01038186,
      "balance_loss_clip": 1.06225979,
      "balance_loss_mlp": 1.02852952,
      "epoch": 0.3104671436301329,
      "flos": 30515271626880.0,
      "grad_norm": 2.3939175469495106,
      "language_loss": 0.78453517,
      "learning_rate": 3.230204091318609e-06,
      "loss": 0.80700946,
      "num_input_tokens_seen": 55464465,
      "step": 2582,
      "time_per_iteration": 3.609772205352783
    },
    {
      "auxiliary_loss_clip": 0.01203554,
      "auxiliary_loss_mlp": 0.01209186,
      "balance_loss_clip": 1.06056798,
      "balance_loss_mlp": 1.00082052,
      "epoch": 0.31058738652077195,
      "flos": 20047240062720.0,
      "grad_norm": 5.629298616919303,
      "language_loss": 0.84727454,
      "learning_rate": 3.2295898197692503e-06,
      "loss": 0.87140191,
      "num_input_tokens_seen": 55483425,
      "step": 2583,
      "time_per_iteration": 2.639660358428955
    },
    {
      "auxiliary_loss_clip": 0.01206963,
      "auxiliary_loss_mlp": 0.01032588,
      "balance_loss_clip": 1.06305957,
      "balance_loss_mlp": 1.02371895,
      "epoch": 0.31070762941141106,
      "flos": 28074639237120.0,
      "grad_norm": 1.9352657237500397,
      "language_loss": 0.79679304,
      "learning_rate": 3.228975361692094e-06,
      "loss": 0.81918859,
      "num_input_tokens_seen": 55504445,
      "step": 2584,
      "time_per_iteration": 3.5433733463287354
    },
    {
      "auxiliary_loss_clip": 0.01213016,
      "auxiliary_loss_mlp": 0.01209994,
      "balance_loss_clip": 1.06031299,
      "balance_loss_mlp": 1.0009979,
      "epoch": 0.31082787230205017,
      "flos": 20521907314560.0,
      "grad_norm": 2.4494128437942218,
      "language_loss": 0.80498743,
      "learning_rate": 3.228360717180352e-06,
      "loss": 0.82921755,
      "num_input_tokens_seen": 55521970,
      "step": 2585,
      "time_per_iteration": 2.6747355461120605
    },
    {
      "auxiliary_loss_clip": 0.01107838,
      "auxiliary_loss_mlp": 0.01200665,
      "balance_loss_clip": 1.03248978,
      "balance_loss_mlp": 1.00011432,
      "epoch": 0.3109481151926892,
      "flos": 62445928723200.0,
      "grad_norm": 0.811171301266563,
      "language_loss": 0.59290707,
      "learning_rate": 3.227745886327266e-06,
      "loss": 0.61599207,
      "num_input_tokens_seen": 55580665,
      "step": 2586,
      "time_per_iteration": 3.1212284564971924
    },
    {
      "auxiliary_loss_clip": 0.01107414,
      "auxiliary_loss_mlp": 0.01006064,
      "balance_loss_clip": 1.03228855,
      "balance_loss_mlp": 1.00493133,
      "epoch": 0.31106835808332833,
      "flos": 44746744723200.0,
      "grad_norm": 0.8127397602277358,
      "language_loss": 0.55820042,
      "learning_rate": 3.227130869226105e-06,
      "loss": 0.57933521,
      "num_input_tokens_seen": 55637825,
      "step": 2587,
      "time_per_iteration": 3.1497654914855957
    },
    {
      "auxiliary_loss_clip": 0.01205474,
      "auxiliary_loss_mlp": 0.01022968,
      "balance_loss_clip": 1.05829406,
      "balance_loss_mlp": 1.01425326,
      "epoch": 0.3111886009739674,
      "flos": 23403056100480.0,
      "grad_norm": 2.8247641709538858,
      "language_loss": 0.83087045,
      "learning_rate": 3.226515665970167e-06,
      "loss": 0.8531549,
      "num_input_tokens_seen": 55655365,
      "step": 2588,
      "time_per_iteration": 2.7264010906219482
    },
    {
      "auxiliary_loss_clip": 0.01207152,
      "auxiliary_loss_mlp": 0.01028776,
      "balance_loss_clip": 1.0611937,
      "balance_loss_mlp": 1.01872659,
      "epoch": 0.3113088438646065,
      "flos": 17530728192000.0,
      "grad_norm": 3.163820069842062,
      "language_loss": 0.86575878,
      "learning_rate": 3.225900276652777e-06,
      "loss": 0.88811803,
      "num_input_tokens_seen": 55672140,
      "step": 2589,
      "time_per_iteration": 2.567561626434326
    },
    {
      "auxiliary_loss_clip": 0.01213712,
      "auxiliary_loss_mlp": 0.01032428,
      "balance_loss_clip": 1.05801272,
      "balance_loss_mlp": 1.02335024,
      "epoch": 0.3114290867552456,
      "flos": 28365802882560.0,
      "grad_norm": 1.6733158924077758,
      "language_loss": 0.75677741,
      "learning_rate": 3.2252847013672906e-06,
      "loss": 0.77923882,
      "num_input_tokens_seen": 55694800,
      "step": 2590,
      "time_per_iteration": 2.7599313259124756
    },
    {
      "auxiliary_loss_clip": 0.01200609,
      "auxiliary_loss_mlp": 0.01025734,
      "balance_loss_clip": 1.05266428,
      "balance_loss_mlp": 1.01653087,
      "epoch": 0.31154932964588467,
      "flos": 27379157126400.0,
      "grad_norm": 3.415176241256042,
      "language_loss": 0.75857186,
      "learning_rate": 3.224668940207089e-06,
      "loss": 0.78083539,
      "num_input_tokens_seen": 55713785,
      "step": 2591,
      "time_per_iteration": 2.6995842456817627
    },
    {
      "auxiliary_loss_clip": 0.01202797,
      "auxiliary_loss_mlp": 0.01037796,
      "balance_loss_clip": 1.04966402,
      "balance_loss_mlp": 1.02810478,
      "epoch": 0.3116695725365238,
      "flos": 26541864120960.0,
      "grad_norm": 2.179579653840301,
      "language_loss": 0.86994016,
      "learning_rate": 3.2240529932655828e-06,
      "loss": 0.89234602,
      "num_input_tokens_seen": 55733050,
      "step": 2592,
      "time_per_iteration": 2.749434232711792
    },
    {
      "auxiliary_loss_clip": 0.0120694,
      "auxiliary_loss_mlp": 0.01033329,
      "balance_loss_clip": 1.05914474,
      "balance_loss_mlp": 1.02420378,
      "epoch": 0.3117898154271629,
      "flos": 21177600134400.0,
      "grad_norm": 2.626301474915312,
      "language_loss": 0.88900566,
      "learning_rate": 3.223436860636211e-06,
      "loss": 0.91140836,
      "num_input_tokens_seen": 55748685,
      "step": 2593,
      "time_per_iteration": 2.631293296813965
    },
    {
      "auxiliary_loss_clip": 0.01205434,
      "auxiliary_loss_mlp": 0.01036389,
      "balance_loss_clip": 1.06192183,
      "balance_loss_mlp": 1.02710903,
      "epoch": 0.31191005831780194,
      "flos": 27272430840960.0,
      "grad_norm": 1.7256618254230849,
      "language_loss": 0.74140126,
      "learning_rate": 3.2228205424124403e-06,
      "loss": 0.76381952,
      "num_input_tokens_seen": 55771840,
      "step": 2594,
      "time_per_iteration": 2.6800386905670166
    },
    {
      "auxiliary_loss_clip": 0.01196924,
      "auxiliary_loss_mlp": 0.01024234,
      "balance_loss_clip": 1.05620813,
      "balance_loss_mlp": 1.01511407,
      "epoch": 0.31203030120844105,
      "flos": 12963501043200.0,
      "grad_norm": 2.398033896943132,
      "language_loss": 0.74618948,
      "learning_rate": 3.222204038687765e-06,
      "loss": 0.76840103,
      "num_input_tokens_seen": 55784975,
      "step": 2595,
      "time_per_iteration": 2.6385035514831543
    },
    {
      "auxiliary_loss_clip": 0.01202753,
      "auxiliary_loss_mlp": 0.01028826,
      "balance_loss_clip": 1.05818987,
      "balance_loss_mlp": 1.02005208,
      "epoch": 0.31215054409908016,
      "flos": 27562014288000.0,
      "grad_norm": 1.832873840165171,
      "language_loss": 0.88114864,
      "learning_rate": 3.221587349555709e-06,
      "loss": 0.90346444,
      "num_input_tokens_seen": 55805235,
      "step": 2596,
      "time_per_iteration": 2.705897331237793
    },
    {
      "auxiliary_loss_clip": 0.01209187,
      "auxiliary_loss_mlp": 0.01024408,
      "balance_loss_clip": 1.05645406,
      "balance_loss_mlp": 1.0150857,
      "epoch": 0.3122707869897192,
      "flos": 21506326427520.0,
      "grad_norm": 2.0080910208573926,
      "language_loss": 0.69618094,
      "learning_rate": 3.2209704751098236e-06,
      "loss": 0.71851695,
      "num_input_tokens_seen": 55824265,
      "step": 2597,
      "time_per_iteration": 2.705024242401123
    },
    {
      "auxiliary_loss_clip": 0.01209361,
      "auxiliary_loss_mlp": 0.01029899,
      "balance_loss_clip": 1.05844903,
      "balance_loss_mlp": 1.02060032,
      "epoch": 0.31239102988035833,
      "flos": 15187017674880.0,
      "grad_norm": 2.0590224978265104,
      "language_loss": 0.82367122,
      "learning_rate": 3.2203534154436875e-06,
      "loss": 0.84606385,
      "num_input_tokens_seen": 55838620,
      "step": 2598,
      "time_per_iteration": 2.653796434402466
    },
    {
      "auxiliary_loss_clip": 0.01210212,
      "auxiliary_loss_mlp": 0.01036322,
      "balance_loss_clip": 1.05253863,
      "balance_loss_mlp": 1.02722597,
      "epoch": 0.31251127277099744,
      "flos": 22053712763520.0,
      "grad_norm": 1.8751376178909351,
      "language_loss": 0.75432956,
      "learning_rate": 3.2197361706509084e-06,
      "loss": 0.77679491,
      "num_input_tokens_seen": 55859375,
      "step": 2599,
      "time_per_iteration": 2.750624895095825
    },
    {
      "auxiliary_loss_clip": 0.0120577,
      "auxiliary_loss_mlp": 0.01031219,
      "balance_loss_clip": 1.06036568,
      "balance_loss_mlp": 1.02094364,
      "epoch": 0.3126315156616365,
      "flos": 15193984913280.0,
      "grad_norm": 3.2357524774043043,
      "language_loss": 0.83644032,
      "learning_rate": 3.2191187408251228e-06,
      "loss": 0.85881025,
      "num_input_tokens_seen": 55876535,
      "step": 2600,
      "time_per_iteration": 2.703653573989868
    },
    {
      "auxiliary_loss_clip": 0.01213139,
      "auxiliary_loss_mlp": 0.01036568,
      "balance_loss_clip": 1.06029177,
      "balance_loss_mlp": 1.02633405,
      "epoch": 0.3127517585522756,
      "flos": 18145338831360.0,
      "grad_norm": 3.5226898142105187,
      "language_loss": 0.78869003,
      "learning_rate": 3.218501126059993e-06,
      "loss": 0.81118715,
      "num_input_tokens_seen": 55891930,
      "step": 2601,
      "time_per_iteration": 2.845562219619751
    },
    {
      "auxiliary_loss_clip": 0.01207119,
      "auxiliary_loss_mlp": 0.01028946,
      "balance_loss_clip": 1.05610883,
      "balance_loss_mlp": 1.01930237,
      "epoch": 0.31287200144291466,
      "flos": 21908633731200.0,
      "grad_norm": 2.4555282419133526,
      "language_loss": 0.81368446,
      "learning_rate": 3.2178833264492116e-06,
      "loss": 0.83604515,
      "num_input_tokens_seen": 55910635,
      "step": 2602,
      "time_per_iteration": 2.6507275104522705
    },
    {
      "auxiliary_loss_clip": 0.01213334,
      "auxiliary_loss_mlp": 0.01028261,
      "balance_loss_clip": 1.06054103,
      "balance_loss_mlp": 1.01936817,
      "epoch": 0.31299224433355377,
      "flos": 29896997800320.0,
      "grad_norm": 1.8274922797176598,
      "language_loss": 0.76026952,
      "learning_rate": 3.217265342086498e-06,
      "loss": 0.78268552,
      "num_input_tokens_seen": 55931125,
      "step": 2603,
      "time_per_iteration": 3.4938557147979736
    },
    {
      "auxiliary_loss_clip": 0.01211308,
      "auxiliary_loss_mlp": 0.01210528,
      "balance_loss_clip": 1.05697513,
      "balance_loss_mlp": 1.00103879,
      "epoch": 0.3131124872241929,
      "flos": 11655886331520.0,
      "grad_norm": 2.223595828745771,
      "language_loss": 0.72752213,
      "learning_rate": 3.216647173065599e-06,
      "loss": 0.75174046,
      "num_input_tokens_seen": 55946590,
      "step": 2604,
      "time_per_iteration": 3.3339970111846924
    },
    {
      "auxiliary_loss_clip": 0.01206854,
      "auxiliary_loss_mlp": 0.01033267,
      "balance_loss_clip": 1.06029058,
      "balance_loss_mlp": 1.02353966,
      "epoch": 0.31323273011483194,
      "flos": 49848785470080.0,
      "grad_norm": 1.9825622538381749,
      "language_loss": 0.7362625,
      "learning_rate": 3.216028819480292e-06,
      "loss": 0.75866371,
      "num_input_tokens_seen": 55967930,
      "step": 2605,
      "time_per_iteration": 2.7971999645233154
    },
    {
      "auxiliary_loss_clip": 0.01193527,
      "auxiliary_loss_mlp": 0.01031679,
      "balance_loss_clip": 1.05458069,
      "balance_loss_mlp": 1.02229726,
      "epoch": 0.31335297300547105,
      "flos": 22601278667520.0,
      "grad_norm": 3.5791662435204445,
      "language_loss": 0.75455725,
      "learning_rate": 3.2154102814243793e-06,
      "loss": 0.77680928,
      "num_input_tokens_seen": 55987070,
      "step": 2606,
      "time_per_iteration": 2.5458943843841553
    },
    {
      "auxiliary_loss_clip": 0.01212686,
      "auxiliary_loss_mlp": 0.01033105,
      "balance_loss_clip": 1.0562551,
      "balance_loss_mlp": 1.02390254,
      "epoch": 0.31347321589611016,
      "flos": 34710858708480.0,
      "grad_norm": 3.3692511441658444,
      "language_loss": 0.67035091,
      "learning_rate": 3.2147915589916937e-06,
      "loss": 0.69280887,
      "num_input_tokens_seen": 56008630,
      "step": 2607,
      "time_per_iteration": 2.6668100357055664
    },
    {
      "auxiliary_loss_clip": 0.01198167,
      "auxiliary_loss_mlp": 0.01027459,
      "balance_loss_clip": 1.05217648,
      "balance_loss_mlp": 1.01829791,
      "epoch": 0.3135934587867492,
      "flos": 19755789108480.0,
      "grad_norm": 2.8792757657934573,
      "language_loss": 0.82736546,
      "learning_rate": 3.2141726522760938e-06,
      "loss": 0.84962171,
      "num_input_tokens_seen": 56026690,
      "step": 2608,
      "time_per_iteration": 2.6767075061798096
    },
    {
      "auxiliary_loss_clip": 0.0110785,
      "auxiliary_loss_mlp": 0.01003491,
      "balance_loss_clip": 1.02961755,
      "balance_loss_mlp": 1.00215554,
      "epoch": 0.3137137016773883,
      "flos": 65815535583360.0,
      "grad_norm": 0.7056359235036728,
      "language_loss": 0.52627808,
      "learning_rate": 3.213553561371469e-06,
      "loss": 0.54739153,
      "num_input_tokens_seen": 56090425,
      "step": 2609,
      "time_per_iteration": 4.193963527679443
    },
    {
      "auxiliary_loss_clip": 0.0120744,
      "auxiliary_loss_mlp": 0.01031395,
      "balance_loss_clip": 1.05463696,
      "balance_loss_mlp": 1.02278233,
      "epoch": 0.31383394456802743,
      "flos": 16252739222400.0,
      "grad_norm": 2.637357144585044,
      "language_loss": 0.96004212,
      "learning_rate": 3.212934286371733e-06,
      "loss": 0.9824304,
      "num_input_tokens_seen": 56107135,
      "step": 2610,
      "time_per_iteration": 2.673328161239624
    },
    {
      "auxiliary_loss_clip": 0.01204734,
      "auxiliary_loss_mlp": 0.01029224,
      "balance_loss_clip": 1.06089318,
      "balance_loss_mlp": 1.01949012,
      "epoch": 0.3139541874586665,
      "flos": 38795517613440.0,
      "grad_norm": 3.1515221659375414,
      "language_loss": 0.83160174,
      "learning_rate": 3.2123148273708304e-06,
      "loss": 0.85394132,
      "num_input_tokens_seen": 56127325,
      "step": 2611,
      "time_per_iteration": 3.729713201522827
    },
    {
      "auxiliary_loss_clip": 0.01201282,
      "auxiliary_loss_mlp": 0.01027901,
      "balance_loss_clip": 1.05999386,
      "balance_loss_mlp": 1.01803648,
      "epoch": 0.3140744303493056,
      "flos": 25046328430080.0,
      "grad_norm": 1.926602090651372,
      "language_loss": 0.76841128,
      "learning_rate": 3.211695184462733e-06,
      "loss": 0.79070312,
      "num_input_tokens_seen": 56148500,
      "step": 2612,
      "time_per_iteration": 2.6878631114959717
    },
    {
      "auxiliary_loss_clip": 0.01118872,
      "auxiliary_loss_mlp": 0.01002008,
      "balance_loss_clip": 1.02757955,
      "balance_loss_mlp": 1.00075054,
      "epoch": 0.3141946732399447,
      "flos": 72504254782080.0,
      "grad_norm": 0.929373515155214,
      "language_loss": 0.60522515,
      "learning_rate": 3.2110753577414383e-06,
      "loss": 0.62643397,
      "num_input_tokens_seen": 56210080,
      "step": 2613,
      "time_per_iteration": 3.2056777477264404
    },
    {
      "auxiliary_loss_clip": 0.01208102,
      "auxiliary_loss_mlp": 0.01025998,
      "balance_loss_clip": 1.05456781,
      "balance_loss_mlp": 1.01662266,
      "epoch": 0.31431491613058377,
      "flos": 19239788280960.0,
      "grad_norm": 1.9886313974511705,
      "language_loss": 0.78690886,
      "learning_rate": 3.2104553473009757e-06,
      "loss": 0.80924988,
      "num_input_tokens_seen": 56228200,
      "step": 2614,
      "time_per_iteration": 2.6734561920166016
    },
    {
      "auxiliary_loss_clip": 0.01203558,
      "auxiliary_loss_mlp": 0.0103011,
      "balance_loss_clip": 1.05131364,
      "balance_loss_mlp": 1.02072263,
      "epoch": 0.3144351590212229,
      "flos": 36210596290560.0,
      "grad_norm": 3.5753598246250116,
      "language_loss": 0.68116927,
      "learning_rate": 3.209835153235399e-06,
      "loss": 0.70350593,
      "num_input_tokens_seen": 56249755,
      "step": 2615,
      "time_per_iteration": 2.828033924102783
    },
    {
      "auxiliary_loss_clip": 0.01196758,
      "auxiliary_loss_mlp": 0.01025498,
      "balance_loss_clip": 1.05130816,
      "balance_loss_mlp": 1.01648593,
      "epoch": 0.314555401911862,
      "flos": 18551740285440.0,
      "grad_norm": 2.503267338131717,
      "language_loss": 0.67560226,
      "learning_rate": 3.2092147756387916e-06,
      "loss": 0.69782484,
      "num_input_tokens_seen": 56270080,
      "step": 2616,
      "time_per_iteration": 2.7311229705810547
    },
    {
      "auxiliary_loss_clip": 0.0119842,
      "auxiliary_loss_mlp": 0.01031853,
      "balance_loss_clip": 1.05262339,
      "balance_loss_mlp": 1.02185106,
      "epoch": 0.31467564480250104,
      "flos": 16362877299840.0,
      "grad_norm": 2.492300781312991,
      "language_loss": 0.83838958,
      "learning_rate": 3.208594214605264e-06,
      "loss": 0.86069226,
      "num_input_tokens_seen": 56288625,
      "step": 2617,
      "time_per_iteration": 2.6044623851776123
    },
    {
      "auxiliary_loss_clip": 0.01194526,
      "auxiliary_loss_mlp": 0.01027902,
      "balance_loss_clip": 1.05374002,
      "balance_loss_mlp": 1.01922369,
      "epoch": 0.31479588769314015,
      "flos": 21652375127040.0,
      "grad_norm": 4.261186945998743,
      "language_loss": 0.77327502,
      "learning_rate": 3.2079734702289553e-06,
      "loss": 0.79549932,
      "num_input_tokens_seen": 56307520,
      "step": 2618,
      "time_per_iteration": 2.678741931915283
    },
    {
      "auxiliary_loss_clip": 0.01101346,
      "auxiliary_loss_mlp": 0.01200349,
      "balance_loss_clip": 1.02615833,
      "balance_loss_mlp": 1.0002408,
      "epoch": 0.3149161305837792,
      "flos": 66051072040320.0,
      "grad_norm": 0.8065074821554253,
      "language_loss": 0.60408509,
      "learning_rate": 3.207352542604031e-06,
      "loss": 0.62710202,
      "num_input_tokens_seen": 56369855,
      "step": 2619,
      "time_per_iteration": 3.2701220512390137
    },
    {
      "auxiliary_loss_clip": 0.01192956,
      "auxiliary_loss_mlp": 0.01031109,
      "balance_loss_clip": 1.05074191,
      "balance_loss_mlp": 1.02248418,
      "epoch": 0.3150363734744183,
      "flos": 28987201192320.0,
      "grad_norm": 1.7270371719212991,
      "language_loss": 0.78512174,
      "learning_rate": 3.2067314318246864e-06,
      "loss": 0.80736238,
      "num_input_tokens_seen": 56390570,
      "step": 2620,
      "time_per_iteration": 2.758711338043213
    },
    {
      "auxiliary_loss_clip": 0.01209089,
      "auxiliary_loss_mlp": 0.01025827,
      "balance_loss_clip": 1.0556699,
      "balance_loss_mlp": 1.01694036,
      "epoch": 0.31515661636505743,
      "flos": 27636600879360.0,
      "grad_norm": 2.703724398497031,
      "language_loss": 0.77816844,
      "learning_rate": 3.206110137985143e-06,
      "loss": 0.80051768,
      "num_input_tokens_seen": 56410775,
      "step": 2621,
      "time_per_iteration": 2.742928981781006
    },
    {
      "auxiliary_loss_clip": 0.01193817,
      "auxiliary_loss_mlp": 0.01028028,
      "balance_loss_clip": 1.04974353,
      "balance_loss_mlp": 1.01826429,
      "epoch": 0.3152768592556965,
      "flos": 24605632465920.0,
      "grad_norm": 2.0308546958453797,
      "language_loss": 0.92035007,
      "learning_rate": 3.2054886611796505e-06,
      "loss": 0.94256854,
      "num_input_tokens_seen": 56429770,
      "step": 2622,
      "time_per_iteration": 2.7515664100646973
    },
    {
      "auxiliary_loss_clip": 0.01100217,
      "auxiliary_loss_mlp": 0.01001059,
      "balance_loss_clip": 1.02754283,
      "balance_loss_mlp": 0.99984264,
      "epoch": 0.3153971021463356,
      "flos": 68476908026880.0,
      "grad_norm": 0.8875489972358996,
      "language_loss": 0.63554287,
      "learning_rate": 3.204867001502487e-06,
      "loss": 0.65655565,
      "num_input_tokens_seen": 56488425,
      "step": 2623,
      "time_per_iteration": 3.153733015060425
    },
    {
      "auxiliary_loss_clip": 0.01200823,
      "auxiliary_loss_mlp": 0.01031763,
      "balance_loss_clip": 1.05887127,
      "balance_loss_mlp": 1.02178502,
      "epoch": 0.3155173450369747,
      "flos": 25593714766080.0,
      "grad_norm": 1.800270166647222,
      "language_loss": 0.80938268,
      "learning_rate": 3.2042451590479567e-06,
      "loss": 0.83170855,
      "num_input_tokens_seen": 56508940,
      "step": 2624,
      "time_per_iteration": 2.6551249027252197
    },
    {
      "auxiliary_loss_clip": 0.01195307,
      "auxiliary_loss_mlp": 0.01031705,
      "balance_loss_clip": 1.05763638,
      "balance_loss_mlp": 1.02254415,
      "epoch": 0.31563758792761376,
      "flos": 24309333175680.0,
      "grad_norm": 2.591301653314764,
      "language_loss": 0.86809492,
      "learning_rate": 3.203623133910394e-06,
      "loss": 0.89036506,
      "num_input_tokens_seen": 56527245,
      "step": 2625,
      "time_per_iteration": 2.6518337726593018
    },
    {
      "auxiliary_loss_clip": 0.01212629,
      "auxiliary_loss_mlp": 0.01036726,
      "balance_loss_clip": 1.05084705,
      "balance_loss_mlp": 1.02766562,
      "epoch": 0.31575783081825287,
      "flos": 31903865550720.0,
      "grad_norm": 2.2248122512579873,
      "language_loss": 0.77334744,
      "learning_rate": 3.203000926184158e-06,
      "loss": 0.79584098,
      "num_input_tokens_seen": 56546170,
      "step": 2626,
      "time_per_iteration": 2.8185205459594727
    },
    {
      "auxiliary_loss_clip": 0.01199153,
      "auxiliary_loss_mlp": 0.01031356,
      "balance_loss_clip": 1.05873013,
      "balance_loss_mlp": 1.0229876,
      "epoch": 0.315878073708892,
      "flos": 30810960385920.0,
      "grad_norm": 1.6978976051635213,
      "language_loss": 0.77524585,
      "learning_rate": 3.202378535963639e-06,
      "loss": 0.79755098,
      "num_input_tokens_seen": 56567085,
      "step": 2627,
      "time_per_iteration": 2.884357452392578
    },
    {
      "auxiliary_loss_clip": 0.01193226,
      "auxiliary_loss_mlp": 0.01210549,
      "balance_loss_clip": 1.05128658,
      "balance_loss_mlp": 1.00108314,
      "epoch": 0.31599831659953104,
      "flos": 22200264253440.0,
      "grad_norm": 1.652546378148362,
      "language_loss": 0.83759487,
      "learning_rate": 3.2017559633432516e-06,
      "loss": 0.86163259,
      "num_input_tokens_seen": 56586715,
      "step": 2628,
      "time_per_iteration": 2.6357362270355225
    },
    {
      "auxiliary_loss_clip": 0.01210941,
      "auxiliary_loss_mlp": 0.01032138,
      "balance_loss_clip": 1.0545187,
      "balance_loss_mlp": 1.02314949,
      "epoch": 0.31611855949017015,
      "flos": 25593463370880.0,
      "grad_norm": 1.8328107605933237,
      "language_loss": 0.66458195,
      "learning_rate": 3.2011332084174398e-06,
      "loss": 0.68701273,
      "num_input_tokens_seen": 56607585,
      "step": 2629,
      "time_per_iteration": 3.6746826171875
    },
    {
      "auxiliary_loss_clip": 0.01199595,
      "auxiliary_loss_mlp": 0.01028815,
      "balance_loss_clip": 1.05708504,
      "balance_loss_mlp": 1.01985002,
      "epoch": 0.31623880238080926,
      "flos": 20594087694720.0,
      "grad_norm": 2.436173564798838,
      "language_loss": 0.89527607,
      "learning_rate": 3.2005102712806756e-06,
      "loss": 0.9175601,
      "num_input_tokens_seen": 56626415,
      "step": 2630,
      "time_per_iteration": 2.700340986251831
    },
    {
      "auxiliary_loss_clip": 0.01204967,
      "auxiliary_loss_mlp": 0.01035884,
      "balance_loss_clip": 1.05695403,
      "balance_loss_mlp": 1.02633584,
      "epoch": 0.3163590452714483,
      "flos": 12784917600000.0,
      "grad_norm": 2.021185456789452,
      "language_loss": 0.72835481,
      "learning_rate": 3.1998871520274575e-06,
      "loss": 0.75076336,
      "num_input_tokens_seen": 56641750,
      "step": 2631,
      "time_per_iteration": 3.5509676933288574
    },
    {
      "auxiliary_loss_clip": 0.01204108,
      "auxiliary_loss_mlp": 0.01035448,
      "balance_loss_clip": 1.05277634,
      "balance_loss_mlp": 1.02612567,
      "epoch": 0.3164792881620874,
      "flos": 23041292273280.0,
      "grad_norm": 1.727832785744645,
      "language_loss": 0.84883058,
      "learning_rate": 3.199263850752312e-06,
      "loss": 0.87122619,
      "num_input_tokens_seen": 56662585,
      "step": 2632,
      "time_per_iteration": 2.7325351238250732
    },
    {
      "auxiliary_loss_clip": 0.0120062,
      "auxiliary_loss_mlp": 0.01029511,
      "balance_loss_clip": 1.05496693,
      "balance_loss_mlp": 1.01976573,
      "epoch": 0.31659953105272653,
      "flos": 18296271780480.0,
      "grad_norm": 3.1760285395628625,
      "language_loss": 0.85593796,
      "learning_rate": 3.198640367549795e-06,
      "loss": 0.87823927,
      "num_input_tokens_seen": 56681480,
      "step": 2633,
      "time_per_iteration": 2.619300365447998
    },
    {
      "auxiliary_loss_clip": 0.01200036,
      "auxiliary_loss_mlp": 0.01209086,
      "balance_loss_clip": 1.05478501,
      "balance_loss_mlp": 1.00100565,
      "epoch": 0.3167197739433656,
      "flos": 25703421880320.0,
      "grad_norm": 1.6652802698400018,
      "language_loss": 0.86300689,
      "learning_rate": 3.198016702514487e-06,
      "loss": 0.88709807,
      "num_input_tokens_seen": 56701760,
      "step": 2634,
      "time_per_iteration": 2.7602221965789795
    },
    {
      "auxiliary_loss_clip": 0.01196514,
      "auxiliary_loss_mlp": 0.01029595,
      "balance_loss_clip": 1.05660415,
      "balance_loss_mlp": 1.02077365,
      "epoch": 0.3168400168340047,
      "flos": 23546016230400.0,
      "grad_norm": 1.7647419607427515,
      "language_loss": 0.8455779,
      "learning_rate": 3.1973928557409972e-06,
      "loss": 0.86783898,
      "num_input_tokens_seen": 56719800,
      "step": 2635,
      "time_per_iteration": 2.7028605937957764
    },
    {
      "auxiliary_loss_clip": 0.01196493,
      "auxiliary_loss_mlp": 0.01029305,
      "balance_loss_clip": 1.058002,
      "balance_loss_mlp": 1.02043605,
      "epoch": 0.31696025972464376,
      "flos": 28366449327360.0,
      "grad_norm": 2.282626419713238,
      "language_loss": 0.71086299,
      "learning_rate": 3.1967688273239636e-06,
      "loss": 0.73312092,
      "num_input_tokens_seen": 56739605,
      "step": 2636,
      "time_per_iteration": 3.5607104301452637
    },
    {
      "auxiliary_loss_clip": 0.01204112,
      "auxiliary_loss_mlp": 0.01031268,
      "balance_loss_clip": 1.05226481,
      "balance_loss_mlp": 1.02240443,
      "epoch": 0.31708050261528287,
      "flos": 16399111144320.0,
      "grad_norm": 2.9051726147232557,
      "language_loss": 0.8243683,
      "learning_rate": 3.1961446173580503e-06,
      "loss": 0.84672207,
      "num_input_tokens_seen": 56756545,
      "step": 2637,
      "time_per_iteration": 2.6480119228363037
    },
    {
      "auxiliary_loss_clip": 0.01201464,
      "auxiliary_loss_mlp": 0.01032173,
      "balance_loss_clip": 1.05589175,
      "balance_loss_mlp": 1.02322638,
      "epoch": 0.317200745505922,
      "flos": 26212347728640.0,
      "grad_norm": 2.020244136449935,
      "language_loss": 0.76947653,
      "learning_rate": 3.1955202259379502e-06,
      "loss": 0.7918129,
      "num_input_tokens_seen": 56778275,
      "step": 2638,
      "time_per_iteration": 3.6247923374176025
    },
    {
      "auxiliary_loss_clip": 0.01197568,
      "auxiliary_loss_mlp": 0.01028552,
      "balance_loss_clip": 1.05427289,
      "balance_loss_mlp": 1.01958132,
      "epoch": 0.31732098839656103,
      "flos": 31350876693120.0,
      "grad_norm": 1.789629488864831,
      "language_loss": 0.82596326,
      "learning_rate": 3.194895653158381e-06,
      "loss": 0.84822452,
      "num_input_tokens_seen": 56797215,
      "step": 2639,
      "time_per_iteration": 2.7117562294006348
    },
    {
      "auxiliary_loss_clip": 0.0110059,
      "auxiliary_loss_mlp": 0.01003764,
      "balance_loss_clip": 1.02827454,
      "balance_loss_mlp": 1.00255442,
      "epoch": 0.31744123128720014,
      "flos": 58989024835200.0,
      "grad_norm": 0.7835111224412404,
      "language_loss": 0.5554812,
      "learning_rate": 3.194270899114093e-06,
      "loss": 0.57652473,
      "num_input_tokens_seen": 56863010,
      "step": 2640,
      "time_per_iteration": 3.2303857803344727
    },
    {
      "auxiliary_loss_clip": 0.01205444,
      "auxiliary_loss_mlp": 0.01031699,
      "balance_loss_clip": 1.057145,
      "balance_loss_mlp": 1.02179265,
      "epoch": 0.31756147417783925,
      "flos": 17417573372160.0,
      "grad_norm": 1.8525212387866485,
      "language_loss": 0.82564449,
      "learning_rate": 3.193645963899858e-06,
      "loss": 0.8480159,
      "num_input_tokens_seen": 56880625,
      "step": 2641,
      "time_per_iteration": 2.589508533477783
    },
    {
      "auxiliary_loss_clip": 0.01195066,
      "auxiliary_loss_mlp": 0.01027879,
      "balance_loss_clip": 1.05292845,
      "balance_loss_mlp": 1.01931405,
      "epoch": 0.3176817170684783,
      "flos": 25481673267840.0,
      "grad_norm": 1.8998338110142636,
      "language_loss": 0.83877301,
      "learning_rate": 3.193020847610479e-06,
      "loss": 0.86100245,
      "num_input_tokens_seen": 56900945,
      "step": 2642,
      "time_per_iteration": 2.8086001873016357
    },
    {
      "auxiliary_loss_clip": 0.01195204,
      "auxiliary_loss_mlp": 0.01028163,
      "balance_loss_clip": 1.05490947,
      "balance_loss_mlp": 1.01923466,
      "epoch": 0.3178019599591174,
      "flos": 24972603765120.0,
      "grad_norm": 5.511832991081724,
      "language_loss": 0.71584404,
      "learning_rate": 3.192395550340787e-06,
      "loss": 0.73807776,
      "num_input_tokens_seen": 56918895,
      "step": 2643,
      "time_per_iteration": 2.669886827468872
    },
    {
      "auxiliary_loss_clip": 0.01200452,
      "auxiliary_loss_mlp": 0.01025692,
      "balance_loss_clip": 1.05611324,
      "balance_loss_mlp": 1.01701975,
      "epoch": 0.31792220284975653,
      "flos": 12422220019200.0,
      "grad_norm": 4.790608182209327,
      "language_loss": 0.76852822,
      "learning_rate": 3.191770072185638e-06,
      "loss": 0.79078966,
      "num_input_tokens_seen": 56935890,
      "step": 2644,
      "time_per_iteration": 2.6501214504241943
    },
    {
      "auxiliary_loss_clip": 0.01198549,
      "auxiliary_loss_mlp": 0.01032409,
      "balance_loss_clip": 1.05670428,
      "balance_loss_mlp": 1.02317572,
      "epoch": 0.3180424457403956,
      "flos": 15485759089920.0,
      "grad_norm": 3.346538387369336,
      "language_loss": 0.72679347,
      "learning_rate": 3.191144413239916e-06,
      "loss": 0.74910307,
      "num_input_tokens_seen": 56952460,
      "step": 2645,
      "time_per_iteration": 2.556757688522339
    },
    {
      "auxiliary_loss_clip": 0.01204491,
      "auxiliary_loss_mlp": 0.01032043,
      "balance_loss_clip": 1.05537868,
      "balance_loss_mlp": 1.02280378,
      "epoch": 0.3181626886310347,
      "flos": 26174964648960.0,
      "grad_norm": 1.9955009079863253,
      "language_loss": 0.88197339,
      "learning_rate": 3.190518573598534e-06,
      "loss": 0.90433878,
      "num_input_tokens_seen": 56969065,
      "step": 2646,
      "time_per_iteration": 2.7233383655548096
    },
    {
      "auxiliary_loss_clip": 0.01209781,
      "auxiliary_loss_mlp": 0.01034597,
      "balance_loss_clip": 1.05144286,
      "balance_loss_mlp": 1.02495909,
      "epoch": 0.3182829315216738,
      "flos": 25483109811840.0,
      "grad_norm": 1.6214962557088324,
      "language_loss": 0.77452683,
      "learning_rate": 3.1898925533564308e-06,
      "loss": 0.79697061,
      "num_input_tokens_seen": 56990535,
      "step": 2647,
      "time_per_iteration": 2.729780912399292
    },
    {
      "auxiliary_loss_clip": 0.01192118,
      "auxiliary_loss_mlp": 0.01036134,
      "balance_loss_clip": 1.04994249,
      "balance_loss_mlp": 1.02693152,
      "epoch": 0.31840317441231286,
      "flos": 18113701927680.0,
      "grad_norm": 2.0400944402650922,
      "language_loss": 0.64049852,
      "learning_rate": 3.1892663526085733e-06,
      "loss": 0.662781,
      "num_input_tokens_seen": 57008910,
      "step": 2648,
      "time_per_iteration": 2.6817662715911865
    },
    {
      "auxiliary_loss_clip": 0.01099902,
      "auxiliary_loss_mlp": 0.01001318,
      "balance_loss_clip": 1.02795541,
      "balance_loss_mlp": 1.00012004,
      "epoch": 0.31852341730295197,
      "flos": 64741948957440.0,
      "grad_norm": 0.7525896771162641,
      "language_loss": 0.56923616,
      "learning_rate": 3.188639971449956e-06,
      "loss": 0.59024835,
      "num_input_tokens_seen": 57074960,
      "step": 2649,
      "time_per_iteration": 3.1177761554718018
    },
    {
      "auxiliary_loss_clip": 0.01200593,
      "auxiliary_loss_mlp": 0.01030748,
      "balance_loss_clip": 1.05964148,
      "balance_loss_mlp": 1.0216825,
      "epoch": 0.318643660193591,
      "flos": 20668135582080.0,
      "grad_norm": 2.255199591936253,
      "language_loss": 0.72388166,
      "learning_rate": 3.1880134099756e-06,
      "loss": 0.74619508,
      "num_input_tokens_seen": 57094595,
      "step": 2650,
      "time_per_iteration": 2.6430325508117676
    },
    {
      "auxiliary_loss_clip": 0.01197197,
      "auxiliary_loss_mlp": 0.01025444,
      "balance_loss_clip": 1.05357659,
      "balance_loss_mlp": 1.0165813,
      "epoch": 0.31876390308423014,
      "flos": 26943345411840.0,
      "grad_norm": 1.852953513646633,
      "language_loss": 0.69352758,
      "learning_rate": 3.1873866682805535e-06,
      "loss": 0.71575397,
      "num_input_tokens_seen": 57115290,
      "step": 2651,
      "time_per_iteration": 2.6806206703186035
    },
    {
      "auxiliary_loss_clip": 0.01207616,
      "auxiliary_loss_mlp": 0.01033706,
      "balance_loss_clip": 1.05688572,
      "balance_loss_mlp": 1.02452111,
      "epoch": 0.31888414597486925,
      "flos": 18041916597120.0,
      "grad_norm": 2.396135850076334,
      "language_loss": 0.88806051,
      "learning_rate": 3.186759746459894e-06,
      "loss": 0.9104737,
      "num_input_tokens_seen": 57134400,
      "step": 2652,
      "time_per_iteration": 2.6324658393859863
    },
    {
      "auxiliary_loss_clip": 0.01201083,
      "auxiliary_loss_mlp": 0.01027457,
      "balance_loss_clip": 1.05394304,
      "balance_loss_mlp": 1.01848054,
      "epoch": 0.3190043888655083,
      "flos": 25149319701120.0,
      "grad_norm": 4.64918469873414,
      "language_loss": 0.79660571,
      "learning_rate": 3.1861326446087246e-06,
      "loss": 0.81889117,
      "num_input_tokens_seen": 57153140,
      "step": 2653,
      "time_per_iteration": 2.7112667560577393
    },
    {
      "auxiliary_loss_clip": 0.01202735,
      "auxiliary_loss_mlp": 0.01033764,
      "balance_loss_clip": 1.05600572,
      "balance_loss_mlp": 1.02405453,
      "epoch": 0.3191246317561474,
      "flos": 22053892331520.0,
      "grad_norm": 2.3749131281729254,
      "language_loss": 0.71795923,
      "learning_rate": 3.1855053628221763e-06,
      "loss": 0.7403242,
      "num_input_tokens_seen": 57172395,
      "step": 2654,
      "time_per_iteration": 2.607849597930908
    },
    {
      "auxiliary_loss_clip": 0.01196553,
      "auxiliary_loss_mlp": 0.01031487,
      "balance_loss_clip": 1.04982066,
      "balance_loss_mlp": 1.02165794,
      "epoch": 0.3192448746467865,
      "flos": 14901815687040.0,
      "grad_norm": 2.76660980148581,
      "language_loss": 0.90055603,
      "learning_rate": 3.184877901195407e-06,
      "loss": 0.92283636,
      "num_input_tokens_seen": 57189090,
      "step": 2655,
      "time_per_iteration": 2.764639139175415
    },
    {
      "auxiliary_loss_clip": 0.01112499,
      "auxiliary_loss_mlp": 0.01002277,
      "balance_loss_clip": 1.03378797,
      "balance_loss_mlp": 1.00070369,
      "epoch": 0.3193651175374256,
      "flos": 67234832657280.0,
      "grad_norm": 0.7930487396562054,
      "language_loss": 0.62753999,
      "learning_rate": 3.184250259823602e-06,
      "loss": 0.64868772,
      "num_input_tokens_seen": 57251620,
      "step": 2656,
      "time_per_iteration": 3.2493197917938232
    },
    {
      "auxiliary_loss_clip": 0.01204663,
      "auxiliary_loss_mlp": 0.0103151,
      "balance_loss_clip": 1.05266953,
      "balance_loss_mlp": 1.02142477,
      "epoch": 0.3194853604280647,
      "flos": 12233077977600.0,
      "grad_norm": 2.5210131612227342,
      "language_loss": 0.81386417,
      "learning_rate": 3.183622438801974e-06,
      "loss": 0.83622587,
      "num_input_tokens_seen": 57266910,
      "step": 2657,
      "time_per_iteration": 3.63411545753479
    },
    {
      "auxiliary_loss_clip": 0.0120036,
      "auxiliary_loss_mlp": 0.01033227,
      "balance_loss_clip": 1.05930781,
      "balance_loss_mlp": 1.0241971,
      "epoch": 0.3196056033187038,
      "flos": 14939917038720.0,
      "grad_norm": 1.8153160106870256,
      "language_loss": 0.75190967,
      "learning_rate": 3.1829944382257637e-06,
      "loss": 0.77424556,
      "num_input_tokens_seen": 57285040,
      "step": 2658,
      "time_per_iteration": 3.499904155731201
    },
    {
      "auxiliary_loss_clip": 0.01201011,
      "auxiliary_loss_mlp": 0.01032193,
      "balance_loss_clip": 1.05851901,
      "balance_loss_mlp": 1.02328837,
      "epoch": 0.31972584620934286,
      "flos": 23768878164480.0,
      "grad_norm": 2.3588728508462697,
      "language_loss": 0.81953466,
      "learning_rate": 3.1823662581902373e-06,
      "loss": 0.84186673,
      "num_input_tokens_seen": 57302725,
      "step": 2659,
      "time_per_iteration": 2.674631118774414
    },
    {
      "auxiliary_loss_clip": 0.01190872,
      "auxiliary_loss_mlp": 0.01028127,
      "balance_loss_clip": 1.0465486,
      "balance_loss_mlp": 1.01855493,
      "epoch": 0.31984608909998197,
      "flos": 21251540280960.0,
      "grad_norm": 2.5654208931523192,
      "language_loss": 0.74487168,
      "learning_rate": 3.1817378987906896e-06,
      "loss": 0.76706171,
      "num_input_tokens_seen": 57322230,
      "step": 2660,
      "time_per_iteration": 2.702141761779785
    },
    {
      "auxiliary_loss_clip": 0.01200552,
      "auxiliary_loss_mlp": 0.0103312,
      "balance_loss_clip": 1.05095756,
      "balance_loss_mlp": 1.02373219,
      "epoch": 0.3199663319906211,
      "flos": 18296235866880.0,
      "grad_norm": 2.4123711410675,
      "language_loss": 0.80090922,
      "learning_rate": 3.181109360122442e-06,
      "loss": 0.82324588,
      "num_input_tokens_seen": 57339820,
      "step": 2661,
      "time_per_iteration": 2.672234535217285
    },
    {
      "auxiliary_loss_clip": 0.01200603,
      "auxiliary_loss_mlp": 0.01028613,
      "balance_loss_clip": 1.05033088,
      "balance_loss_mlp": 1.019279,
      "epoch": 0.32008657488126013,
      "flos": 18733627779840.0,
      "grad_norm": 2.250267907922968,
      "language_loss": 0.78717488,
      "learning_rate": 3.1804806422808445e-06,
      "loss": 0.80946708,
      "num_input_tokens_seen": 57356955,
      "step": 2662,
      "time_per_iteration": 3.438459873199463
    },
    {
      "auxiliary_loss_clip": 0.01192313,
      "auxiliary_loss_mlp": 0.01037892,
      "balance_loss_clip": 1.05148983,
      "balance_loss_mlp": 1.02819443,
      "epoch": 0.32020681777189924,
      "flos": 20595344670720.0,
      "grad_norm": 1.997645620796689,
      "language_loss": 0.73113275,
      "learning_rate": 3.1798517453612714e-06,
      "loss": 0.75343478,
      "num_input_tokens_seen": 57376760,
      "step": 2663,
      "time_per_iteration": 2.6796975135803223
    },
    {
      "auxiliary_loss_clip": 0.01198934,
      "auxiliary_loss_mlp": 0.01034764,
      "balance_loss_clip": 1.05900061,
      "balance_loss_mlp": 1.02572179,
      "epoch": 0.32032706066253835,
      "flos": 35261692750080.0,
      "grad_norm": 2.1600201470312195,
      "language_loss": 0.75391781,
      "learning_rate": 3.1792226694591265e-06,
      "loss": 0.77625477,
      "num_input_tokens_seen": 57398145,
      "step": 2664,
      "time_per_iteration": 3.6371493339538574
    },
    {
      "auxiliary_loss_clip": 0.01199237,
      "auxiliary_loss_mlp": 0.01027427,
      "balance_loss_clip": 1.05259562,
      "balance_loss_mlp": 1.01854014,
      "epoch": 0.3204473035531774,
      "flos": 15304230731520.0,
      "grad_norm": 12.603608456585622,
      "language_loss": 0.80432177,
      "learning_rate": 3.178593414669841e-06,
      "loss": 0.82658839,
      "num_input_tokens_seen": 57416730,
      "step": 2665,
      "time_per_iteration": 2.687460422515869
    },
    {
      "auxiliary_loss_clip": 0.01204947,
      "auxiliary_loss_mlp": 0.01029407,
      "balance_loss_clip": 1.05826998,
      "balance_loss_mlp": 1.01983476,
      "epoch": 0.3205675464438165,
      "flos": 24462564595200.0,
      "grad_norm": 3.8086381564504554,
      "language_loss": 0.70863032,
      "learning_rate": 3.1779639810888707e-06,
      "loss": 0.7309739,
      "num_input_tokens_seen": 57436325,
      "step": 2666,
      "time_per_iteration": 2.6524498462677
    },
    {
      "auxiliary_loss_clip": 0.01199452,
      "auxiliary_loss_mlp": 0.01035102,
      "balance_loss_clip": 1.058146,
      "balance_loss_mlp": 1.02545846,
      "epoch": 0.3206877893344556,
      "flos": 22456235548800.0,
      "grad_norm": 2.1693383662751864,
      "language_loss": 0.76188034,
      "learning_rate": 3.1773343688117013e-06,
      "loss": 0.78422588,
      "num_input_tokens_seen": 57457235,
      "step": 2667,
      "time_per_iteration": 2.652186870574951
    },
    {
      "auxiliary_loss_clip": 0.01208031,
      "auxiliary_loss_mlp": 0.01209409,
      "balance_loss_clip": 1.05533373,
      "balance_loss_mlp": 1.00106239,
      "epoch": 0.3208080322250947,
      "flos": 20412236113920.0,
      "grad_norm": 3.415189390073905,
      "language_loss": 0.84260595,
      "learning_rate": 3.1767045779338445e-06,
      "loss": 0.86678028,
      "num_input_tokens_seen": 57474895,
      "step": 2668,
      "time_per_iteration": 2.6725780963897705
    },
    {
      "auxiliary_loss_clip": 0.01196992,
      "auxiliary_loss_mlp": 0.01029279,
      "balance_loss_clip": 1.05272698,
      "balance_loss_mlp": 1.02107728,
      "epoch": 0.3209282751157338,
      "flos": 21762118154880.0,
      "grad_norm": 2.778228346667605,
      "language_loss": 0.9143064,
      "learning_rate": 3.176074608550839e-06,
      "loss": 0.93656909,
      "num_input_tokens_seen": 57490715,
      "step": 2669,
      "time_per_iteration": 2.619267463684082
    },
    {
      "auxiliary_loss_clip": 0.01210646,
      "auxiliary_loss_mlp": 0.01033722,
      "balance_loss_clip": 1.04992867,
      "balance_loss_mlp": 1.02445984,
      "epoch": 0.32104851800637285,
      "flos": 22055041566720.0,
      "grad_norm": 2.256696992105593,
      "language_loss": 0.81864321,
      "learning_rate": 3.17544446075825e-06,
      "loss": 0.84108692,
      "num_input_tokens_seen": 57509880,
      "step": 2670,
      "time_per_iteration": 2.7454521656036377
    },
    {
      "auxiliary_loss_clip": 0.01204867,
      "auxiliary_loss_mlp": 0.01031262,
      "balance_loss_clip": 1.05289197,
      "balance_loss_mlp": 1.02245843,
      "epoch": 0.32116876089701196,
      "flos": 37012301896320.0,
      "grad_norm": 1.8268070890795647,
      "language_loss": 0.71186292,
      "learning_rate": 3.174814134651671e-06,
      "loss": 0.7342242,
      "num_input_tokens_seen": 57532430,
      "step": 2671,
      "time_per_iteration": 2.739452362060547
    },
    {
      "auxiliary_loss_clip": 0.01193553,
      "auxiliary_loss_mlp": 0.01030444,
      "balance_loss_clip": 1.05627751,
      "balance_loss_mlp": 1.02132463,
      "epoch": 0.3212890037876511,
      "flos": 21979233912960.0,
      "grad_norm": 1.739896420850263,
      "language_loss": 0.80665886,
      "learning_rate": 3.1741836303267215e-06,
      "loss": 0.82889891,
      "num_input_tokens_seen": 57551965,
      "step": 2672,
      "time_per_iteration": 2.64749813079834
    },
    {
      "auxiliary_loss_clip": 0.01195827,
      "auxiliary_loss_mlp": 0.01031838,
      "balance_loss_clip": 1.05766189,
      "balance_loss_mlp": 1.02286792,
      "epoch": 0.32140924667829013,
      "flos": 10342345875840.0,
      "grad_norm": 1.919449645451129,
      "language_loss": 0.75461286,
      "learning_rate": 3.1735529478790496e-06,
      "loss": 0.77688944,
      "num_input_tokens_seen": 57569955,
      "step": 2673,
      "time_per_iteration": 2.5472605228424072
    },
    {
      "auxiliary_loss_clip": 0.01200304,
      "auxiliary_loss_mlp": 0.0103502,
      "balance_loss_clip": 1.05554593,
      "balance_loss_mlp": 1.02503574,
      "epoch": 0.32152948956892924,
      "flos": 50798910072960.0,
      "grad_norm": 1.9004005008309994,
      "language_loss": 0.79278046,
      "learning_rate": 3.172922087404328e-06,
      "loss": 0.81513369,
      "num_input_tokens_seen": 57592215,
      "step": 2674,
      "time_per_iteration": 2.9728050231933594
    },
    {
      "auxiliary_loss_clip": 0.01099766,
      "auxiliary_loss_mlp": 0.01004666,
      "balance_loss_clip": 1.02759075,
      "balance_loss_mlp": 1.00347352,
      "epoch": 0.32164973245956835,
      "flos": 63863250549120.0,
      "grad_norm": 0.772901510184787,
      "language_loss": 0.55213308,
      "learning_rate": 3.1722910489982586e-06,
      "loss": 0.57317734,
      "num_input_tokens_seen": 57652575,
      "step": 2675,
      "time_per_iteration": 3.2420902252197266
    },
    {
      "auxiliary_loss_clip": 0.01197539,
      "auxiliary_loss_mlp": 0.01033297,
      "balance_loss_clip": 1.05264854,
      "balance_loss_mlp": 1.02331376,
      "epoch": 0.3217699753502074,
      "flos": 23513948363520.0,
      "grad_norm": 1.6554491974785706,
      "language_loss": 0.80265415,
      "learning_rate": 3.1716598327565694e-06,
      "loss": 0.8249625,
      "num_input_tokens_seen": 57672215,
      "step": 2676,
      "time_per_iteration": 2.752798318862915
    },
    {
      "auxiliary_loss_clip": 0.01195615,
      "auxiliary_loss_mlp": 0.01030992,
      "balance_loss_clip": 1.05718362,
      "balance_loss_mlp": 1.02166343,
      "epoch": 0.3218902182408465,
      "flos": 19062533640960.0,
      "grad_norm": 2.299660359612843,
      "language_loss": 0.84249604,
      "learning_rate": 3.171028438775015e-06,
      "loss": 0.86476207,
      "num_input_tokens_seen": 57691410,
      "step": 2677,
      "time_per_iteration": 2.6656291484832764
    },
    {
      "auxiliary_loss_clip": 0.0119687,
      "auxiliary_loss_mlp": 0.01029709,
      "balance_loss_clip": 1.05731821,
      "balance_loss_mlp": 1.0207088,
      "epoch": 0.3220104611314856,
      "flos": 20375571306240.0,
      "grad_norm": 1.9296160497109698,
      "language_loss": 0.84465617,
      "learning_rate": 3.170396867149377e-06,
      "loss": 0.86692202,
      "num_input_tokens_seen": 57709415,
      "step": 2678,
      "time_per_iteration": 2.5664212703704834
    },
    {
      "auxiliary_loss_clip": 0.01199662,
      "auxiliary_loss_mlp": 0.01035486,
      "balance_loss_clip": 1.05113626,
      "balance_loss_mlp": 1.02591324,
      "epoch": 0.3221307040221247,
      "flos": 20117014231680.0,
      "grad_norm": 1.929362987902816,
      "language_loss": 0.86688584,
      "learning_rate": 3.1697651179754653e-06,
      "loss": 0.88923728,
      "num_input_tokens_seen": 57728075,
      "step": 2679,
      "time_per_iteration": 2.7306864261627197
    },
    {
      "auxiliary_loss_clip": 0.01205756,
      "auxiliary_loss_mlp": 0.01029748,
      "balance_loss_clip": 1.05807912,
      "balance_loss_mlp": 1.02079535,
      "epoch": 0.3222509469127638,
      "flos": 23987789602560.0,
      "grad_norm": 1.7249637546482717,
      "language_loss": 0.73198092,
      "learning_rate": 3.1691331913491153e-06,
      "loss": 0.75433594,
      "num_input_tokens_seen": 57750645,
      "step": 2680,
      "time_per_iteration": 2.761404514312744
    },
    {
      "auxiliary_loss_clip": 0.01196328,
      "auxiliary_loss_mlp": 0.01030152,
      "balance_loss_clip": 1.05508113,
      "balance_loss_mlp": 1.02164054,
      "epoch": 0.32237118980340285,
      "flos": 17675735397120.0,
      "grad_norm": 2.3011136347214167,
      "language_loss": 0.850586,
      "learning_rate": 3.1685010873661898e-06,
      "loss": 0.87285078,
      "num_input_tokens_seen": 57769820,
      "step": 2681,
      "time_per_iteration": 2.6250436305999756
    },
    {
      "auxiliary_loss_clip": 0.01197691,
      "auxiliary_loss_mlp": 0.01032726,
      "balance_loss_clip": 1.05591631,
      "balance_loss_mlp": 1.02293301,
      "epoch": 0.32249143269404196,
      "flos": 23147982645120.0,
      "grad_norm": 1.988254840429406,
      "language_loss": 0.79863894,
      "learning_rate": 3.167868806122578e-06,
      "loss": 0.82094306,
      "num_input_tokens_seen": 57788870,
      "step": 2682,
      "time_per_iteration": 2.6150062084198
    },
    {
      "auxiliary_loss_clip": 0.01203534,
      "auxiliary_loss_mlp": 0.01032422,
      "balance_loss_clip": 1.05399704,
      "balance_loss_mlp": 1.02322519,
      "epoch": 0.32261167558468107,
      "flos": 24422308427520.0,
      "grad_norm": 2.354867347730338,
      "language_loss": 0.66448522,
      "learning_rate": 3.1672363477141968e-06,
      "loss": 0.68684471,
      "num_input_tokens_seen": 57808165,
      "step": 2683,
      "time_per_iteration": 2.676356315612793
    },
    {
      "auxiliary_loss_clip": 0.01204294,
      "auxiliary_loss_mlp": 0.01037165,
      "balance_loss_clip": 1.05289948,
      "balance_loss_mlp": 1.0273664,
      "epoch": 0.3227319184753201,
      "flos": 30367175852160.0,
      "grad_norm": 2.512159381464269,
      "language_loss": 0.84921229,
      "learning_rate": 3.1666037122369903e-06,
      "loss": 0.87162685,
      "num_input_tokens_seen": 57828825,
      "step": 2684,
      "time_per_iteration": 3.677708148956299
    },
    {
      "auxiliary_loss_clip": 0.01196415,
      "auxiliary_loss_mlp": 0.01029449,
      "balance_loss_clip": 1.052109,
      "balance_loss_mlp": 1.02085423,
      "epoch": 0.32285216136595923,
      "flos": 16946174257920.0,
      "grad_norm": 2.4121417110486663,
      "language_loss": 0.86727268,
      "learning_rate": 3.165970899786928e-06,
      "loss": 0.88953137,
      "num_input_tokens_seen": 57846740,
      "step": 2685,
      "time_per_iteration": 3.569166421890259
    },
    {
      "auxiliary_loss_clip": 0.01205709,
      "auxiliary_loss_mlp": 0.01026424,
      "balance_loss_clip": 1.05098939,
      "balance_loss_mlp": 1.01726842,
      "epoch": 0.32297240425659834,
      "flos": 21981532383360.0,
      "grad_norm": 2.0005285080472825,
      "language_loss": 0.75375128,
      "learning_rate": 3.1653379104600067e-06,
      "loss": 0.77607262,
      "num_input_tokens_seen": 57866885,
      "step": 2686,
      "time_per_iteration": 2.7229931354522705
    },
    {
      "auxiliary_loss_clip": 0.01196552,
      "auxiliary_loss_mlp": 0.0102932,
      "balance_loss_clip": 1.05385458,
      "balance_loss_mlp": 1.02073085,
      "epoch": 0.3230926471472374,
      "flos": 22748045639040.0,
      "grad_norm": 1.4189764653922925,
      "language_loss": 0.69430113,
      "learning_rate": 3.164704744352251e-06,
      "loss": 0.71655983,
      "num_input_tokens_seen": 57887690,
      "step": 2687,
      "time_per_iteration": 2.719109535217285
    },
    {
      "auxiliary_loss_clip": 0.01195147,
      "auxiliary_loss_mlp": 0.01032503,
      "balance_loss_clip": 1.05277324,
      "balance_loss_mlp": 1.0241344,
      "epoch": 0.3232128900378765,
      "flos": 16942977947520.0,
      "grad_norm": 1.9841942404265382,
      "language_loss": 0.81176376,
      "learning_rate": 3.164071401559713e-06,
      "loss": 0.83404028,
      "num_input_tokens_seen": 57905090,
      "step": 2688,
      "time_per_iteration": 2.737208604812622
    },
    {
      "auxiliary_loss_clip": 0.01202042,
      "auxiliary_loss_mlp": 0.01032396,
      "balance_loss_clip": 1.05440044,
      "balance_loss_mlp": 1.023265,
      "epoch": 0.3233331329285156,
      "flos": 24023736138240.0,
      "grad_norm": 2.2056943238963367,
      "language_loss": 0.71368897,
      "learning_rate": 3.1634378821784674e-06,
      "loss": 0.73603338,
      "num_input_tokens_seen": 57925305,
      "step": 2689,
      "time_per_iteration": 3.654986619949341
    },
    {
      "auxiliary_loss_clip": 0.01204636,
      "auxiliary_loss_mlp": 0.01027978,
      "balance_loss_clip": 1.05366123,
      "balance_loss_mlp": 1.01920986,
      "epoch": 0.3234533758191547,
      "flos": 18113845582080.0,
      "grad_norm": 3.271269279951908,
      "language_loss": 0.7402702,
      "learning_rate": 3.1628041863046208e-06,
      "loss": 0.76259637,
      "num_input_tokens_seen": 57942720,
      "step": 2690,
      "time_per_iteration": 2.659047842025757
    },
    {
      "auxiliary_loss_clip": 0.01199222,
      "auxiliary_loss_mlp": 0.01031444,
      "balance_loss_clip": 1.05504537,
      "balance_loss_mlp": 1.02151966,
      "epoch": 0.3235736187097938,
      "flos": 16946138344320.0,
      "grad_norm": 2.5397035055352544,
      "language_loss": 0.91303867,
      "learning_rate": 3.162170314034304e-06,
      "loss": 0.93534529,
      "num_input_tokens_seen": 57960135,
      "step": 2691,
      "time_per_iteration": 3.546694755554199
    },
    {
      "auxiliary_loss_clip": 0.01198652,
      "auxiliary_loss_mlp": 0.0103329,
      "balance_loss_clip": 1.05646253,
      "balance_loss_mlp": 1.02360964,
      "epoch": 0.3236938616004329,
      "flos": 22127150119680.0,
      "grad_norm": 1.937426119132582,
      "language_loss": 0.80957127,
      "learning_rate": 3.1615362654636738e-06,
      "loss": 0.8318907,
      "num_input_tokens_seen": 57980875,
      "step": 2692,
      "time_per_iteration": 2.6630961894989014
    },
    {
      "auxiliary_loss_clip": 0.01193834,
      "auxiliary_loss_mlp": 0.01031477,
      "balance_loss_clip": 1.05307138,
      "balance_loss_mlp": 1.02321029,
      "epoch": 0.32381410449107195,
      "flos": 17164618819200.0,
      "grad_norm": 2.1753385303929598,
      "language_loss": 0.87097764,
      "learning_rate": 3.1609020406889163e-06,
      "loss": 0.89323074,
      "num_input_tokens_seen": 57998310,
      "step": 2693,
      "time_per_iteration": 2.782418727874756
    },
    {
      "auxiliary_loss_clip": 0.01202055,
      "auxiliary_loss_mlp": 0.01031017,
      "balance_loss_clip": 1.05267417,
      "balance_loss_mlp": 1.02080035,
      "epoch": 0.32393434738171106,
      "flos": 16578125550720.0,
      "grad_norm": 1.64823148154409,
      "language_loss": 0.84959114,
      "learning_rate": 3.1602676398062416e-06,
      "loss": 0.8719219,
      "num_input_tokens_seen": 58017220,
      "step": 2694,
      "time_per_iteration": 2.6401660442352295
    },
    {
      "auxiliary_loss_clip": 0.01194758,
      "auxiliary_loss_mlp": 0.01027516,
      "balance_loss_clip": 1.05462313,
      "balance_loss_mlp": 1.01815808,
      "epoch": 0.3240545902723502,
      "flos": 25483612602240.0,
      "grad_norm": 3.445913442917519,
      "language_loss": 0.61571372,
      "learning_rate": 3.1596330629118886e-06,
      "loss": 0.63793647,
      "num_input_tokens_seen": 58037190,
      "step": 2695,
      "time_per_iteration": 2.6672091484069824
    },
    {
      "auxiliary_loss_clip": 0.01198749,
      "auxiliary_loss_mlp": 0.0103102,
      "balance_loss_clip": 1.0502193,
      "balance_loss_mlp": 1.02144122,
      "epoch": 0.32417483316298923,
      "flos": 35845851634560.0,
      "grad_norm": 1.913160856844998,
      "language_loss": 0.73129106,
      "learning_rate": 3.1589983101021223e-06,
      "loss": 0.7535888,
      "num_input_tokens_seen": 58055820,
      "step": 2696,
      "time_per_iteration": 2.8751676082611084
    },
    {
      "auxiliary_loss_clip": 0.01200889,
      "auxiliary_loss_mlp": 0.01027169,
      "balance_loss_clip": 1.05292296,
      "balance_loss_mlp": 1.01830029,
      "epoch": 0.32429507605362834,
      "flos": 30080501406720.0,
      "grad_norm": 6.398426216199275,
      "language_loss": 0.85117042,
      "learning_rate": 3.1583633814732337e-06,
      "loss": 0.87345105,
      "num_input_tokens_seen": 58075340,
      "step": 2697,
      "time_per_iteration": 2.7302942276000977
    },
    {
      "auxiliary_loss_clip": 0.01193227,
      "auxiliary_loss_mlp": 0.01036865,
      "balance_loss_clip": 1.05463302,
      "balance_loss_mlp": 1.0279007,
      "epoch": 0.3244153189442674,
      "flos": 18223265387520.0,
      "grad_norm": 2.463478598376056,
      "language_loss": 0.71317619,
      "learning_rate": 3.157728277121541e-06,
      "loss": 0.73547709,
      "num_input_tokens_seen": 58093515,
      "step": 2698,
      "time_per_iteration": 2.659496307373047
    },
    {
      "auxiliary_loss_clip": 0.01193702,
      "auxiliary_loss_mlp": 0.01026618,
      "balance_loss_clip": 1.05220342,
      "balance_loss_mlp": 1.01732516,
      "epoch": 0.3245355618349065,
      "flos": 17710317216000.0,
      "grad_norm": 2.4549920150703324,
      "language_loss": 0.78578836,
      "learning_rate": 3.1570929971433897e-06,
      "loss": 0.8079915,
      "num_input_tokens_seen": 58109300,
      "step": 2699,
      "time_per_iteration": 2.566309928894043
    },
    {
      "auxiliary_loss_clip": 0.01197156,
      "auxiliary_loss_mlp": 0.01037342,
      "balance_loss_clip": 1.05677009,
      "balance_loss_mlp": 1.02858567,
      "epoch": 0.3246558047255456,
      "flos": 23440798316160.0,
      "grad_norm": 2.4366852413232643,
      "language_loss": 0.83911788,
      "learning_rate": 3.1564575416351504e-06,
      "loss": 0.86146283,
      "num_input_tokens_seen": 58128000,
      "step": 2700,
      "time_per_iteration": 2.6804471015930176
    },
    {
      "auxiliary_loss_clip": 0.01195789,
      "auxiliary_loss_mlp": 0.01029223,
      "balance_loss_clip": 1.05598247,
      "balance_loss_mlp": 1.01987112,
      "epoch": 0.32477604761618467,
      "flos": 21760861178880.0,
      "grad_norm": 2.0823063361756455,
      "language_loss": 0.74956453,
      "learning_rate": 3.155821910693221e-06,
      "loss": 0.77181458,
      "num_input_tokens_seen": 58147415,
      "step": 2701,
      "time_per_iteration": 2.5681874752044678
    },
    {
      "auxiliary_loss_clip": 0.01200221,
      "auxiliary_loss_mlp": 0.01032854,
      "balance_loss_clip": 1.0530405,
      "balance_loss_mlp": 1.02335286,
      "epoch": 0.3248962905068238,
      "flos": 19828328624640.0,
      "grad_norm": 6.133629594164037,
      "language_loss": 0.86089176,
      "learning_rate": 3.1551861044140275e-06,
      "loss": 0.88322246,
      "num_input_tokens_seen": 58167050,
      "step": 2702,
      "time_per_iteration": 2.7537429332733154
    },
    {
      "auxiliary_loss_clip": 0.01200887,
      "auxiliary_loss_mlp": 0.01030369,
      "balance_loss_clip": 1.05213082,
      "balance_loss_mlp": 1.02159524,
      "epoch": 0.3250165333974629,
      "flos": 23948215793280.0,
      "grad_norm": 1.9323012957141792,
      "language_loss": 0.7752893,
      "learning_rate": 3.15455012289402e-06,
      "loss": 0.79760188,
      "num_input_tokens_seen": 58186695,
      "step": 2703,
      "time_per_iteration": 2.70719575881958
    },
    {
      "auxiliary_loss_clip": 0.01201086,
      "auxiliary_loss_mlp": 0.01032338,
      "balance_loss_clip": 1.05736351,
      "balance_loss_mlp": 1.02263427,
      "epoch": 0.32513677628810195,
      "flos": 23989333887360.0,
      "grad_norm": 1.9780584162901447,
      "language_loss": 0.84122366,
      "learning_rate": 3.153913966229677e-06,
      "loss": 0.86355793,
      "num_input_tokens_seen": 58205815,
      "step": 2704,
      "time_per_iteration": 2.6781725883483887
    },
    {
      "auxiliary_loss_clip": 0.01109831,
      "auxiliary_loss_mlp": 0.01001949,
      "balance_loss_clip": 1.03137302,
      "balance_loss_mlp": 1.00069702,
      "epoch": 0.32525701917874106,
      "flos": 70655790009600.0,
      "grad_norm": 0.6366558336341019,
      "language_loss": 0.50231117,
      "learning_rate": 3.1532776345175027e-06,
      "loss": 0.52342892,
      "num_input_tokens_seen": 58270960,
      "step": 2705,
      "time_per_iteration": 3.1707870960235596
    },
    {
      "auxiliary_loss_clip": 0.01194238,
      "auxiliary_loss_mlp": 0.01032373,
      "balance_loss_clip": 1.05598569,
      "balance_loss_mlp": 1.0230391,
      "epoch": 0.32537726206938017,
      "flos": 19682639061120.0,
      "grad_norm": 1.794017643974846,
      "language_loss": 0.78504872,
      "learning_rate": 3.1526411278540285e-06,
      "loss": 0.80731487,
      "num_input_tokens_seen": 58289390,
      "step": 2706,
      "time_per_iteration": 2.6031057834625244
    },
    {
      "auxiliary_loss_clip": 0.01208068,
      "auxiliary_loss_mlp": 0.01033524,
      "balance_loss_clip": 1.05328941,
      "balance_loss_mlp": 1.02365947,
      "epoch": 0.3254975049600192,
      "flos": 28760999293440.0,
      "grad_norm": 2.172682712685635,
      "language_loss": 0.81097162,
      "learning_rate": 3.1520044463358116e-06,
      "loss": 0.83338761,
      "num_input_tokens_seen": 58306120,
      "step": 2707,
      "time_per_iteration": 2.7643063068389893
    },
    {
      "auxiliary_loss_clip": 0.0119938,
      "auxiliary_loss_mlp": 0.01028171,
      "balance_loss_clip": 1.0565846,
      "balance_loss_mlp": 1.01896846,
      "epoch": 0.32561774785065833,
      "flos": 18877378008960.0,
      "grad_norm": 1.5682206919156418,
      "language_loss": 0.80476916,
      "learning_rate": 3.151367590059436e-06,
      "loss": 0.82704461,
      "num_input_tokens_seen": 58324545,
      "step": 2708,
      "time_per_iteration": 2.6144886016845703
    },
    {
      "auxiliary_loss_clip": 0.01198072,
      "auxiliary_loss_mlp": 0.01209883,
      "balance_loss_clip": 1.05688918,
      "balance_loss_mlp": 1.00118828,
      "epoch": 0.32573799074129745,
      "flos": 23112107936640.0,
      "grad_norm": 2.089431318434712,
      "language_loss": 0.86656988,
      "learning_rate": 3.1507305591215117e-06,
      "loss": 0.89064938,
      "num_input_tokens_seen": 58342455,
      "step": 2709,
      "time_per_iteration": 2.654576539993286
    },
    {
      "auxiliary_loss_clip": 0.01111217,
      "auxiliary_loss_mlp": 0.01001883,
      "balance_loss_clip": 1.03298974,
      "balance_loss_mlp": 1.00057817,
      "epoch": 0.3258582336319365,
      "flos": 71237650423680.0,
      "grad_norm": 0.6696229841113791,
      "language_loss": 0.55684006,
      "learning_rate": 3.150093353618677e-06,
      "loss": 0.57797104,
      "num_input_tokens_seen": 58407185,
      "step": 2710,
      "time_per_iteration": 3.232940912246704
    },
    {
      "auxiliary_loss_clip": 0.01201248,
      "auxiliary_loss_mlp": 0.01029561,
      "balance_loss_clip": 1.05286908,
      "balance_loss_mlp": 1.02005434,
      "epoch": 0.3259784765225756,
      "flos": 22456020067200.0,
      "grad_norm": 3.2715762181547445,
      "language_loss": 0.88704443,
      "learning_rate": 3.149455973647596e-06,
      "loss": 0.90935248,
      "num_input_tokens_seen": 58425245,
      "step": 2711,
      "time_per_iteration": 3.7299225330352783
    },
    {
      "auxiliary_loss_clip": 0.01194105,
      "auxiliary_loss_mlp": 0.01026214,
      "balance_loss_clip": 1.04759204,
      "balance_loss_mlp": 1.01575339,
      "epoch": 0.32609871941321467,
      "flos": 20484811543680.0,
      "grad_norm": 2.3204417104166883,
      "language_loss": 0.77275985,
      "learning_rate": 3.1488184193049563e-06,
      "loss": 0.79496312,
      "num_input_tokens_seen": 58444780,
      "step": 2712,
      "time_per_iteration": 3.577704429626465
    },
    {
      "auxiliary_loss_clip": 0.01197903,
      "auxiliary_loss_mlp": 0.01029396,
      "balance_loss_clip": 1.05862117,
      "balance_loss_mlp": 1.0204438,
      "epoch": 0.3262189623038538,
      "flos": 22416805393920.0,
      "grad_norm": 2.2510609765166376,
      "language_loss": 0.72379005,
      "learning_rate": 3.1481806906874767e-06,
      "loss": 0.74606305,
      "num_input_tokens_seen": 58466090,
      "step": 2713,
      "time_per_iteration": 2.6596975326538086
    },
    {
      "auxiliary_loss_clip": 0.01195395,
      "auxiliary_loss_mlp": 0.01028045,
      "balance_loss_clip": 1.05649602,
      "balance_loss_mlp": 1.01960528,
      "epoch": 0.3263392051944929,
      "flos": 20923496346240.0,
      "grad_norm": 1.593707362077002,
      "language_loss": 0.87833291,
      "learning_rate": 3.147542787891899e-06,
      "loss": 0.90056729,
      "num_input_tokens_seen": 58485435,
      "step": 2714,
      "time_per_iteration": 2.665440082550049
    },
    {
      "auxiliary_loss_clip": 0.01202886,
      "auxiliary_loss_mlp": 0.01032844,
      "balance_loss_clip": 1.05578053,
      "balance_loss_mlp": 1.02292001,
      "epoch": 0.32645944808513194,
      "flos": 24025172682240.0,
      "grad_norm": 2.161219018899487,
      "language_loss": 0.75325906,
      "learning_rate": 3.1469047110149926e-06,
      "loss": 0.77561635,
      "num_input_tokens_seen": 58504175,
      "step": 2715,
      "time_per_iteration": 2.6812968254089355
    },
    {
      "auxiliary_loss_clip": 0.01199984,
      "auxiliary_loss_mlp": 0.01027888,
      "balance_loss_clip": 1.05065942,
      "balance_loss_mlp": 1.01881576,
      "epoch": 0.32657969097577105,
      "flos": 21032413361280.0,
      "grad_norm": 1.8963494145153432,
      "language_loss": 0.85198438,
      "learning_rate": 3.146266460153554e-06,
      "loss": 0.87426317,
      "num_input_tokens_seen": 58523885,
      "step": 2716,
      "time_per_iteration": 3.5847747325897217
    },
    {
      "auxiliary_loss_clip": 0.01202693,
      "auxiliary_loss_mlp": 0.012099,
      "balance_loss_clip": 1.05624413,
      "balance_loss_mlp": 1.00120378,
      "epoch": 0.32669993386641016,
      "flos": 22710267509760.0,
      "grad_norm": 1.5870483994857476,
      "language_loss": 0.80360186,
      "learning_rate": 3.145628035404404e-06,
      "loss": 0.82772774,
      "num_input_tokens_seen": 58543085,
      "step": 2717,
      "time_per_iteration": 2.671298027038574
    },
    {
      "auxiliary_loss_clip": 0.01109818,
      "auxiliary_loss_mlp": 0.0100114,
      "balance_loss_clip": 1.03304136,
      "balance_loss_mlp": 0.99979347,
      "epoch": 0.3268201767570492,
      "flos": 72105718406400.0,
      "grad_norm": 2.761635423638334,
      "language_loss": 0.57491553,
      "learning_rate": 3.1449894368643922e-06,
      "loss": 0.59602523,
      "num_input_tokens_seen": 58605400,
      "step": 2718,
      "time_per_iteration": 4.139817953109741
    },
    {
      "auxiliary_loss_clip": 0.01203819,
      "auxiliary_loss_mlp": 0.01031006,
      "balance_loss_clip": 1.05616856,
      "balance_loss_mlp": 1.02170181,
      "epoch": 0.32694041964768833,
      "flos": 24535175938560.0,
      "grad_norm": 1.912662244111736,
      "language_loss": 0.71338511,
      "learning_rate": 3.1443506646303934e-06,
      "loss": 0.73573339,
      "num_input_tokens_seen": 58626700,
      "step": 2719,
      "time_per_iteration": 2.726626396179199
    },
    {
      "auxiliary_loss_clip": 0.01202738,
      "auxiliary_loss_mlp": 0.01027527,
      "balance_loss_clip": 1.05521393,
      "balance_loss_mlp": 1.01779413,
      "epoch": 0.32706066253832744,
      "flos": 33183003755520.0,
      "grad_norm": 2.4522258849194976,
      "language_loss": 0.66989136,
      "learning_rate": 3.1437117187993086e-06,
      "loss": 0.69219398,
      "num_input_tokens_seen": 58649020,
      "step": 2720,
      "time_per_iteration": 2.8354012966156006
    },
    {
      "auxiliary_loss_clip": 0.01197233,
      "auxiliary_loss_mlp": 0.01034812,
      "balance_loss_clip": 1.04980707,
      "balance_loss_mlp": 1.02550173,
      "epoch": 0.3271809054289665,
      "flos": 24061622008320.0,
      "grad_norm": 1.986685334992588,
      "language_loss": 0.80201256,
      "learning_rate": 3.143072599468065e-06,
      "loss": 0.82433301,
      "num_input_tokens_seen": 58668845,
      "step": 2721,
      "time_per_iteration": 2.7210464477539062
    },
    {
      "auxiliary_loss_clip": 0.01202441,
      "auxiliary_loss_mlp": 0.01027189,
      "balance_loss_clip": 1.05575275,
      "balance_loss_mlp": 1.01806331,
      "epoch": 0.3273011483196056,
      "flos": 38253769712640.0,
      "grad_norm": 1.6711080625451267,
      "language_loss": 0.75788283,
      "learning_rate": 3.1424333067336174e-06,
      "loss": 0.78017914,
      "num_input_tokens_seen": 58691610,
      "step": 2722,
      "time_per_iteration": 2.84368896484375
    },
    {
      "auxiliary_loss_clip": 0.01204163,
      "auxiliary_loss_mlp": 0.01032198,
      "balance_loss_clip": 1.05432224,
      "balance_loss_mlp": 1.0220232,
      "epoch": 0.3274213912102447,
      "flos": 29054389582080.0,
      "grad_norm": 2.029695566965948,
      "language_loss": 0.78047305,
      "learning_rate": 3.141793840692945e-06,
      "loss": 0.80283666,
      "num_input_tokens_seen": 58712360,
      "step": 2723,
      "time_per_iteration": 2.6846022605895996
    },
    {
      "auxiliary_loss_clip": 0.01193815,
      "auxiliary_loss_mlp": 0.01032495,
      "balance_loss_clip": 1.05282974,
      "balance_loss_mlp": 1.02237439,
      "epoch": 0.32754163410088377,
      "flos": 29133249891840.0,
      "grad_norm": 2.0304309268741507,
      "language_loss": 0.6144622,
      "learning_rate": 3.1411542014430553e-06,
      "loss": 0.63672531,
      "num_input_tokens_seen": 58733440,
      "step": 2724,
      "time_per_iteration": 2.7160181999206543
    },
    {
      "auxiliary_loss_clip": 0.01200571,
      "auxiliary_loss_mlp": 0.01027913,
      "balance_loss_clip": 1.04881299,
      "balance_loss_mlp": 1.01941955,
      "epoch": 0.3276618769915229,
      "flos": 20631075724800.0,
      "grad_norm": 2.3044181194012556,
      "language_loss": 0.8178212,
      "learning_rate": 3.1405143890809804e-06,
      "loss": 0.84010607,
      "num_input_tokens_seen": 58752735,
      "step": 2725,
      "time_per_iteration": 2.756392002105713
    },
    {
      "auxiliary_loss_clip": 0.01199422,
      "auxiliary_loss_mlp": 0.01030444,
      "balance_loss_clip": 1.0533936,
      "balance_loss_mlp": 1.02172971,
      "epoch": 0.327782119882162,
      "flos": 18657425076480.0,
      "grad_norm": 1.9050182958732818,
      "language_loss": 0.69745398,
      "learning_rate": 3.1398744037037796e-06,
      "loss": 0.71975261,
      "num_input_tokens_seen": 58772070,
      "step": 2726,
      "time_per_iteration": 2.672581672668457
    },
    {
      "auxiliary_loss_clip": 0.01203399,
      "auxiliary_loss_mlp": 0.01028352,
      "balance_loss_clip": 1.05712032,
      "balance_loss_mlp": 1.01987576,
      "epoch": 0.32790236277280105,
      "flos": 21795802133760.0,
      "grad_norm": 2.0210870360831352,
      "language_loss": 0.84527582,
      "learning_rate": 3.139234245408538e-06,
      "loss": 0.86759329,
      "num_input_tokens_seen": 58790950,
      "step": 2727,
      "time_per_iteration": 2.8508646488189697
    },
    {
      "auxiliary_loss_clip": 0.01205998,
      "auxiliary_loss_mlp": 0.01209418,
      "balance_loss_clip": 1.05465531,
      "balance_loss_mlp": 1.00132346,
      "epoch": 0.32802260566344016,
      "flos": 23331414424320.0,
      "grad_norm": 1.3477999007475845,
      "language_loss": 0.76060778,
      "learning_rate": 3.1385939142923666e-06,
      "loss": 0.78476197,
      "num_input_tokens_seen": 58813340,
      "step": 2728,
      "time_per_iteration": 2.7314889430999756
    },
    {
      "auxiliary_loss_clip": 0.0120367,
      "auxiliary_loss_mlp": 0.01029326,
      "balance_loss_clip": 1.05318666,
      "balance_loss_mlp": 1.01975965,
      "epoch": 0.3281428485540792,
      "flos": 24206988349440.0,
      "grad_norm": 2.050775193020076,
      "language_loss": 0.78070593,
      "learning_rate": 3.137953410452405e-06,
      "loss": 0.80303591,
      "num_input_tokens_seen": 58833610,
      "step": 2729,
      "time_per_iteration": 2.6518959999084473
    },
    {
      "auxiliary_loss_clip": 0.01198119,
      "auxiliary_loss_mlp": 0.01029934,
      "balance_loss_clip": 1.05084443,
      "balance_loss_mlp": 1.02128589,
      "epoch": 0.3282630914447183,
      "flos": 34128962380800.0,
      "grad_norm": 1.6147519585080237,
      "language_loss": 0.74549687,
      "learning_rate": 3.1373127339858146e-06,
      "loss": 0.76777738,
      "num_input_tokens_seen": 58856210,
      "step": 2730,
      "time_per_iteration": 2.7763397693634033
    },
    {
      "auxiliary_loss_clip": 0.01198011,
      "auxiliary_loss_mlp": 0.01027664,
      "balance_loss_clip": 1.04992938,
      "balance_loss_mlp": 1.01969481,
      "epoch": 0.32838333433535744,
      "flos": 27600726170880.0,
      "grad_norm": 2.360745916837081,
      "language_loss": 0.74465656,
      "learning_rate": 3.136671884989787e-06,
      "loss": 0.76691335,
      "num_input_tokens_seen": 58876120,
      "step": 2731,
      "time_per_iteration": 2.7871103286743164
    },
    {
      "auxiliary_loss_clip": 0.01212823,
      "auxiliary_loss_mlp": 0.0102782,
      "balance_loss_clip": 1.05042124,
      "balance_loss_mlp": 1.01853347,
      "epoch": 0.3285035772259965,
      "flos": 12349500935040.0,
      "grad_norm": 2.202474068049025,
      "language_loss": 0.87032485,
      "learning_rate": 3.1360308635615383e-06,
      "loss": 0.89273131,
      "num_input_tokens_seen": 58894660,
      "step": 2732,
      "time_per_iteration": 2.7951579093933105
    },
    {
      "auxiliary_loss_clip": 0.01207132,
      "auxiliary_loss_mlp": 0.01027239,
      "balance_loss_clip": 1.05441606,
      "balance_loss_mlp": 1.01723742,
      "epoch": 0.3286238201166356,
      "flos": 24316084932480.0,
      "grad_norm": 1.8660181036886534,
      "language_loss": 0.79041541,
      "learning_rate": 3.135389669798311e-06,
      "loss": 0.8127591,
      "num_input_tokens_seen": 58912720,
      "step": 2733,
      "time_per_iteration": 2.7243456840515137
    },
    {
      "auxiliary_loss_clip": 0.01197114,
      "auxiliary_loss_mlp": 0.0120899,
      "balance_loss_clip": 1.05365336,
      "balance_loss_mlp": 1.00129342,
      "epoch": 0.3287440630072747,
      "flos": 21392812471680.0,
      "grad_norm": 2.269697561861382,
      "language_loss": 0.80168766,
      "learning_rate": 3.134748303797373e-06,
      "loss": 0.82574868,
      "num_input_tokens_seen": 58930090,
      "step": 2734,
      "time_per_iteration": 2.7320659160614014
    },
    {
      "auxiliary_loss_clip": 0.01207641,
      "auxiliary_loss_mlp": 0.01033319,
      "balance_loss_clip": 1.04940581,
      "balance_loss_mlp": 1.02315676,
      "epoch": 0.32886430589791377,
      "flos": 23732536579200.0,
      "grad_norm": 4.403235882128202,
      "language_loss": 0.81192446,
      "learning_rate": 3.1341067656560203e-06,
      "loss": 0.83433402,
      "num_input_tokens_seen": 58947935,
      "step": 2735,
      "time_per_iteration": 2.7063376903533936
    },
    {
      "auxiliary_loss_clip": 0.01210859,
      "auxiliary_loss_mlp": 0.01026044,
      "balance_loss_clip": 1.05476761,
      "balance_loss_mlp": 1.0163641,
      "epoch": 0.3289845487885529,
      "flos": 22418708814720.0,
      "grad_norm": 2.731566568717726,
      "language_loss": 0.86672568,
      "learning_rate": 3.133465055471572e-06,
      "loss": 0.88909471,
      "num_input_tokens_seen": 58967720,
      "step": 2736,
      "time_per_iteration": 2.680724859237671
    },
    {
      "auxiliary_loss_clip": 0.0119907,
      "auxiliary_loss_mlp": 0.01028426,
      "balance_loss_clip": 1.05128038,
      "balance_loss_mlp": 1.01943195,
      "epoch": 0.329104791679192,
      "flos": 19682603147520.0,
      "grad_norm": 2.5729166130307415,
      "language_loss": 0.66274923,
      "learning_rate": 3.1328231733413767e-06,
      "loss": 0.6850242,
      "num_input_tokens_seen": 58984360,
      "step": 2737,
      "time_per_iteration": 3.6616759300231934
    },
    {
      "auxiliary_loss_clip": 0.01195693,
      "auxiliary_loss_mlp": 0.01033444,
      "balance_loss_clip": 1.05492008,
      "balance_loss_mlp": 1.02368093,
      "epoch": 0.32922503456983104,
      "flos": 15997234803840.0,
      "grad_norm": 2.404941350134084,
      "language_loss": 0.91141617,
      "learning_rate": 3.1321811193628067e-06,
      "loss": 0.93370759,
      "num_input_tokens_seen": 59002505,
      "step": 2738,
      "time_per_iteration": 3.5429983139038086
    },
    {
      "auxiliary_loss_clip": 0.01201929,
      "auxiliary_loss_mlp": 0.01209556,
      "balance_loss_clip": 1.05797541,
      "balance_loss_mlp": 1.00113952,
      "epoch": 0.32934527746047015,
      "flos": 26834069260800.0,
      "grad_norm": 1.950935735489035,
      "language_loss": 0.69989836,
      "learning_rate": 3.131538893633261e-06,
      "loss": 0.72401321,
      "num_input_tokens_seen": 59022065,
      "step": 2739,
      "time_per_iteration": 2.6827893257141113
    },
    {
      "auxiliary_loss_clip": 0.01198139,
      "auxiliary_loss_mlp": 0.01030871,
      "balance_loss_clip": 1.0571816,
      "balance_loss_mlp": 1.02218008,
      "epoch": 0.32946552035110926,
      "flos": 23403774372480.0,
      "grad_norm": 2.3524155559264117,
      "language_loss": 0.78110778,
      "learning_rate": 3.130896496250165e-06,
      "loss": 0.80339789,
      "num_input_tokens_seen": 59041890,
      "step": 2740,
      "time_per_iteration": 2.657207489013672
    },
    {
      "auxiliary_loss_clip": 0.01198302,
      "auxiliary_loss_mlp": 0.01030328,
      "balance_loss_clip": 1.05687165,
      "balance_loss_mlp": 1.02151847,
      "epoch": 0.3295857632417483,
      "flos": 14172470029440.0,
      "grad_norm": 2.3328148401191533,
      "language_loss": 0.86606926,
      "learning_rate": 3.1302539273109693e-06,
      "loss": 0.88835555,
      "num_input_tokens_seen": 59058715,
      "step": 2741,
      "time_per_iteration": 2.5981712341308594
    },
    {
      "auxiliary_loss_clip": 0.01193731,
      "auxiliary_loss_mlp": 0.01031427,
      "balance_loss_clip": 1.05333185,
      "balance_loss_mlp": 1.02143705,
      "epoch": 0.32970600613238743,
      "flos": 22196708807040.0,
      "grad_norm": 1.9496181113940636,
      "language_loss": 0.80502498,
      "learning_rate": 3.1296111869131513e-06,
      "loss": 0.82727659,
      "num_input_tokens_seen": 59076140,
      "step": 2742,
      "time_per_iteration": 2.725405216217041
    },
    {
      "auxiliary_loss_clip": 0.01196252,
      "auxiliary_loss_mlp": 0.0103113,
      "balance_loss_clip": 1.05680883,
      "balance_loss_mlp": 1.02215958,
      "epoch": 0.32982624902302654,
      "flos": 22053784590720.0,
      "grad_norm": 2.0031607558328943,
      "language_loss": 0.85870683,
      "learning_rate": 3.1289682751542153e-06,
      "loss": 0.88098061,
      "num_input_tokens_seen": 59095700,
      "step": 2743,
      "time_per_iteration": 3.6307291984558105
    },
    {
      "auxiliary_loss_clip": 0.01197717,
      "auxiliary_loss_mlp": 0.01028841,
      "balance_loss_clip": 1.0557059,
      "balance_loss_mlp": 1.01968551,
      "epoch": 0.3299464919136656,
      "flos": 18661626967680.0,
      "grad_norm": 1.8800688405339796,
      "language_loss": 0.71347415,
      "learning_rate": 3.1283251921316883e-06,
      "loss": 0.73573971,
      "num_input_tokens_seen": 59113445,
      "step": 2744,
      "time_per_iteration": 2.6083478927612305
    },
    {
      "auxiliary_loss_clip": 0.01203523,
      "auxiliary_loss_mlp": 0.01033774,
      "balance_loss_clip": 1.05228806,
      "balance_loss_mlp": 1.02414203,
      "epoch": 0.3300667348043047,
      "flos": 13407357404160.0,
      "grad_norm": 1.9687762184585378,
      "language_loss": 0.81353027,
      "learning_rate": 3.1276819379431277e-06,
      "loss": 0.83590329,
      "num_input_tokens_seen": 59131535,
      "step": 2745,
      "time_per_iteration": 3.6122078895568848
    },
    {
      "auxiliary_loss_clip": 0.01210533,
      "auxiliary_loss_mlp": 0.01209888,
      "balance_loss_clip": 1.05575895,
      "balance_loss_mlp": 1.00107121,
      "epoch": 0.33018697769494376,
      "flos": 15742556398080.0,
      "grad_norm": 2.2314431189084294,
      "language_loss": 0.75632244,
      "learning_rate": 3.1270385126861134e-06,
      "loss": 0.78052664,
      "num_input_tokens_seen": 59149520,
      "step": 2746,
      "time_per_iteration": 2.6397268772125244
    },
    {
      "auxiliary_loss_clip": 0.01199273,
      "auxiliary_loss_mlp": 0.01036577,
      "balance_loss_clip": 1.05821967,
      "balance_loss_mlp": 1.02627766,
      "epoch": 0.3303072205855829,
      "flos": 18258601392000.0,
      "grad_norm": 1.7181149674185008,
      "language_loss": 0.82009375,
      "learning_rate": 3.1263949164582533e-06,
      "loss": 0.84245217,
      "num_input_tokens_seen": 59169170,
      "step": 2747,
      "time_per_iteration": 2.6626815795898438
    },
    {
      "auxiliary_loss_clip": 0.01197527,
      "auxiliary_loss_mlp": 0.01028349,
      "balance_loss_clip": 1.05549645,
      "balance_loss_mlp": 1.01909256,
      "epoch": 0.330427463476222,
      "flos": 17749424148480.0,
      "grad_norm": 1.9788270127446614,
      "language_loss": 0.78455234,
      "learning_rate": 3.1257511493571797e-06,
      "loss": 0.80681109,
      "num_input_tokens_seen": 59187675,
      "step": 2748,
      "time_per_iteration": 2.6252334117889404
    },
    {
      "auxiliary_loss_clip": 0.01207456,
      "auxiliary_loss_mlp": 0.0103011,
      "balance_loss_clip": 1.05418825,
      "balance_loss_mlp": 1.02105594,
      "epoch": 0.33054770636686104,
      "flos": 27162580072320.0,
      "grad_norm": 1.8680033611064242,
      "language_loss": 0.7840749,
      "learning_rate": 3.125107211480552e-06,
      "loss": 0.80645055,
      "num_input_tokens_seen": 59207610,
      "step": 2749,
      "time_per_iteration": 2.7705278396606445
    },
    {
      "auxiliary_loss_clip": 0.01204009,
      "auxiliary_loss_mlp": 0.01031538,
      "balance_loss_clip": 1.04825091,
      "balance_loss_mlp": 1.02255583,
      "epoch": 0.33066794925750015,
      "flos": 20117193799680.0,
      "grad_norm": 1.7107025853734192,
      "language_loss": 0.80081797,
      "learning_rate": 3.124463102926054e-06,
      "loss": 0.8231734,
      "num_input_tokens_seen": 59226945,
      "step": 2750,
      "time_per_iteration": 2.7192726135253906
    },
    {
      "auxiliary_loss_clip": 0.01110273,
      "auxiliary_loss_mlp": 0.01004565,
      "balance_loss_clip": 1.03755665,
      "balance_loss_mlp": 1.0035342,
      "epoch": 0.33078819214813926,
      "flos": 70642609718400.0,
      "grad_norm": 0.7720941356996767,
      "language_loss": 0.61596775,
      "learning_rate": 3.1238188237913984e-06,
      "loss": 0.63711613,
      "num_input_tokens_seen": 59291485,
      "step": 2751,
      "time_per_iteration": 3.284609317779541
    },
    {
      "auxiliary_loss_clip": 0.01202471,
      "auxiliary_loss_mlp": 0.01033087,
      "balance_loss_clip": 1.05952501,
      "balance_loss_mlp": 1.02369893,
      "epoch": 0.3309084350387783,
      "flos": 21141940907520.0,
      "grad_norm": 1.971914437021134,
      "language_loss": 0.76652861,
      "learning_rate": 3.1231743741743202e-06,
      "loss": 0.78888428,
      "num_input_tokens_seen": 59310990,
      "step": 2752,
      "time_per_iteration": 2.619908094406128
    },
    {
      "auxiliary_loss_clip": 0.01194343,
      "auxiliary_loss_mlp": 0.010318,
      "balance_loss_clip": 1.05315161,
      "balance_loss_mlp": 1.02305627,
      "epoch": 0.3310286779294174,
      "flos": 14209350318720.0,
      "grad_norm": 3.0440329667049992,
      "language_loss": 0.83872646,
      "learning_rate": 3.122529754172582e-06,
      "loss": 0.8609879,
      "num_input_tokens_seen": 59327875,
      "step": 2753,
      "time_per_iteration": 2.5967957973480225
    },
    {
      "auxiliary_loss_clip": 0.01200462,
      "auxiliary_loss_mlp": 0.01033994,
      "balance_loss_clip": 1.05698442,
      "balance_loss_mlp": 1.02482092,
      "epoch": 0.33114892082005654,
      "flos": 20778130005120.0,
      "grad_norm": 3.0049798356740136,
      "language_loss": 0.72967625,
      "learning_rate": 3.1218849638839736e-06,
      "loss": 0.75202078,
      "num_input_tokens_seen": 59347135,
      "step": 2754,
      "time_per_iteration": 2.712536334991455
    },
    {
      "auxiliary_loss_clip": 0.01191271,
      "auxiliary_loss_mlp": 0.01035217,
      "balance_loss_clip": 1.04639566,
      "balance_loss_mlp": 1.02519166,
      "epoch": 0.3312691637106956,
      "flos": 17090750499840.0,
      "grad_norm": 3.446581974741475,
      "language_loss": 0.78671217,
      "learning_rate": 3.121240003406307e-06,
      "loss": 0.80897707,
      "num_input_tokens_seen": 59365985,
      "step": 2755,
      "time_per_iteration": 2.6416115760803223
    },
    {
      "auxiliary_loss_clip": 0.01208365,
      "auxiliary_loss_mlp": 0.01028252,
      "balance_loss_clip": 1.05555797,
      "balance_loss_mlp": 1.01853609,
      "epoch": 0.3313894066013347,
      "flos": 29456230008960.0,
      "grad_norm": 1.9299294274306604,
      "language_loss": 0.73041272,
      "learning_rate": 3.120594872837425e-06,
      "loss": 0.75277889,
      "num_input_tokens_seen": 59384655,
      "step": 2756,
      "time_per_iteration": 2.774571657180786
    },
    {
      "auxiliary_loss_clip": 0.01113359,
      "auxiliary_loss_mlp": 0.01200438,
      "balance_loss_clip": 1.03799534,
      "balance_loss_mlp": 1.00026536,
      "epoch": 0.3315096494919738,
      "flos": 61419242280960.0,
      "grad_norm": 0.8187382097934185,
      "language_loss": 0.62299275,
      "learning_rate": 3.1199495722751906e-06,
      "loss": 0.6461308,
      "num_input_tokens_seen": 59444185,
      "step": 2757,
      "time_per_iteration": 3.2244162559509277
    },
    {
      "auxiliary_loss_clip": 0.01206691,
      "auxiliary_loss_mlp": 0.01035014,
      "balance_loss_clip": 1.04989886,
      "balance_loss_mlp": 1.02599549,
      "epoch": 0.33162989238261287,
      "flos": 21653057485440.0,
      "grad_norm": 1.6318062956208732,
      "language_loss": 0.84017909,
      "learning_rate": 3.1193041018174972e-06,
      "loss": 0.86259615,
      "num_input_tokens_seen": 59464900,
      "step": 2758,
      "time_per_iteration": 2.698575258255005
    },
    {
      "auxiliary_loss_clip": 0.01203991,
      "auxiliary_loss_mlp": 0.01025506,
      "balance_loss_clip": 1.05852675,
      "balance_loss_mlp": 1.01604056,
      "epoch": 0.331750135273252,
      "flos": 22674787850880.0,
      "grad_norm": 2.068069659829069,
      "language_loss": 0.95221013,
      "learning_rate": 3.118658461562261e-06,
      "loss": 0.97450519,
      "num_input_tokens_seen": 59481000,
      "step": 2759,
      "time_per_iteration": 2.6699469089508057
    },
    {
      "auxiliary_loss_clip": 0.01201023,
      "auxiliary_loss_mlp": 0.01029644,
      "balance_loss_clip": 1.05652356,
      "balance_loss_mlp": 1.01947522,
      "epoch": 0.33187037816389103,
      "flos": 22746896403840.0,
      "grad_norm": 1.4306722704397647,
      "language_loss": 0.85160983,
      "learning_rate": 3.118012651607426e-06,
      "loss": 0.87391651,
      "num_input_tokens_seen": 59502605,
      "step": 2760,
      "time_per_iteration": 2.7019259929656982
    },
    {
      "auxiliary_loss_clip": 0.01197626,
      "auxiliary_loss_mlp": 0.01034162,
      "balance_loss_clip": 1.05808628,
      "balance_loss_mlp": 1.02401137,
      "epoch": 0.33199062105453014,
      "flos": 19203769918080.0,
      "grad_norm": 2.1377694800400207,
      "language_loss": 0.83625299,
      "learning_rate": 3.1173666720509603e-06,
      "loss": 0.85857093,
      "num_input_tokens_seen": 59519540,
      "step": 2761,
      "time_per_iteration": 2.61018967628479
    },
    {
      "auxiliary_loss_clip": 0.01204652,
      "auxiliary_loss_mlp": 0.01029497,
      "balance_loss_clip": 1.0534544,
      "balance_loss_mlp": 1.02042508,
      "epoch": 0.33211086394516925,
      "flos": 31577006764800.0,
      "grad_norm": 1.8088782413164195,
      "language_loss": 0.68284154,
      "learning_rate": 3.116720522990859e-06,
      "loss": 0.70518303,
      "num_input_tokens_seen": 59540415,
      "step": 2762,
      "time_per_iteration": 2.6600735187530518
    },
    {
      "auxiliary_loss_clip": 0.01208257,
      "auxiliary_loss_mlp": 0.01034208,
      "balance_loss_clip": 1.05037546,
      "balance_loss_mlp": 1.02502942,
      "epoch": 0.3322311068358083,
      "flos": 17932496791680.0,
      "grad_norm": 2.4756878056359746,
      "language_loss": 0.62085962,
      "learning_rate": 3.116074204525142e-06,
      "loss": 0.6432842,
      "num_input_tokens_seen": 59558590,
      "step": 2763,
      "time_per_iteration": 2.7061994075775146
    },
    {
      "auxiliary_loss_clip": 0.01189696,
      "auxiliary_loss_mlp": 0.01026011,
      "balance_loss_clip": 1.05458784,
      "balance_loss_mlp": 1.01753473,
      "epoch": 0.3323513497264474,
      "flos": 32269831269120.0,
      "grad_norm": 1.8427250200866834,
      "language_loss": 0.83733577,
      "learning_rate": 3.1154277167518553e-06,
      "loss": 0.85949284,
      "num_input_tokens_seen": 59580205,
      "step": 2764,
      "time_per_iteration": 3.6499407291412354
    },
    {
      "auxiliary_loss_clip": 0.01108033,
      "auxiliary_loss_mlp": 0.01001099,
      "balance_loss_clip": 1.03069997,
      "balance_loss_mlp": 1.00005555,
      "epoch": 0.33247159261708653,
      "flos": 52668674588160.0,
      "grad_norm": 0.7796124305461043,
      "language_loss": 0.59455168,
      "learning_rate": 3.114781059769072e-06,
      "loss": 0.61564296,
      "num_input_tokens_seen": 59631530,
      "step": 2765,
      "time_per_iteration": 3.899827718734741
    },
    {
      "auxiliary_loss_clip": 0.01199387,
      "auxiliary_loss_mlp": 0.01030553,
      "balance_loss_clip": 1.05355036,
      "balance_loss_mlp": 1.02121878,
      "epoch": 0.3325918355077256,
      "flos": 27125232906240.0,
      "grad_norm": 2.6284698900927577,
      "language_loss": 0.67442364,
      "learning_rate": 3.1141342336748874e-06,
      "loss": 0.69672298,
      "num_input_tokens_seen": 59651090,
      "step": 2766,
      "time_per_iteration": 2.70115065574646
    },
    {
      "auxiliary_loss_clip": 0.0119515,
      "auxiliary_loss_mlp": 0.01032526,
      "balance_loss_clip": 1.05529368,
      "balance_loss_mlp": 1.02360368,
      "epoch": 0.3327120783983647,
      "flos": 23664414435840.0,
      "grad_norm": 2.1656378054998933,
      "language_loss": 0.82251638,
      "learning_rate": 3.1134872385674253e-06,
      "loss": 0.84479314,
      "num_input_tokens_seen": 59675245,
      "step": 2767,
      "time_per_iteration": 2.7224695682525635
    },
    {
      "auxiliary_loss_clip": 0.01204051,
      "auxiliary_loss_mlp": 0.01027382,
      "balance_loss_clip": 1.05202651,
      "balance_loss_mlp": 1.01791084,
      "epoch": 0.3328323212890038,
      "flos": 19171378828800.0,
      "grad_norm": 1.7415911026075632,
      "language_loss": 0.85863787,
      "learning_rate": 3.1128400745448353e-06,
      "loss": 0.88095224,
      "num_input_tokens_seen": 59694625,
      "step": 2768,
      "time_per_iteration": 2.694366216659546
    },
    {
      "auxiliary_loss_clip": 0.01198326,
      "auxiliary_loss_mlp": 0.01030022,
      "balance_loss_clip": 1.05591464,
      "balance_loss_mlp": 1.02127767,
      "epoch": 0.33295256417964286,
      "flos": 37706347463040.0,
      "grad_norm": 2.099256452889705,
      "language_loss": 0.62909591,
      "learning_rate": 3.11219274170529e-06,
      "loss": 0.65137935,
      "num_input_tokens_seen": 59716435,
      "step": 2769,
      "time_per_iteration": 3.7759530544281006
    },
    {
      "auxiliary_loss_clip": 0.01192023,
      "auxiliary_loss_mlp": 0.01036394,
      "balance_loss_clip": 1.05172563,
      "balance_loss_mlp": 1.02822244,
      "epoch": 0.333072807070282,
      "flos": 26505989412480.0,
      "grad_norm": 1.8832715226045842,
      "language_loss": 0.81923509,
      "learning_rate": 3.1115452401469903e-06,
      "loss": 0.84151924,
      "num_input_tokens_seen": 59736835,
      "step": 2770,
      "time_per_iteration": 2.6387710571289062
    },
    {
      "auxiliary_loss_clip": 0.01194871,
      "auxiliary_loss_mlp": 0.01028402,
      "balance_loss_clip": 1.04706347,
      "balance_loss_mlp": 1.01950336,
      "epoch": 0.3331930499609211,
      "flos": 21430913823360.0,
      "grad_norm": 2.007984718176959,
      "language_loss": 0.86535984,
      "learning_rate": 3.1108975699681613e-06,
      "loss": 0.88759255,
      "num_input_tokens_seen": 59754230,
      "step": 2771,
      "time_per_iteration": 2.6947388648986816
    },
    {
      "auxiliary_loss_clip": 0.01199856,
      "auxiliary_loss_mlp": 0.0103397,
      "balance_loss_clip": 1.05212975,
      "balance_loss_mlp": 1.02532148,
      "epoch": 0.33331329285156014,
      "flos": 20659947281280.0,
      "grad_norm": 1.78274558320932,
      "language_loss": 0.71565866,
      "learning_rate": 3.1102497312670542e-06,
      "loss": 0.73799694,
      "num_input_tokens_seen": 59772235,
      "step": 2772,
      "time_per_iteration": 3.5372722148895264
    },
    {
      "auxiliary_loss_clip": 0.01189486,
      "auxiliary_loss_mlp": 0.01032463,
      "balance_loss_clip": 1.05248308,
      "balance_loss_mlp": 1.02324176,
      "epoch": 0.33343353574219925,
      "flos": 28001596930560.0,
      "grad_norm": 1.7728078792418847,
      "language_loss": 0.80354166,
      "learning_rate": 3.109601724141946e-06,
      "loss": 0.8257612,
      "num_input_tokens_seen": 59791230,
      "step": 2773,
      "time_per_iteration": 2.6858956813812256
    },
    {
      "auxiliary_loss_clip": 0.01197374,
      "auxiliary_loss_mlp": 0.01025718,
      "balance_loss_clip": 1.05480504,
      "balance_loss_mlp": 1.01705194,
      "epoch": 0.33355377863283836,
      "flos": 23764963582080.0,
      "grad_norm": 2.002750495225092,
      "language_loss": 0.68229765,
      "learning_rate": 3.108953548691138e-06,
      "loss": 0.70452857,
      "num_input_tokens_seen": 59811315,
      "step": 2774,
      "time_per_iteration": 2.6629061698913574
    },
    {
      "auxiliary_loss_clip": 0.01198108,
      "auxiliary_loss_mlp": 0.01030128,
      "balance_loss_clip": 1.05766654,
      "balance_loss_mlp": 1.02085304,
      "epoch": 0.3336740215234774,
      "flos": 37779677078400.0,
      "grad_norm": 2.3767589293856854,
      "language_loss": 0.72763813,
      "learning_rate": 3.108305205012959e-06,
      "loss": 0.74992049,
      "num_input_tokens_seen": 59832010,
      "step": 2775,
      "time_per_iteration": 2.7817747592926025
    },
    {
      "auxiliary_loss_clip": 0.01196539,
      "auxiliary_loss_mlp": 0.01023966,
      "balance_loss_clip": 1.05287027,
      "balance_loss_mlp": 1.0148226,
      "epoch": 0.3337942644141165,
      "flos": 25519056347520.0,
      "grad_norm": 2.7513251458801693,
      "language_loss": 0.8778311,
      "learning_rate": 3.107656693205761e-06,
      "loss": 0.90003616,
      "num_input_tokens_seen": 59851450,
      "step": 2776,
      "time_per_iteration": 2.7686376571655273
    },
    {
      "auxiliary_loss_clip": 0.01200603,
      "auxiliary_loss_mlp": 0.01029785,
      "balance_loss_clip": 1.05804038,
      "balance_loss_mlp": 1.01991463,
      "epoch": 0.3339145073047556,
      "flos": 25989844930560.0,
      "grad_norm": 2.0590184603712207,
      "language_loss": 0.70402062,
      "learning_rate": 3.107008013367924e-06,
      "loss": 0.72632456,
      "num_input_tokens_seen": 59870245,
      "step": 2777,
      "time_per_iteration": 2.639850616455078
    },
    {
      "auxiliary_loss_clip": 0.01200677,
      "auxiliary_loss_mlp": 0.01028496,
      "balance_loss_clip": 1.0504446,
      "balance_loss_mlp": 1.01899505,
      "epoch": 0.3340347501953947,
      "flos": 19062569554560.0,
      "grad_norm": 2.395090381256645,
      "language_loss": 0.86896062,
      "learning_rate": 3.1063591655978507e-06,
      "loss": 0.8912524,
      "num_input_tokens_seen": 59886195,
      "step": 2778,
      "time_per_iteration": 2.6783621311187744
    },
    {
      "auxiliary_loss_clip": 0.01192701,
      "auxiliary_loss_mlp": 0.01027626,
      "balance_loss_clip": 1.04664993,
      "balance_loss_mlp": 1.01835179,
      "epoch": 0.3341549930860338,
      "flos": 18109715518080.0,
      "grad_norm": 2.071391192164593,
      "language_loss": 0.80260235,
      "learning_rate": 3.105710149993972e-06,
      "loss": 0.82480562,
      "num_input_tokens_seen": 59905525,
      "step": 2779,
      "time_per_iteration": 2.806314468383789
    },
    {
      "auxiliary_loss_clip": 0.01197214,
      "auxiliary_loss_mlp": 0.01026054,
      "balance_loss_clip": 1.05614305,
      "balance_loss_mlp": 1.01696467,
      "epoch": 0.33427523597667286,
      "flos": 22674967418880.0,
      "grad_norm": 1.8740428893014496,
      "language_loss": 0.85342151,
      "learning_rate": 3.1050609666547427e-06,
      "loss": 0.87565422,
      "num_input_tokens_seen": 59925085,
      "step": 2780,
      "time_per_iteration": 2.6399359703063965
    },
    {
      "auxiliary_loss_clip": 0.01210674,
      "auxiliary_loss_mlp": 0.01036205,
      "balance_loss_clip": 1.05532682,
      "balance_loss_mlp": 1.02754474,
      "epoch": 0.33439547886731197,
      "flos": 22638338524800.0,
      "grad_norm": 1.959026028165239,
      "language_loss": 0.77327949,
      "learning_rate": 3.104411615678644e-06,
      "loss": 0.79574835,
      "num_input_tokens_seen": 59943935,
      "step": 2781,
      "time_per_iteration": 2.6496100425720215
    },
    {
      "auxiliary_loss_clip": 0.01195743,
      "auxiliary_loss_mlp": 0.01029458,
      "balance_loss_clip": 1.053527,
      "balance_loss_mlp": 1.01972508,
      "epoch": 0.3345157217579511,
      "flos": 24096383395200.0,
      "grad_norm": 2.5973344917697645,
      "language_loss": 0.73247296,
      "learning_rate": 3.1037620971641803e-06,
      "loss": 0.75472498,
      "num_input_tokens_seen": 59963725,
      "step": 2782,
      "time_per_iteration": 2.901736259460449
    },
    {
      "auxiliary_loss_clip": 0.01199561,
      "auxiliary_loss_mlp": 0.01032648,
      "balance_loss_clip": 1.05895317,
      "balance_loss_mlp": 1.023242,
      "epoch": 0.33463596464859013,
      "flos": 18989491334400.0,
      "grad_norm": 2.807285223269154,
      "language_loss": 0.65120864,
      "learning_rate": 3.1031124112098844e-06,
      "loss": 0.6735307,
      "num_input_tokens_seen": 59981935,
      "step": 2783,
      "time_per_iteration": 2.643108606338501
    },
    {
      "auxiliary_loss_clip": 0.01202996,
      "auxiliary_loss_mlp": 0.0102659,
      "balance_loss_clip": 1.05564213,
      "balance_loss_mlp": 1.01781058,
      "epoch": 0.33475620753922924,
      "flos": 20375607219840.0,
      "grad_norm": 1.94268465627646,
      "language_loss": 0.72393709,
      "learning_rate": 3.1024625579143127e-06,
      "loss": 0.74623299,
      "num_input_tokens_seen": 59999455,
      "step": 2784,
      "time_per_iteration": 2.666144609451294
    },
    {
      "auxiliary_loss_clip": 0.01195219,
      "auxiliary_loss_mlp": 0.01030561,
      "balance_loss_clip": 1.05637372,
      "balance_loss_mlp": 1.02121496,
      "epoch": 0.33487645042986836,
      "flos": 18182578256640.0,
      "grad_norm": 1.8304635242201697,
      "language_loss": 0.72912335,
      "learning_rate": 3.101812537376048e-06,
      "loss": 0.75138116,
      "num_input_tokens_seen": 60018475,
      "step": 2785,
      "time_per_iteration": 2.623419761657715
    },
    {
      "auxiliary_loss_clip": 0.01194264,
      "auxiliary_loss_mlp": 0.01208811,
      "balance_loss_clip": 1.0531466,
      "balance_loss_mlp": 1.0007565,
      "epoch": 0.3349966933205074,
      "flos": 25848824135040.0,
      "grad_norm": 4.3629749817273025,
      "language_loss": 0.84216607,
      "learning_rate": 3.1011623496936973e-06,
      "loss": 0.86619675,
      "num_input_tokens_seen": 60036770,
      "step": 2786,
      "time_per_iteration": 2.7205846309661865
    },
    {
      "auxiliary_loss_clip": 0.01195529,
      "auxiliary_loss_mlp": 0.01027004,
      "balance_loss_clip": 1.05794716,
      "balance_loss_mlp": 1.01809323,
      "epoch": 0.3351169362111465,
      "flos": 28111447699200.0,
      "grad_norm": 1.9999565160718922,
      "language_loss": 0.69634777,
      "learning_rate": 3.100511994965893e-06,
      "loss": 0.71857309,
      "num_input_tokens_seen": 60056725,
      "step": 2787,
      "time_per_iteration": 2.738480567932129
    },
    {
      "auxiliary_loss_clip": 0.01192038,
      "auxiliary_loss_mlp": 0.01030234,
      "balance_loss_clip": 1.05531502,
      "balance_loss_mlp": 1.02144802,
      "epoch": 0.33523717910178563,
      "flos": 22673315393280.0,
      "grad_norm": 18.994111475338922,
      "language_loss": 0.84721375,
      "learning_rate": 3.0998614732912947e-06,
      "loss": 0.86943644,
      "num_input_tokens_seen": 60076100,
      "step": 2788,
      "time_per_iteration": 2.662501335144043
    },
    {
      "auxiliary_loss_clip": 0.01199653,
      "auxiliary_loss_mlp": 0.01032012,
      "balance_loss_clip": 1.05815148,
      "balance_loss_mlp": 1.02298141,
      "epoch": 0.3353574219924247,
      "flos": 15669801400320.0,
      "grad_norm": 2.2146025286470485,
      "language_loss": 0.68188733,
      "learning_rate": 3.0992107847685855e-06,
      "loss": 0.70420396,
      "num_input_tokens_seen": 60093815,
      "step": 2789,
      "time_per_iteration": 2.7105348110198975
    },
    {
      "auxiliary_loss_clip": 0.01204989,
      "auxiliary_loss_mlp": 0.01034075,
      "balance_loss_clip": 1.05853927,
      "balance_loss_mlp": 1.02468169,
      "epoch": 0.3354776648830638,
      "flos": 24790644443520.0,
      "grad_norm": 2.2143368587674037,
      "language_loss": 0.79158592,
      "learning_rate": 3.0985599294964736e-06,
      "loss": 0.81397665,
      "num_input_tokens_seen": 60113370,
      "step": 2790,
      "time_per_iteration": 2.7021102905273438
    },
    {
      "auxiliary_loss_clip": 0.01213817,
      "auxiliary_loss_mlp": 0.01035371,
      "balance_loss_clip": 1.05402291,
      "balance_loss_mlp": 1.02538729,
      "epoch": 0.33559790777370285,
      "flos": 28694852398080.0,
      "grad_norm": 2.1909988793420303,
      "language_loss": 0.7034179,
      "learning_rate": 3.097908907573695e-06,
      "loss": 0.72590971,
      "num_input_tokens_seen": 60131350,
      "step": 2791,
      "time_per_iteration": 3.692608594894409
    },
    {
      "auxiliary_loss_clip": 0.01206645,
      "auxiliary_loss_mlp": 0.01030071,
      "balance_loss_clip": 1.05204296,
      "balance_loss_mlp": 1.02153611,
      "epoch": 0.33571815066434196,
      "flos": 22235779825920.0,
      "grad_norm": 2.6141784039587384,
      "language_loss": 0.89998698,
      "learning_rate": 3.0972577190990067e-06,
      "loss": 0.92235422,
      "num_input_tokens_seen": 60149830,
      "step": 2792,
      "time_per_iteration": 3.8356659412384033
    },
    {
      "auxiliary_loss_clip": 0.01207376,
      "auxiliary_loss_mlp": 0.01031869,
      "balance_loss_clip": 1.05363381,
      "balance_loss_mlp": 1.02281523,
      "epoch": 0.3358383935549811,
      "flos": 23842279607040.0,
      "grad_norm": 1.9864973472936065,
      "language_loss": 0.80151844,
      "learning_rate": 3.096606364171196e-06,
      "loss": 0.82391089,
      "num_input_tokens_seen": 60169620,
      "step": 2793,
      "time_per_iteration": 2.6508266925811768
    },
    {
      "auxiliary_loss_clip": 0.01187269,
      "auxiliary_loss_mlp": 0.0102998,
      "balance_loss_clip": 1.04915452,
      "balance_loss_mlp": 1.02084827,
      "epoch": 0.33595863644562013,
      "flos": 22267308988800.0,
      "grad_norm": 1.796376745614975,
      "language_loss": 0.84970176,
      "learning_rate": 3.0959548428890703e-06,
      "loss": 0.87187421,
      "num_input_tokens_seen": 60188490,
      "step": 2794,
      "time_per_iteration": 2.721787214279175
    },
    {
      "auxiliary_loss_clip": 0.01198029,
      "auxiliary_loss_mlp": 0.01032012,
      "balance_loss_clip": 1.05911422,
      "balance_loss_mlp": 1.02274919,
      "epoch": 0.33607887933625924,
      "flos": 20119779578880.0,
      "grad_norm": 1.6098041721485719,
      "language_loss": 0.84383458,
      "learning_rate": 3.095303155351468e-06,
      "loss": 0.866135,
      "num_input_tokens_seen": 60208695,
      "step": 2795,
      "time_per_iteration": 2.6338706016540527
    },
    {
      "auxiliary_loss_clip": 0.01197675,
      "auxiliary_loss_mlp": 0.01028027,
      "balance_loss_clip": 1.05235338,
      "balance_loss_mlp": 1.01924109,
      "epoch": 0.33619912222689835,
      "flos": 19318109886720.0,
      "grad_norm": 2.1840029258069107,
      "language_loss": 0.797297,
      "learning_rate": 3.0946513016572464e-06,
      "loss": 0.81955397,
      "num_input_tokens_seen": 60227600,
      "step": 2796,
      "time_per_iteration": 3.6519603729248047
    },
    {
      "auxiliary_loss_clip": 0.0120287,
      "auxiliary_loss_mlp": 0.01032961,
      "balance_loss_clip": 1.05662417,
      "balance_loss_mlp": 1.02361453,
      "epoch": 0.3363193651175374,
      "flos": 16800664262400.0,
      "grad_norm": 5.628821644481781,
      "language_loss": 0.77081823,
      "learning_rate": 3.0939992819052938e-06,
      "loss": 0.79317653,
      "num_input_tokens_seen": 60245110,
      "step": 2797,
      "time_per_iteration": 2.6158604621887207
    },
    {
      "auxiliary_loss_clip": 0.01201719,
      "auxiliary_loss_mlp": 0.01026824,
      "balance_loss_clip": 1.05514598,
      "balance_loss_mlp": 1.01761532,
      "epoch": 0.3364396080081765,
      "flos": 23550289948800.0,
      "grad_norm": 2.1933498088671604,
      "language_loss": 0.81258857,
      "learning_rate": 3.0933470961945193e-06,
      "loss": 0.83487397,
      "num_input_tokens_seen": 60263405,
      "step": 2798,
      "time_per_iteration": 2.7127861976623535
    },
    {
      "auxiliary_loss_clip": 0.01199262,
      "auxiliary_loss_mlp": 0.01032274,
      "balance_loss_clip": 1.05828023,
      "balance_loss_mlp": 1.02370918,
      "epoch": 0.3365598508988156,
      "flos": 28037902602240.0,
      "grad_norm": 2.158613820697869,
      "language_loss": 0.6822449,
      "learning_rate": 3.0926947446238597e-06,
      "loss": 0.70456022,
      "num_input_tokens_seen": 60282975,
      "step": 2799,
      "time_per_iteration": 3.6414830684661865
    },
    {
      "auxiliary_loss_clip": 0.01203221,
      "auxiliary_loss_mlp": 0.01030119,
      "balance_loss_clip": 1.05431962,
      "balance_loss_mlp": 1.02040911,
      "epoch": 0.3366800937894547,
      "flos": 16982767238400.0,
      "grad_norm": 2.6897402847624066,
      "language_loss": 0.82897151,
      "learning_rate": 3.092042227292276e-06,
      "loss": 0.85130489,
      "num_input_tokens_seen": 60299810,
      "step": 2800,
      "time_per_iteration": 2.6313235759735107
    },
    {
      "auxiliary_loss_clip": 0.01192395,
      "auxiliary_loss_mlp": 0.01025942,
      "balance_loss_clip": 1.05733752,
      "balance_loss_mlp": 1.01777649,
      "epoch": 0.3368003366800938,
      "flos": 23915321913600.0,
      "grad_norm": 1.5766106769944241,
      "language_loss": 0.88371623,
      "learning_rate": 3.0913895442987557e-06,
      "loss": 0.90589958,
      "num_input_tokens_seen": 60320775,
      "step": 2801,
      "time_per_iteration": 2.7195863723754883
    },
    {
      "auxiliary_loss_clip": 0.01204052,
      "auxiliary_loss_mlp": 0.01209001,
      "balance_loss_clip": 1.05585885,
      "balance_loss_mlp": 1.00077724,
      "epoch": 0.3369205795707329,
      "flos": 24791219061120.0,
      "grad_norm": 2.1025335553229194,
      "language_loss": 0.86295736,
      "learning_rate": 3.090736695742308e-06,
      "loss": 0.88708794,
      "num_input_tokens_seen": 60341905,
      "step": 2802,
      "time_per_iteration": 2.6697747707366943
    },
    {
      "auxiliary_loss_clip": 0.01196464,
      "auxiliary_loss_mlp": 0.01027989,
      "balance_loss_clip": 1.04859567,
      "balance_loss_mlp": 1.01954865,
      "epoch": 0.33704082246137196,
      "flos": 17931096161280.0,
      "grad_norm": 2.913458119194158,
      "language_loss": 0.52540827,
      "learning_rate": 3.0900836817219713e-06,
      "loss": 0.54765278,
      "num_input_tokens_seen": 60358335,
      "step": 2803,
      "time_per_iteration": 2.721740484237671
    },
    {
      "auxiliary_loss_clip": 0.01195171,
      "auxiliary_loss_mlp": 0.01026071,
      "balance_loss_clip": 1.05684412,
      "balance_loss_mlp": 1.01783991,
      "epoch": 0.33716106535201107,
      "flos": 21286517149440.0,
      "grad_norm": 1.7403502473688102,
      "language_loss": 0.84027243,
      "learning_rate": 3.089430502336807e-06,
      "loss": 0.86248481,
      "num_input_tokens_seen": 60378305,
      "step": 2804,
      "time_per_iteration": 2.597853660583496
    },
    {
      "auxiliary_loss_clip": 0.01200119,
      "auxiliary_loss_mlp": 0.01028227,
      "balance_loss_clip": 1.05508697,
      "balance_loss_mlp": 1.01883292,
      "epoch": 0.3372813082426502,
      "flos": 18402962152320.0,
      "grad_norm": 2.5201703824452446,
      "language_loss": 0.90485764,
      "learning_rate": 3.088777157685902e-06,
      "loss": 0.92714101,
      "num_input_tokens_seen": 60393895,
      "step": 2805,
      "time_per_iteration": 2.607654094696045
    },
    {
      "auxiliary_loss_clip": 0.01194555,
      "auxiliary_loss_mlp": 0.01025965,
      "balance_loss_clip": 1.05421138,
      "balance_loss_mlp": 1.01737642,
      "epoch": 0.33740155113328923,
      "flos": 17201391367680.0,
      "grad_norm": 2.659423843434639,
      "language_loss": 0.85624361,
      "learning_rate": 3.088123647868367e-06,
      "loss": 0.87844884,
      "num_input_tokens_seen": 60410445,
      "step": 2806,
      "time_per_iteration": 2.6349432468414307
    },
    {
      "auxiliary_loss_clip": 0.01201942,
      "auxiliary_loss_mlp": 0.01028336,
      "balance_loss_clip": 1.05501628,
      "balance_loss_mlp": 1.01962197,
      "epoch": 0.33752179402392835,
      "flos": 29058950609280.0,
      "grad_norm": 1.8784052358915702,
      "language_loss": 0.81201208,
      "learning_rate": 3.0874699729833405e-06,
      "loss": 0.83431482,
      "num_input_tokens_seen": 60431815,
      "step": 2807,
      "time_per_iteration": 2.6977450847625732
    },
    {
      "auxiliary_loss_clip": 0.01196049,
      "auxiliary_loss_mlp": 0.01027011,
      "balance_loss_clip": 1.05415845,
      "balance_loss_mlp": 1.01784933,
      "epoch": 0.3376420369145674,
      "flos": 25080730680960.0,
      "grad_norm": 1.7262060568903226,
      "language_loss": 0.79658294,
      "learning_rate": 3.086816133129983e-06,
      "loss": 0.81881356,
      "num_input_tokens_seen": 60452075,
      "step": 2808,
      "time_per_iteration": 2.7214772701263428
    },
    {
      "auxiliary_loss_clip": 0.01196642,
      "auxiliary_loss_mlp": 0.01029086,
      "balance_loss_clip": 1.0597744,
      "balance_loss_mlp": 1.02003217,
      "epoch": 0.3377622798052065,
      "flos": 27490624007040.0,
      "grad_norm": 2.04921782454361,
      "language_loss": 0.75779188,
      "learning_rate": 3.0861621284074826e-06,
      "loss": 0.78004909,
      "num_input_tokens_seen": 60472600,
      "step": 2809,
      "time_per_iteration": 2.6255972385406494
    },
    {
      "auxiliary_loss_clip": 0.01208219,
      "auxiliary_loss_mlp": 0.01027367,
      "balance_loss_clip": 1.05718529,
      "balance_loss_mlp": 1.01917768,
      "epoch": 0.3378825226958456,
      "flos": 21975211589760.0,
      "grad_norm": 1.5861923154044864,
      "language_loss": 0.73323053,
      "learning_rate": 3.085507958915051e-06,
      "loss": 0.75558639,
      "num_input_tokens_seen": 60491030,
      "step": 2810,
      "time_per_iteration": 2.803650379180908
    },
    {
      "auxiliary_loss_clip": 0.01196597,
      "auxiliary_loss_mlp": 0.01030492,
      "balance_loss_clip": 1.05546713,
      "balance_loss_mlp": 1.02066302,
      "epoch": 0.3380027655864847,
      "flos": 42523189200000.0,
      "grad_norm": 1.9992173547605117,
      "language_loss": 0.71239901,
      "learning_rate": 3.084853624751925e-06,
      "loss": 0.73466992,
      "num_input_tokens_seen": 60512615,
      "step": 2811,
      "time_per_iteration": 2.8230223655700684
    },
    {
      "auxiliary_loss_clip": 0.0120643,
      "auxiliary_loss_mlp": 0.01029078,
      "balance_loss_clip": 1.05589247,
      "balance_loss_mlp": 1.02025688,
      "epoch": 0.3381230084771238,
      "flos": 26725080418560.0,
      "grad_norm": 1.8483928960870306,
      "language_loss": 0.85865033,
      "learning_rate": 3.0841991260173668e-06,
      "loss": 0.88100541,
      "num_input_tokens_seen": 60532520,
      "step": 2812,
      "time_per_iteration": 2.7303810119628906
    },
    {
      "auxiliary_loss_clip": 0.01198527,
      "auxiliary_loss_mlp": 0.01027098,
      "balance_loss_clip": 1.05862713,
      "balance_loss_mlp": 1.01754904,
      "epoch": 0.3382432513677629,
      "flos": 22710375250560.0,
      "grad_norm": 1.976746430901719,
      "language_loss": 0.80864191,
      "learning_rate": 3.0835444628106634e-06,
      "loss": 0.83089817,
      "num_input_tokens_seen": 60551500,
      "step": 2813,
      "time_per_iteration": 2.5913567543029785
    },
    {
      "auxiliary_loss_clip": 0.0119464,
      "auxiliary_loss_mlp": 0.01209449,
      "balance_loss_clip": 1.0570128,
      "balance_loss_mlp": 1.00090313,
      "epoch": 0.33836349425840195,
      "flos": 22122409524480.0,
      "grad_norm": 1.8133812749540164,
      "language_loss": 0.83257782,
      "learning_rate": 3.082889635231126e-06,
      "loss": 0.85661876,
      "num_input_tokens_seen": 60570160,
      "step": 2814,
      "time_per_iteration": 2.76240611076355
    },
    {
      "auxiliary_loss_clip": 0.0120363,
      "auxiliary_loss_mlp": 0.01027518,
      "balance_loss_clip": 1.05325925,
      "balance_loss_mlp": 1.01765382,
      "epoch": 0.33848373714904106,
      "flos": 27308090067840.0,
      "grad_norm": 2.9797303487422617,
      "language_loss": 0.76849318,
      "learning_rate": 3.0822346433780925e-06,
      "loss": 0.79080462,
      "num_input_tokens_seen": 60590885,
      "step": 2815,
      "time_per_iteration": 2.7154550552368164
    },
    {
      "auxiliary_loss_clip": 0.01198214,
      "auxiliary_loss_mlp": 0.01027307,
      "balance_loss_clip": 1.05406618,
      "balance_loss_mlp": 1.01767516,
      "epoch": 0.3386039800396802,
      "flos": 25848716394240.0,
      "grad_norm": 7.879661139620971,
      "language_loss": 0.8720597,
      "learning_rate": 3.0815794873509237e-06,
      "loss": 0.894315,
      "num_input_tokens_seen": 60609170,
      "step": 2816,
      "time_per_iteration": 2.742644786834717
    },
    {
      "auxiliary_loss_clip": 0.01195534,
      "auxiliary_loss_mlp": 0.01026369,
      "balance_loss_clip": 1.05724764,
      "balance_loss_mlp": 1.0172379,
      "epoch": 0.33872422293031923,
      "flos": 18880646146560.0,
      "grad_norm": 1.7961971060329183,
      "language_loss": 0.72914505,
      "learning_rate": 3.0809241672490066e-06,
      "loss": 0.75136411,
      "num_input_tokens_seen": 60627340,
      "step": 2817,
      "time_per_iteration": 3.568558931350708
    },
    {
      "auxiliary_loss_clip": 0.01203775,
      "auxiliary_loss_mlp": 0.01025972,
      "balance_loss_clip": 1.0566051,
      "balance_loss_mlp": 1.01695347,
      "epoch": 0.33884446582095834,
      "flos": 23146977064320.0,
      "grad_norm": 1.6464987729599143,
      "language_loss": 0.85147363,
      "learning_rate": 3.080268683171753e-06,
      "loss": 0.87377107,
      "num_input_tokens_seen": 60647630,
      "step": 2818,
      "time_per_iteration": 2.6406936645507812
    },
    {
      "auxiliary_loss_clip": 0.01196838,
      "auxiliary_loss_mlp": 0.01026093,
      "balance_loss_clip": 1.05370903,
      "balance_loss_mlp": 1.0171169,
      "epoch": 0.33896470871159745,
      "flos": 15997342544640.0,
      "grad_norm": 2.9156802421115384,
      "language_loss": 0.89448261,
      "learning_rate": 3.0796130352185985e-06,
      "loss": 0.91671193,
      "num_input_tokens_seen": 60664485,
      "step": 2819,
      "time_per_iteration": 3.5558383464813232
    },
    {
      "auxiliary_loss_clip": 0.01185746,
      "auxiliary_loss_mlp": 0.01209937,
      "balance_loss_clip": 1.0486201,
      "balance_loss_mlp": 1.00098288,
      "epoch": 0.3390849516022365,
      "flos": 34495754112000.0,
      "grad_norm": 1.9739053629714722,
      "language_loss": 0.66461194,
      "learning_rate": 3.0789572234890057e-06,
      "loss": 0.68856883,
      "num_input_tokens_seen": 60686125,
      "step": 2820,
      "time_per_iteration": 2.707578659057617
    },
    {
      "auxiliary_loss_clip": 0.01200461,
      "auxiliary_loss_mlp": 0.01031261,
      "balance_loss_clip": 1.05638385,
      "balance_loss_mlp": 1.02117586,
      "epoch": 0.3392051944928756,
      "flos": 16180307447040.0,
      "grad_norm": 1.7760063123074825,
      "language_loss": 0.77664632,
      "learning_rate": 3.0783012480824596e-06,
      "loss": 0.79896355,
      "num_input_tokens_seen": 60705270,
      "step": 2821,
      "time_per_iteration": 2.726325511932373
    },
    {
      "auxiliary_loss_clip": 0.01196453,
      "auxiliary_loss_mlp": 0.01036641,
      "balance_loss_clip": 1.05727565,
      "balance_loss_mlp": 1.02709866,
      "epoch": 0.33932543738351467,
      "flos": 17086656349440.0,
      "grad_norm": 10.310547276835992,
      "language_loss": 0.74812746,
      "learning_rate": 3.077645109098471e-06,
      "loss": 0.7704584,
      "num_input_tokens_seen": 60721540,
      "step": 2822,
      "time_per_iteration": 3.576075792312622
    },
    {
      "auxiliary_loss_clip": 0.01189663,
      "auxiliary_loss_mlp": 0.01032978,
      "balance_loss_clip": 1.05145633,
      "balance_loss_mlp": 1.02389455,
      "epoch": 0.3394456802741538,
      "flos": 22126970551680.0,
      "grad_norm": 2.1275610137321834,
      "language_loss": 0.72646916,
      "learning_rate": 3.076988806636577e-06,
      "loss": 0.74869555,
      "num_input_tokens_seen": 60739300,
      "step": 2823,
      "time_per_iteration": 2.6904971599578857
    },
    {
      "auxiliary_loss_clip": 0.01205336,
      "auxiliary_loss_mlp": 0.01209332,
      "balance_loss_clip": 1.05704856,
      "balance_loss_mlp": 1.00082862,
      "epoch": 0.3395659231647929,
      "flos": 25226887121280.0,
      "grad_norm": 3.001191370103993,
      "language_loss": 0.89016628,
      "learning_rate": 3.0763323407963377e-06,
      "loss": 0.91431296,
      "num_input_tokens_seen": 60758910,
      "step": 2824,
      "time_per_iteration": 2.722111225128174
    },
    {
      "auxiliary_loss_clip": 0.01198868,
      "auxiliary_loss_mlp": 0.01030747,
      "balance_loss_clip": 1.05486488,
      "balance_loss_mlp": 1.0223012,
      "epoch": 0.33968616605543195,
      "flos": 29096477343360.0,
      "grad_norm": 1.997280125373871,
      "language_loss": 0.80263233,
      "learning_rate": 3.075675711677337e-06,
      "loss": 0.82492852,
      "num_input_tokens_seen": 60779005,
      "step": 2825,
      "time_per_iteration": 2.7077345848083496
    },
    {
      "auxiliary_loss_clip": 0.01194579,
      "auxiliary_loss_mlp": 0.01031092,
      "balance_loss_clip": 1.05454838,
      "balance_loss_mlp": 1.0215199,
      "epoch": 0.33980640894607106,
      "flos": 21433966479360.0,
      "grad_norm": 2.7235793270402544,
      "language_loss": 0.78343874,
      "learning_rate": 3.0750189193791865e-06,
      "loss": 0.80569541,
      "num_input_tokens_seen": 60798590,
      "step": 2826,
      "time_per_iteration": 3.5873398780822754
    },
    {
      "auxiliary_loss_clip": 0.01194405,
      "auxiliary_loss_mlp": 0.01028217,
      "balance_loss_clip": 1.05387151,
      "balance_loss_mlp": 1.0186801,
      "epoch": 0.33992665183671017,
      "flos": 32490035596800.0,
      "grad_norm": 1.8892694834993056,
      "language_loss": 0.70557535,
      "learning_rate": 3.0743619640015203e-06,
      "loss": 0.72780156,
      "num_input_tokens_seen": 60818840,
      "step": 2827,
      "time_per_iteration": 2.812485694885254
    },
    {
      "auxiliary_loss_clip": 0.01205059,
      "auxiliary_loss_mlp": 0.01028663,
      "balance_loss_clip": 1.05302799,
      "balance_loss_mlp": 1.01926327,
      "epoch": 0.3400468947273492,
      "flos": 17055414495360.0,
      "grad_norm": 2.0977435680746837,
      "language_loss": 0.92660278,
      "learning_rate": 3.073704845643999e-06,
      "loss": 0.94893998,
      "num_input_tokens_seen": 60835965,
      "step": 2828,
      "time_per_iteration": 2.649986982345581
    },
    {
      "auxiliary_loss_clip": 0.01200264,
      "auxiliary_loss_mlp": 0.01037156,
      "balance_loss_clip": 1.05440807,
      "balance_loss_mlp": 1.02729702,
      "epoch": 0.34016713761798834,
      "flos": 16872988296960.0,
      "grad_norm": 2.676175133157097,
      "language_loss": 0.78171074,
      "learning_rate": 3.0730475644063063e-06,
      "loss": 0.80408496,
      "num_input_tokens_seen": 60851065,
      "step": 2829,
      "time_per_iteration": 2.672074556350708
    },
    {
      "auxiliary_loss_clip": 0.01192438,
      "auxiliary_loss_mlp": 0.01208487,
      "balance_loss_clip": 1.05059958,
      "balance_loss_mlp": 1.00090563,
      "epoch": 0.34028738050862745,
      "flos": 21907161273600.0,
      "grad_norm": 1.7218621212284444,
      "language_loss": 0.65202487,
      "learning_rate": 3.072390120388151e-06,
      "loss": 0.67603415,
      "num_input_tokens_seen": 60869390,
      "step": 2830,
      "time_per_iteration": 2.650834798812866
    },
    {
      "auxiliary_loss_clip": 0.01198725,
      "auxiliary_loss_mlp": 0.01028696,
      "balance_loss_clip": 1.05469477,
      "balance_loss_mlp": 1.01905227,
      "epoch": 0.3404076233992665,
      "flos": 22746034477440.0,
      "grad_norm": 2.4427850191943743,
      "language_loss": 0.71637809,
      "learning_rate": 3.071732513689267e-06,
      "loss": 0.73865235,
      "num_input_tokens_seen": 60887925,
      "step": 2831,
      "time_per_iteration": 2.639039993286133
    },
    {
      "auxiliary_loss_clip": 0.01200496,
      "auxiliary_loss_mlp": 0.01030643,
      "balance_loss_clip": 1.05909228,
      "balance_loss_mlp": 1.02088571,
      "epoch": 0.3405278662899056,
      "flos": 17052361839360.0,
      "grad_norm": 3.451405036991576,
      "language_loss": 0.67926562,
      "learning_rate": 3.0710747444094134e-06,
      "loss": 0.70157707,
      "num_input_tokens_seen": 60905955,
      "step": 2832,
      "time_per_iteration": 2.66194486618042
    },
    {
      "auxiliary_loss_clip": 0.0120215,
      "auxiliary_loss_mlp": 0.01031245,
      "balance_loss_clip": 1.05398011,
      "balance_loss_mlp": 1.02163696,
      "epoch": 0.3406481091805447,
      "flos": 42813131783040.0,
      "grad_norm": 3.2071178840139924,
      "language_loss": 0.65712494,
      "learning_rate": 3.070416812648372e-06,
      "loss": 0.67945886,
      "num_input_tokens_seen": 60929405,
      "step": 2833,
      "time_per_iteration": 2.870413064956665
    },
    {
      "auxiliary_loss_clip": 0.01199896,
      "auxiliary_loss_mlp": 0.01027148,
      "balance_loss_clip": 1.04892755,
      "balance_loss_mlp": 1.01855612,
      "epoch": 0.3407683520711838,
      "flos": 26761457917440.0,
      "grad_norm": 2.9220044723439424,
      "language_loss": 0.6555742,
      "learning_rate": 3.069758718505951e-06,
      "loss": 0.67784464,
      "num_input_tokens_seen": 60951145,
      "step": 2834,
      "time_per_iteration": 2.677917718887329
    },
    {
      "auxiliary_loss_clip": 0.01197052,
      "auxiliary_loss_mlp": 0.01033598,
      "balance_loss_clip": 1.05944204,
      "balance_loss_mlp": 1.02413869,
      "epoch": 0.3408885949618229,
      "flos": 28767643309440.0,
      "grad_norm": 1.772813021359223,
      "language_loss": 0.79986501,
      "learning_rate": 3.0691004620819836e-06,
      "loss": 0.82217145,
      "num_input_tokens_seen": 60971275,
      "step": 2835,
      "time_per_iteration": 2.6651973724365234
    },
    {
      "auxiliary_loss_clip": 0.01121079,
      "auxiliary_loss_mlp": 0.01001453,
      "balance_loss_clip": 1.02911234,
      "balance_loss_mlp": 1.00027871,
      "epoch": 0.341008837852462,
      "flos": 63576252881280.0,
      "grad_norm": 0.8033516204368965,
      "language_loss": 0.60191274,
      "learning_rate": 3.0684420434763254e-06,
      "loss": 0.62313807,
      "num_input_tokens_seen": 61037460,
      "step": 2836,
      "time_per_iteration": 3.2791528701782227
    },
    {
      "auxiliary_loss_clip": 0.01191954,
      "auxiliary_loss_mlp": 0.01031667,
      "balance_loss_clip": 1.05220151,
      "balance_loss_mlp": 1.02319384,
      "epoch": 0.34112908074310105,
      "flos": 20812173120000.0,
      "grad_norm": 1.8097350454193297,
      "language_loss": 0.77212423,
      "learning_rate": 3.06778346278886e-06,
      "loss": 0.79436046,
      "num_input_tokens_seen": 61056295,
      "step": 2837,
      "time_per_iteration": 2.7053117752075195
    },
    {
      "auxiliary_loss_clip": 0.01199618,
      "auxiliary_loss_mlp": 0.01023498,
      "balance_loss_clip": 1.06009364,
      "balance_loss_mlp": 1.01422334,
      "epoch": 0.34124932363374016,
      "flos": 24976446520320.0,
      "grad_norm": 1.7863663427274288,
      "language_loss": 0.78840822,
      "learning_rate": 3.0671247201194906e-06,
      "loss": 0.81063938,
      "num_input_tokens_seen": 61078430,
      "step": 2838,
      "time_per_iteration": 2.6870615482330322
    },
    {
      "auxiliary_loss_clip": 0.01206418,
      "auxiliary_loss_mlp": 0.01031442,
      "balance_loss_clip": 1.05340195,
      "balance_loss_mlp": 1.02189279,
      "epoch": 0.3413695665243792,
      "flos": 28402970480640.0,
      "grad_norm": 1.7178540686009887,
      "language_loss": 0.75676137,
      "learning_rate": 3.066465815568151e-06,
      "loss": 0.77914,
      "num_input_tokens_seen": 61099260,
      "step": 2839,
      "time_per_iteration": 2.7109367847442627
    },
    {
      "auxiliary_loss_clip": 0.01198715,
      "auxiliary_loss_mlp": 0.01024499,
      "balance_loss_clip": 1.05385685,
      "balance_loss_mlp": 1.01558173,
      "epoch": 0.34148980941501833,
      "flos": 25302012416640.0,
      "grad_norm": 1.6812672307285306,
      "language_loss": 0.68713164,
      "learning_rate": 3.0658067492347947e-06,
      "loss": 0.70936382,
      "num_input_tokens_seen": 61121900,
      "step": 2840,
      "time_per_iteration": 2.7120087146759033
    },
    {
      "auxiliary_loss_clip": 0.0120284,
      "auxiliary_loss_mlp": 0.01027743,
      "balance_loss_clip": 1.04691505,
      "balance_loss_mlp": 1.0179199,
      "epoch": 0.34161005230565744,
      "flos": 17530081747200.0,
      "grad_norm": 2.930121889389593,
      "language_loss": 0.66665298,
      "learning_rate": 3.065147521219402e-06,
      "loss": 0.68895882,
      "num_input_tokens_seen": 61141155,
      "step": 2841,
      "time_per_iteration": 2.7144718170166016
    },
    {
      "auxiliary_loss_clip": 0.01193254,
      "auxiliary_loss_mlp": 0.01032115,
      "balance_loss_clip": 1.0566628,
      "balance_loss_mlp": 1.02285814,
      "epoch": 0.3417302951962965,
      "flos": 43650101566080.0,
      "grad_norm": 1.421654495684915,
      "language_loss": 0.74290454,
      "learning_rate": 3.064488131621977e-06,
      "loss": 0.76515818,
      "num_input_tokens_seen": 61164480,
      "step": 2842,
      "time_per_iteration": 2.919564723968506
    },
    {
      "auxiliary_loss_clip": 0.01191101,
      "auxiliary_loss_mlp": 0.01029012,
      "balance_loss_clip": 1.05255997,
      "balance_loss_mlp": 1.01941538,
      "epoch": 0.3418505380869356,
      "flos": 30882207012480.0,
      "grad_norm": 1.8073513880182908,
      "language_loss": 0.73976207,
      "learning_rate": 3.063828580542549e-06,
      "loss": 0.76196319,
      "num_input_tokens_seen": 61185675,
      "step": 2843,
      "time_per_iteration": 2.8121588230133057
    },
    {
      "auxiliary_loss_clip": 0.01197772,
      "auxiliary_loss_mlp": 0.01033909,
      "balance_loss_clip": 1.05236685,
      "balance_loss_mlp": 1.02515268,
      "epoch": 0.3419707809775747,
      "flos": 19463871277440.0,
      "grad_norm": 2.2292089209994534,
      "language_loss": 0.735524,
      "learning_rate": 3.0631688680811706e-06,
      "loss": 0.75784081,
      "num_input_tokens_seen": 61205300,
      "step": 2844,
      "time_per_iteration": 3.6299796104431152
    },
    {
      "auxiliary_loss_clip": 0.01196599,
      "auxiliary_loss_mlp": 0.01031164,
      "balance_loss_clip": 1.05732524,
      "balance_loss_mlp": 1.02190173,
      "epoch": 0.3420910238682138,
      "flos": 28727818104960.0,
      "grad_norm": 7.954750712444567,
      "language_loss": 0.75698733,
      "learning_rate": 3.062508994337921e-06,
      "loss": 0.77926493,
      "num_input_tokens_seen": 61224905,
      "step": 2845,
      "time_per_iteration": 2.698932647705078
    },
    {
      "auxiliary_loss_clip": 0.01200134,
      "auxiliary_loss_mlp": 0.01028924,
      "balance_loss_clip": 1.05488968,
      "balance_loss_mlp": 1.01947057,
      "epoch": 0.3422112667588529,
      "flos": 21397265758080.0,
      "grad_norm": 2.6799044622006885,
      "language_loss": 0.79283136,
      "learning_rate": 3.0618489594129013e-06,
      "loss": 0.81512201,
      "num_input_tokens_seen": 61243045,
      "step": 2846,
      "time_per_iteration": 3.5613651275634766
    },
    {
      "auxiliary_loss_clip": 0.01207978,
      "auxiliary_loss_mlp": 0.0103015,
      "balance_loss_clip": 1.0552876,
      "balance_loss_mlp": 1.02132297,
      "epoch": 0.342331509649492,
      "flos": 13881450038400.0,
      "grad_norm": 2.62648900532115,
      "language_loss": 0.71452975,
      "learning_rate": 3.061188763406239e-06,
      "loss": 0.73691106,
      "num_input_tokens_seen": 61259190,
      "step": 2847,
      "time_per_iteration": 2.730022430419922
    },
    {
      "auxiliary_loss_clip": 0.01195584,
      "auxiliary_loss_mlp": 0.01034242,
      "balance_loss_clip": 1.05328274,
      "balance_loss_mlp": 1.02471137,
      "epoch": 0.34245175254013105,
      "flos": 28621450955520.0,
      "grad_norm": 2.4796768201941446,
      "language_loss": 0.82501662,
      "learning_rate": 3.060528406418085e-06,
      "loss": 0.84731477,
      "num_input_tokens_seen": 61279040,
      "step": 2848,
      "time_per_iteration": 2.715163230895996
    },
    {
      "auxiliary_loss_clip": 0.01194065,
      "auxiliary_loss_mlp": 0.010297,
      "balance_loss_clip": 1.05368161,
      "balance_loss_mlp": 1.02087283,
      "epoch": 0.34257199543077016,
      "flos": 34127058960000.0,
      "grad_norm": 1.6171189689779863,
      "language_loss": 0.61986119,
      "learning_rate": 3.0598678885486145e-06,
      "loss": 0.6420989,
      "num_input_tokens_seen": 61301580,
      "step": 2849,
      "time_per_iteration": 3.676306962966919
    },
    {
      "auxiliary_loss_clip": 0.01206366,
      "auxiliary_loss_mlp": 0.01209443,
      "balance_loss_clip": 1.05199397,
      "balance_loss_mlp": 1.00108778,
      "epoch": 0.34269223832140927,
      "flos": 19974018188160.0,
      "grad_norm": 2.0810848761256064,
      "language_loss": 0.74868441,
      "learning_rate": 3.0592072098980282e-06,
      "loss": 0.77284247,
      "num_input_tokens_seen": 61321240,
      "step": 2850,
      "time_per_iteration": 2.69330096244812
    },
    {
      "auxiliary_loss_clip": 0.0119375,
      "auxiliary_loss_mlp": 0.01031555,
      "balance_loss_clip": 1.05295289,
      "balance_loss_mlp": 1.02234066,
      "epoch": 0.3428124812120483,
      "flos": 27235658292480.0,
      "grad_norm": 1.7443528302344862,
      "language_loss": 0.7303344,
      "learning_rate": 3.0585463705665514e-06,
      "loss": 0.75258744,
      "num_input_tokens_seen": 61341615,
      "step": 2851,
      "time_per_iteration": 2.69640851020813
    },
    {
      "auxiliary_loss_clip": 0.01199528,
      "auxiliary_loss_mlp": 0.010332,
      "balance_loss_clip": 1.04848027,
      "balance_loss_mlp": 1.02405643,
      "epoch": 0.34293272410268744,
      "flos": 24570871079040.0,
      "grad_norm": 2.4789672082626435,
      "language_loss": 0.71220016,
      "learning_rate": 3.0578853706544304e-06,
      "loss": 0.73452741,
      "num_input_tokens_seen": 61359005,
      "step": 2852,
      "time_per_iteration": 3.630303144454956
    },
    {
      "auxiliary_loss_clip": 0.01204809,
      "auxiliary_loss_mlp": 0.01209695,
      "balance_loss_clip": 1.05230689,
      "balance_loss_mlp": 1.00110507,
      "epoch": 0.34305296699332655,
      "flos": 21506865131520.0,
      "grad_norm": 2.212731471635902,
      "language_loss": 0.65484613,
      "learning_rate": 3.0572242102619404e-06,
      "loss": 0.6789912,
      "num_input_tokens_seen": 61376160,
      "step": 2853,
      "time_per_iteration": 2.720919132232666
    },
    {
      "auxiliary_loss_clip": 0.01198295,
      "auxiliary_loss_mlp": 0.01028651,
      "balance_loss_clip": 1.05465531,
      "balance_loss_mlp": 1.01944828,
      "epoch": 0.3431732098839656,
      "flos": 24056665931520.0,
      "grad_norm": 2.167534618631957,
      "language_loss": 0.80623007,
      "learning_rate": 3.0565628894893784e-06,
      "loss": 0.82849956,
      "num_input_tokens_seen": 61396795,
      "step": 2854,
      "time_per_iteration": 2.7741832733154297
    },
    {
      "auxiliary_loss_clip": 0.01190178,
      "auxiliary_loss_mlp": 0.01030624,
      "balance_loss_clip": 1.05507278,
      "balance_loss_mlp": 1.0214088,
      "epoch": 0.3432934527746047,
      "flos": 16800879744000.0,
      "grad_norm": 1.7848665185126174,
      "language_loss": 0.74777973,
      "learning_rate": 3.0559014084370655e-06,
      "loss": 0.76998776,
      "num_input_tokens_seen": 61415320,
      "step": 2855,
      "time_per_iteration": 2.6389291286468506
    },
    {
      "auxiliary_loss_clip": 0.01204553,
      "auxiliary_loss_mlp": 0.01029283,
      "balance_loss_clip": 1.05437446,
      "balance_loss_mlp": 1.01935887,
      "epoch": 0.34341369566524377,
      "flos": 23439720908160.0,
      "grad_norm": 1.7807381376837657,
      "language_loss": 0.78993082,
      "learning_rate": 3.055239767205349e-06,
      "loss": 0.81226909,
      "num_input_tokens_seen": 61437070,
      "step": 2856,
      "time_per_iteration": 2.6703736782073975
    },
    {
      "auxiliary_loss_clip": 0.01200579,
      "auxiliary_loss_mlp": 0.01031409,
      "balance_loss_clip": 1.06137216,
      "balance_loss_mlp": 1.02258158,
      "epoch": 0.3435339385558829,
      "flos": 17267466435840.0,
      "grad_norm": 2.459454329574069,
      "language_loss": 0.78339708,
      "learning_rate": 3.054577965894599e-06,
      "loss": 0.80571693,
      "num_input_tokens_seen": 61453215,
      "step": 2857,
      "time_per_iteration": 2.6328108310699463
    },
    {
      "auxiliary_loss_clip": 0.01210805,
      "auxiliary_loss_mlp": 0.0102779,
      "balance_loss_clip": 1.05964768,
      "balance_loss_mlp": 1.01734734,
      "epoch": 0.343654181446522,
      "flos": 22199366413440.0,
      "grad_norm": 1.7663323094884933,
      "language_loss": 0.70376664,
      "learning_rate": 3.0539160046052094e-06,
      "loss": 0.7261526,
      "num_input_tokens_seen": 61472915,
      "step": 2858,
      "time_per_iteration": 2.656914472579956
    },
    {
      "auxiliary_loss_clip": 0.01195414,
      "auxiliary_loss_mlp": 0.01031954,
      "balance_loss_clip": 1.053859,
      "balance_loss_mlp": 1.02121902,
      "epoch": 0.34377442433716104,
      "flos": 19901801894400.0,
      "grad_norm": 2.5751296166406767,
      "language_loss": 0.70548058,
      "learning_rate": 3.0532538834376003e-06,
      "loss": 0.72775435,
      "num_input_tokens_seen": 61492475,
      "step": 2859,
      "time_per_iteration": 2.708272933959961
    },
    {
      "auxiliary_loss_clip": 0.01202514,
      "auxiliary_loss_mlp": 0.0103125,
      "balance_loss_clip": 1.05648661,
      "balance_loss_mlp": 1.02188921,
      "epoch": 0.34389466722780015,
      "flos": 22197678474240.0,
      "grad_norm": 1.8698867958397853,
      "language_loss": 0.78378356,
      "learning_rate": 3.0525916024922143e-06,
      "loss": 0.80612111,
      "num_input_tokens_seen": 61511660,
      "step": 2860,
      "time_per_iteration": 2.6329379081726074
    },
    {
      "auxiliary_loss_clip": 0.01200489,
      "auxiliary_loss_mlp": 0.01030872,
      "balance_loss_clip": 1.0550977,
      "balance_loss_mlp": 1.02156198,
      "epoch": 0.34401491011843927,
      "flos": 18624567110400.0,
      "grad_norm": 2.639108510455606,
      "language_loss": 0.84259069,
      "learning_rate": 3.0519291618695193e-06,
      "loss": 0.86490428,
      "num_input_tokens_seen": 61529060,
      "step": 2861,
      "time_per_iteration": 2.6793947219848633
    },
    {
      "auxiliary_loss_clip": 0.0119583,
      "auxiliary_loss_mlp": 0.01030168,
      "balance_loss_clip": 1.04997635,
      "balance_loss_mlp": 1.02094173,
      "epoch": 0.3441351530090783,
      "flos": 17858197509120.0,
      "grad_norm": 24.66834687253494,
      "language_loss": 0.75689483,
      "learning_rate": 3.0512665616700065e-06,
      "loss": 0.77915478,
      "num_input_tokens_seen": 61548125,
      "step": 2862,
      "time_per_iteration": 2.644728899002075
    },
    {
      "auxiliary_loss_clip": 0.01196945,
      "auxiliary_loss_mlp": 0.01033068,
      "balance_loss_clip": 1.04864967,
      "balance_loss_mlp": 1.02397871,
      "epoch": 0.34425539589971743,
      "flos": 23112754381440.0,
      "grad_norm": 2.209013213084495,
      "language_loss": 0.89097452,
      "learning_rate": 3.0506038019941933e-06,
      "loss": 0.91327465,
      "num_input_tokens_seen": 61568135,
      "step": 2863,
      "time_per_iteration": 2.776232957839966
    },
    {
      "auxiliary_loss_clip": 0.01203996,
      "auxiliary_loss_mlp": 0.01027987,
      "balance_loss_clip": 1.05371642,
      "balance_loss_mlp": 1.01870084,
      "epoch": 0.34437563879035654,
      "flos": 21907699977600.0,
      "grad_norm": 3.2509278749351447,
      "language_loss": 0.6775552,
      "learning_rate": 3.049940882942617e-06,
      "loss": 0.69987512,
      "num_input_tokens_seen": 61586920,
      "step": 2864,
      "time_per_iteration": 2.820765972137451
    },
    {
      "auxiliary_loss_clip": 0.01197495,
      "auxiliary_loss_mlp": 0.01033847,
      "balance_loss_clip": 1.05763292,
      "balance_loss_mlp": 1.0241729,
      "epoch": 0.3444958816809956,
      "flos": 23076915586560.0,
      "grad_norm": 3.6010427952578854,
      "language_loss": 0.80312842,
      "learning_rate": 3.0492778046158448e-06,
      "loss": 0.82544178,
      "num_input_tokens_seen": 61608340,
      "step": 2865,
      "time_per_iteration": 2.676616907119751
    },
    {
      "auxiliary_loss_clip": 0.01196984,
      "auxiliary_loss_mlp": 0.0102789,
      "balance_loss_clip": 1.05713725,
      "balance_loss_mlp": 1.01938462,
      "epoch": 0.3446161245716347,
      "flos": 21908633731200.0,
      "grad_norm": 2.218271396032908,
      "language_loss": 0.76957715,
      "learning_rate": 3.0486145671144633e-06,
      "loss": 0.79182589,
      "num_input_tokens_seen": 61628130,
      "step": 2866,
      "time_per_iteration": 2.6620934009552
    },
    {
      "auxiliary_loss_clip": 0.01186182,
      "auxiliary_loss_mlp": 0.0103233,
      "balance_loss_clip": 1.04466367,
      "balance_loss_mlp": 1.02243578,
      "epoch": 0.3447363674622738,
      "flos": 25112834461440.0,
      "grad_norm": 2.9193117285665804,
      "language_loss": 0.77553564,
      "learning_rate": 3.047951170539086e-06,
      "loss": 0.79772079,
      "num_input_tokens_seen": 61647755,
      "step": 2867,
      "time_per_iteration": 2.730525255203247
    },
    {
      "auxiliary_loss_clip": 0.01198801,
      "auxiliary_loss_mlp": 0.01036657,
      "balance_loss_clip": 1.05566597,
      "balance_loss_mlp": 1.02815735,
      "epoch": 0.3448566103529129,
      "flos": 11984684451840.0,
      "grad_norm": 2.1316305721831963,
      "language_loss": 0.84732306,
      "learning_rate": 3.047287614990349e-06,
      "loss": 0.86967766,
      "num_input_tokens_seen": 61665675,
      "step": 2868,
      "time_per_iteration": 2.6789536476135254
    },
    {
      "auxiliary_loss_clip": 0.0119459,
      "auxiliary_loss_mlp": 0.01031184,
      "balance_loss_clip": 1.05425739,
      "balance_loss_mlp": 1.02126002,
      "epoch": 0.344976853243552,
      "flos": 40187882465280.0,
      "grad_norm": 2.4452309993294703,
      "language_loss": 0.6201793,
      "learning_rate": 3.046623900568914e-06,
      "loss": 0.64243704,
      "num_input_tokens_seen": 61688240,
      "step": 2869,
      "time_per_iteration": 2.7988898754119873
    },
    {
      "auxiliary_loss_clip": 0.01198047,
      "auxiliary_loss_mlp": 0.01030667,
      "balance_loss_clip": 1.0547694,
      "balance_loss_mlp": 1.02071941,
      "epoch": 0.34509709613419104,
      "flos": 28723652127360.0,
      "grad_norm": 2.731072581526885,
      "language_loss": 0.70221627,
      "learning_rate": 3.045960027375465e-06,
      "loss": 0.7245034,
      "num_input_tokens_seen": 61706075,
      "step": 2870,
      "time_per_iteration": 2.7351186275482178
    },
    {
      "auxiliary_loss_clip": 0.01204843,
      "auxiliary_loss_mlp": 0.01029728,
      "balance_loss_clip": 1.05712938,
      "balance_loss_mlp": 1.01949358,
      "epoch": 0.34521733902483015,
      "flos": 29967597982080.0,
      "grad_norm": 4.9826625841904715,
      "language_loss": 0.83012438,
      "learning_rate": 3.045295995510711e-06,
      "loss": 0.85246998,
      "num_input_tokens_seen": 61723045,
      "step": 2871,
      "time_per_iteration": 4.623464107513428
    },
    {
      "auxiliary_loss_clip": 0.0119816,
      "auxiliary_loss_mlp": 0.01030733,
      "balance_loss_clip": 1.05599844,
      "balance_loss_mlp": 1.02192903,
      "epoch": 0.34533758191546926,
      "flos": 27923059843200.0,
      "grad_norm": 1.9708527211063078,
      "language_loss": 0.73657995,
      "learning_rate": 3.0446318050753865e-06,
      "loss": 0.75886887,
      "num_input_tokens_seen": 61743525,
      "step": 2872,
      "time_per_iteration": 2.7203352451324463
    },
    {
      "auxiliary_loss_clip": 0.01191508,
      "auxiliary_loss_mlp": 0.01031907,
      "balance_loss_clip": 1.05531454,
      "balance_loss_mlp": 1.02332962,
      "epoch": 0.3454578248061083,
      "flos": 27125879351040.0,
      "grad_norm": 2.0084194738978347,
      "language_loss": 0.77610338,
      "learning_rate": 3.0439674561702474e-06,
      "loss": 0.79833746,
      "num_input_tokens_seen": 61763025,
      "step": 2873,
      "time_per_iteration": 2.720691204071045
    },
    {
      "auxiliary_loss_clip": 0.01195928,
      "auxiliary_loss_mlp": 0.0102958,
      "balance_loss_clip": 1.05551708,
      "balance_loss_mlp": 1.02050185,
      "epoch": 0.3455780676967474,
      "flos": 19024899166080.0,
      "grad_norm": 2.215477418502816,
      "language_loss": 0.88392049,
      "learning_rate": 3.043302948896076e-06,
      "loss": 0.90617555,
      "num_input_tokens_seen": 61781630,
      "step": 2874,
      "time_per_iteration": 2.6148695945739746
    },
    {
      "auxiliary_loss_clip": 0.01195371,
      "auxiliary_loss_mlp": 0.01032311,
      "balance_loss_clip": 1.05131745,
      "balance_loss_mlp": 1.02241123,
      "epoch": 0.34569831058738654,
      "flos": 34496005507200.0,
      "grad_norm": 2.803057257866328,
      "language_loss": 0.60692656,
      "learning_rate": 3.0426382833536756e-06,
      "loss": 0.62920338,
      "num_input_tokens_seen": 61804985,
      "step": 2875,
      "time_per_iteration": 3.803445339202881
    },
    {
      "auxiliary_loss_clip": 0.01200523,
      "auxiliary_loss_mlp": 0.01027296,
      "balance_loss_clip": 1.05298126,
      "balance_loss_mlp": 1.01824844,
      "epoch": 0.3458185534780256,
      "flos": 31138681098240.0,
      "grad_norm": 1.9855718310022903,
      "language_loss": 0.78326309,
      "learning_rate": 3.041973459643877e-06,
      "loss": 0.80554128,
      "num_input_tokens_seen": 61824440,
      "step": 2876,
      "time_per_iteration": 2.8508338928222656
    },
    {
      "auxiliary_loss_clip": 0.01199101,
      "auxiliary_loss_mlp": 0.0102575,
      "balance_loss_clip": 1.04954779,
      "balance_loss_mlp": 1.01656473,
      "epoch": 0.3459387963686647,
      "flos": 32452508862720.0,
      "grad_norm": 2.04265898611729,
      "language_loss": 0.67012292,
      "learning_rate": 3.0413084778675334e-06,
      "loss": 0.69237149,
      "num_input_tokens_seen": 61845690,
      "step": 2877,
      "time_per_iteration": 2.8190364837646484
    },
    {
      "auxiliary_loss_clip": 0.01195204,
      "auxiliary_loss_mlp": 0.01208887,
      "balance_loss_clip": 1.05333877,
      "balance_loss_mlp": 1.00111485,
      "epoch": 0.3460590392593038,
      "flos": 24675658030080.0,
      "grad_norm": 2.8238544677604622,
      "language_loss": 0.84144115,
      "learning_rate": 3.0406433381255214e-06,
      "loss": 0.86548209,
      "num_input_tokens_seen": 61863725,
      "step": 2878,
      "time_per_iteration": 3.7068164348602295
    },
    {
      "auxiliary_loss_clip": 0.01200586,
      "auxiliary_loss_mlp": 0.01025533,
      "balance_loss_clip": 1.06159282,
      "balance_loss_mlp": 1.01717913,
      "epoch": 0.34617928214994287,
      "flos": 18807316531200.0,
      "grad_norm": 2.4189312367474627,
      "language_loss": 0.82505369,
      "learning_rate": 3.0399780405187425e-06,
      "loss": 0.84731495,
      "num_input_tokens_seen": 61882720,
      "step": 2879,
      "time_per_iteration": 2.665994644165039
    },
    {
      "auxiliary_loss_clip": 0.01194891,
      "auxiliary_loss_mlp": 0.01024238,
      "balance_loss_clip": 1.05633736,
      "balance_loss_mlp": 1.01594663,
      "epoch": 0.346299525040582,
      "flos": 24857653265280.0,
      "grad_norm": 2.1370977533187334,
      "language_loss": 0.78636146,
      "learning_rate": 3.0393125851481216e-06,
      "loss": 0.80855274,
      "num_input_tokens_seen": 61902595,
      "step": 2880,
      "time_per_iteration": 2.7113733291625977
    },
    {
      "auxiliary_loss_clip": 0.01202116,
      "auxiliary_loss_mlp": 0.01024401,
      "balance_loss_clip": 1.05705476,
      "balance_loss_mlp": 1.0158354,
      "epoch": 0.3464197679312211,
      "flos": 16434914025600.0,
      "grad_norm": 2.989799532187806,
      "language_loss": 0.86317801,
      "learning_rate": 3.038646972114608e-06,
      "loss": 0.88544315,
      "num_input_tokens_seen": 61918920,
      "step": 2881,
      "time_per_iteration": 2.711107015609741
    },
    {
      "auxiliary_loss_clip": 0.01201057,
      "auxiliary_loss_mlp": 0.01032589,
      "balance_loss_clip": 1.05691564,
      "balance_loss_mlp": 1.023911,
      "epoch": 0.34654001082186014,
      "flos": 22382474970240.0,
      "grad_norm": 1.89655777050148,
      "language_loss": 0.67559135,
      "learning_rate": 3.037981201519174e-06,
      "loss": 0.69792771,
      "num_input_tokens_seen": 61939520,
      "step": 2882,
      "time_per_iteration": 2.7195262908935547
    },
    {
      "auxiliary_loss_clip": 0.01201006,
      "auxiliary_loss_mlp": 0.01032858,
      "balance_loss_clip": 1.0610292,
      "balance_loss_mlp": 1.02372098,
      "epoch": 0.34666025371249926,
      "flos": 19573901614080.0,
      "grad_norm": 2.311379489678107,
      "language_loss": 0.71684182,
      "learning_rate": 3.0373152734628175e-06,
      "loss": 0.73918045,
      "num_input_tokens_seen": 61957800,
      "step": 2883,
      "time_per_iteration": 2.6503360271453857
    },
    {
      "auxiliary_loss_clip": 0.01190916,
      "auxiliary_loss_mlp": 0.01026035,
      "balance_loss_clip": 1.05197549,
      "balance_loss_mlp": 1.01695704,
      "epoch": 0.34678049660313837,
      "flos": 15267637751040.0,
      "grad_norm": 2.2932038599473477,
      "language_loss": 0.76172465,
      "learning_rate": 3.0366491880465584e-06,
      "loss": 0.78389418,
      "num_input_tokens_seen": 61975820,
      "step": 2884,
      "time_per_iteration": 2.5714666843414307
    },
    {
      "auxiliary_loss_clip": 0.01202883,
      "auxiliary_loss_mlp": 0.01029794,
      "balance_loss_clip": 1.06356072,
      "balance_loss_mlp": 1.02036452,
      "epoch": 0.3469007394937774,
      "flos": 21181550630400.0,
      "grad_norm": 2.918945831550973,
      "language_loss": 0.82047367,
      "learning_rate": 3.035982945371443e-06,
      "loss": 0.8428005,
      "num_input_tokens_seen": 61997515,
      "step": 2885,
      "time_per_iteration": 2.6764063835144043
    },
    {
      "auxiliary_loss_clip": 0.01206938,
      "auxiliary_loss_mlp": 0.01029007,
      "balance_loss_clip": 1.0558331,
      "balance_loss_mlp": 1.01940417,
      "epoch": 0.34702098238441653,
      "flos": 22375471818240.0,
      "grad_norm": 2.4270970168904706,
      "language_loss": 0.85519511,
      "learning_rate": 3.035316545538537e-06,
      "loss": 0.87755454,
      "num_input_tokens_seen": 62016310,
      "step": 2886,
      "time_per_iteration": 2.607307195663452
    },
    {
      "auxiliary_loss_clip": 0.01199457,
      "auxiliary_loss_mlp": 0.01029067,
      "balance_loss_clip": 1.05805624,
      "balance_loss_mlp": 1.02054405,
      "epoch": 0.3471412252750556,
      "flos": 22929430343040.0,
      "grad_norm": 2.0532329822563264,
      "language_loss": 0.79448932,
      "learning_rate": 3.034649988648935e-06,
      "loss": 0.81677455,
      "num_input_tokens_seen": 62036075,
      "step": 2887,
      "time_per_iteration": 2.677419662475586
    },
    {
      "auxiliary_loss_clip": 0.0120368,
      "auxiliary_loss_mlp": 0.0102539,
      "balance_loss_clip": 1.05574143,
      "balance_loss_mlp": 1.01638401,
      "epoch": 0.3472614681656947,
      "flos": 21324259365120.0,
      "grad_norm": 1.731989079884638,
      "language_loss": 0.80863631,
      "learning_rate": 3.033983274803752e-06,
      "loss": 0.83092701,
      "num_input_tokens_seen": 62055865,
      "step": 2888,
      "time_per_iteration": 2.6010961532592773
    },
    {
      "auxiliary_loss_clip": 0.01196027,
      "auxiliary_loss_mlp": 0.01031885,
      "balance_loss_clip": 1.05284846,
      "balance_loss_mlp": 1.02265573,
      "epoch": 0.3473817110563338,
      "flos": 23475739271040.0,
      "grad_norm": 10.566672059899902,
      "language_loss": 0.72855043,
      "learning_rate": 3.0333164041041283e-06,
      "loss": 0.75082964,
      "num_input_tokens_seen": 62072180,
      "step": 2889,
      "time_per_iteration": 2.643549919128418
    },
    {
      "auxiliary_loss_clip": 0.01207616,
      "auxiliary_loss_mlp": 0.01025793,
      "balance_loss_clip": 1.04930544,
      "balance_loss_mlp": 1.0171864,
      "epoch": 0.34750195394697286,
      "flos": 22346025644160.0,
      "grad_norm": 2.679606585315806,
      "language_loss": 0.72192723,
      "learning_rate": 3.032649376651228e-06,
      "loss": 0.74426132,
      "num_input_tokens_seen": 62091600,
      "step": 2890,
      "time_per_iteration": 2.7106831073760986
    },
    {
      "auxiliary_loss_clip": 0.01203213,
      "auxiliary_loss_mlp": 0.01026579,
      "balance_loss_clip": 1.05325842,
      "balance_loss_mlp": 1.01693463,
      "epoch": 0.347622196837612,
      "flos": 29095004885760.0,
      "grad_norm": 2.2227476998885556,
      "language_loss": 0.75757396,
      "learning_rate": 3.031982192546238e-06,
      "loss": 0.77987194,
      "num_input_tokens_seen": 62114695,
      "step": 2891,
      "time_per_iteration": 2.7828209400177
    },
    {
      "auxiliary_loss_clip": 0.01200965,
      "auxiliary_loss_mlp": 0.01029867,
      "balance_loss_clip": 1.05768692,
      "balance_loss_mlp": 1.02112854,
      "epoch": 0.3477424397282511,
      "flos": 22455732758400.0,
      "grad_norm": 2.1173691475747667,
      "language_loss": 0.94929719,
      "learning_rate": 3.0313148518903696e-06,
      "loss": 0.97160554,
      "num_input_tokens_seen": 62134520,
      "step": 2892,
      "time_per_iteration": 2.6877601146698
    },
    {
      "auxiliary_loss_clip": 0.01204441,
      "auxiliary_loss_mlp": 0.01025593,
      "balance_loss_clip": 1.05808985,
      "balance_loss_mlp": 1.01644993,
      "epoch": 0.34786268261889014,
      "flos": 15778790242560.0,
      "grad_norm": 2.518221373762116,
      "language_loss": 0.81133199,
      "learning_rate": 3.030647354784859e-06,
      "loss": 0.83363229,
      "num_input_tokens_seen": 62151560,
      "step": 2893,
      "time_per_iteration": 2.695913314819336
    },
    {
      "auxiliary_loss_clip": 0.01200625,
      "auxiliary_loss_mlp": 0.01031324,
      "balance_loss_clip": 1.0517714,
      "balance_loss_mlp": 1.02204347,
      "epoch": 0.34798292550952925,
      "flos": 20777627214720.0,
      "grad_norm": 2.071949490384783,
      "language_loss": 0.77377617,
      "learning_rate": 3.029979701330964e-06,
      "loss": 0.79609561,
      "num_input_tokens_seen": 62170985,
      "step": 2894,
      "time_per_iteration": 2.6679742336273193
    },
    {
      "auxiliary_loss_clip": 0.01204992,
      "auxiliary_loss_mlp": 0.01030961,
      "balance_loss_clip": 1.055884,
      "balance_loss_mlp": 1.0220741,
      "epoch": 0.34810316840016836,
      "flos": 19937820257280.0,
      "grad_norm": 2.0342332745682965,
      "language_loss": 0.80454004,
      "learning_rate": 3.029311891629966e-06,
      "loss": 0.82689965,
      "num_input_tokens_seen": 62189440,
      "step": 2895,
      "time_per_iteration": 2.702807903289795
    },
    {
      "auxiliary_loss_clip": 0.01198557,
      "auxiliary_loss_mlp": 0.01033326,
      "balance_loss_clip": 1.0569346,
      "balance_loss_mlp": 1.02421868,
      "epoch": 0.3482234112908074,
      "flos": 23623296341760.0,
      "grad_norm": 2.1298779792459888,
      "language_loss": 0.74379563,
      "learning_rate": 3.0286439257831744e-06,
      "loss": 0.76611447,
      "num_input_tokens_seen": 62208910,
      "step": 2896,
      "time_per_iteration": 2.7270164489746094
    },
    {
      "auxiliary_loss_clip": 0.01201932,
      "auxiliary_loss_mlp": 0.01038041,
      "balance_loss_clip": 1.05979753,
      "balance_loss_mlp": 1.02741361,
      "epoch": 0.3483436541814465,
      "flos": 23986712194560.0,
      "grad_norm": 2.014658443688505,
      "language_loss": 0.71282697,
      "learning_rate": 3.0279758038919156e-06,
      "loss": 0.73522669,
      "num_input_tokens_seen": 62227135,
      "step": 2897,
      "time_per_iteration": 2.6415038108825684
    },
    {
      "auxiliary_loss_clip": 0.01198724,
      "auxiliary_loss_mlp": 0.01029617,
      "balance_loss_clip": 1.05687475,
      "balance_loss_mlp": 1.02052128,
      "epoch": 0.34846389707208564,
      "flos": 22638338524800.0,
      "grad_norm": 2.0987459364860723,
      "language_loss": 0.78417718,
      "learning_rate": 3.0273075260575455e-06,
      "loss": 0.80646062,
      "num_input_tokens_seen": 62246035,
      "step": 2898,
      "time_per_iteration": 4.562588691711426
    },
    {
      "auxiliary_loss_clip": 0.01203518,
      "auxiliary_loss_mlp": 0.01033877,
      "balance_loss_clip": 1.05582345,
      "balance_loss_mlp": 1.02380347,
      "epoch": 0.3485841399627247,
      "flos": 21792857218560.0,
      "grad_norm": 1.8195680797280451,
      "language_loss": 0.80900443,
      "learning_rate": 3.0266390923814396e-06,
      "loss": 0.8313784,
      "num_input_tokens_seen": 62264095,
      "step": 2899,
      "time_per_iteration": 2.717331647872925
    },
    {
      "auxiliary_loss_clip": 0.01207475,
      "auxiliary_loss_mlp": 0.01034532,
      "balance_loss_clip": 1.06165171,
      "balance_loss_mlp": 1.02435803,
      "epoch": 0.3487043828533638,
      "flos": 17019036996480.0,
      "grad_norm": 1.9713135159486201,
      "language_loss": 0.82363319,
      "learning_rate": 3.0259705029650008e-06,
      "loss": 0.84605324,
      "num_input_tokens_seen": 62282025,
      "step": 2900,
      "time_per_iteration": 2.6333446502685547
    },
    {
      "auxiliary_loss_clip": 0.01199715,
      "auxiliary_loss_mlp": 0.01025571,
      "balance_loss_clip": 1.05504358,
      "balance_loss_mlp": 1.01735747,
      "epoch": 0.34882462574400286,
      "flos": 22601135013120.0,
      "grad_norm": 2.0093285557831724,
      "language_loss": 0.73028147,
      "learning_rate": 3.025301757909652e-06,
      "loss": 0.75253439,
      "num_input_tokens_seen": 62302220,
      "step": 2901,
      "time_per_iteration": 2.8183038234710693
    },
    {
      "auxiliary_loss_clip": 0.01205982,
      "auxiliary_loss_mlp": 0.01210003,
      "balance_loss_clip": 1.05346966,
      "balance_loss_mlp": 1.00119162,
      "epoch": 0.34894486863464197,
      "flos": 29861518141440.0,
      "grad_norm": 1.522219323364293,
      "language_loss": 0.80554724,
      "learning_rate": 3.024632857316842e-06,
      "loss": 0.82970715,
      "num_input_tokens_seen": 62323535,
      "step": 2902,
      "time_per_iteration": 3.6139206886291504
    },
    {
      "auxiliary_loss_clip": 0.01203775,
      "auxiliary_loss_mlp": 0.0102713,
      "balance_loss_clip": 1.06152773,
      "balance_loss_mlp": 1.01811779,
      "epoch": 0.3490651115252811,
      "flos": 22122265870080.0,
      "grad_norm": 1.8428851843933927,
      "language_loss": 0.77489185,
      "learning_rate": 3.0239638012880412e-06,
      "loss": 0.79720092,
      "num_input_tokens_seen": 62343430,
      "step": 2903,
      "time_per_iteration": 2.6656782627105713
    },
    {
      "auxiliary_loss_clip": 0.0119596,
      "auxiliary_loss_mlp": 0.01024915,
      "balance_loss_clip": 1.04918551,
      "balance_loss_mlp": 1.01549125,
      "epoch": 0.34918535441592014,
      "flos": 12676682943360.0,
      "grad_norm": 3.46876486779693,
      "language_loss": 0.81361532,
      "learning_rate": 3.0232945899247466e-06,
      "loss": 0.83582413,
      "num_input_tokens_seen": 62360365,
      "step": 2904,
      "time_per_iteration": 2.6995327472686768
    },
    {
      "auxiliary_loss_clip": 0.01200304,
      "auxiliary_loss_mlp": 0.01037015,
      "balance_loss_clip": 1.05607915,
      "balance_loss_mlp": 1.02709675,
      "epoch": 0.34930559730655925,
      "flos": 23185617120000.0,
      "grad_norm": 2.5209077110614246,
      "language_loss": 0.77375931,
      "learning_rate": 3.022625223328476e-06,
      "loss": 0.7961325,
      "num_input_tokens_seen": 62382105,
      "step": 2905,
      "time_per_iteration": 3.7315194606781006
    },
    {
      "auxiliary_loss_clip": 0.01206588,
      "auxiliary_loss_mlp": 0.01031892,
      "balance_loss_clip": 1.05946088,
      "balance_loss_mlp": 1.02090061,
      "epoch": 0.34942584019719836,
      "flos": 22855023319680.0,
      "grad_norm": 1.505478614676838,
      "language_loss": 0.69405478,
      "learning_rate": 3.0219557016007723e-06,
      "loss": 0.71643955,
      "num_input_tokens_seen": 62402235,
      "step": 2906,
      "time_per_iteration": 2.72121000289917
    },
    {
      "auxiliary_loss_clip": 0.01196125,
      "auxiliary_loss_mlp": 0.01029393,
      "balance_loss_clip": 1.05772853,
      "balance_loss_mlp": 1.02023745,
      "epoch": 0.3495460830878374,
      "flos": 24426043441920.0,
      "grad_norm": 1.914516308863709,
      "language_loss": 0.69970202,
      "learning_rate": 3.021286024843202e-06,
      "loss": 0.72195721,
      "num_input_tokens_seen": 62420430,
      "step": 2907,
      "time_per_iteration": 2.688716411590576
    },
    {
      "auxiliary_loss_clip": 0.01114563,
      "auxiliary_loss_mlp": 0.01001029,
      "balance_loss_clip": 1.04250407,
      "balance_loss_mlp": 0.99987239,
      "epoch": 0.3496663259784765,
      "flos": 70008749389440.0,
      "grad_norm": 1.0654941465468817,
      "language_loss": 0.64745021,
      "learning_rate": 3.0206161931573526e-06,
      "loss": 0.66860616,
      "num_input_tokens_seen": 62472980,
      "step": 2908,
      "time_per_iteration": 3.170686960220337
    },
    {
      "auxiliary_loss_clip": 0.01197527,
      "auxiliary_loss_mlp": 0.01029598,
      "balance_loss_clip": 1.05186582,
      "balance_loss_mlp": 1.02108657,
      "epoch": 0.34978656886911563,
      "flos": 28692805322880.0,
      "grad_norm": 1.7840989837184786,
      "language_loss": 0.9305414,
      "learning_rate": 3.0199462066448388e-06,
      "loss": 0.95281267,
      "num_input_tokens_seen": 62495175,
      "step": 2909,
      "time_per_iteration": 2.7074501514434814
    },
    {
      "auxiliary_loss_clip": 0.01200529,
      "auxiliary_loss_mlp": 0.01029557,
      "balance_loss_clip": 1.06019688,
      "balance_loss_mlp": 1.02006805,
      "epoch": 0.3499068117597547,
      "flos": 21142156389120.0,
      "grad_norm": 1.8312077895968524,
      "language_loss": 0.69817781,
      "learning_rate": 3.019276065407296e-06,
      "loss": 0.72047865,
      "num_input_tokens_seen": 62514295,
      "step": 2910,
      "time_per_iteration": 2.6670937538146973
    },
    {
      "auxiliary_loss_clip": 0.01209463,
      "auxiliary_loss_mlp": 0.01035597,
      "balance_loss_clip": 1.05262446,
      "balance_loss_mlp": 1.02637649,
      "epoch": 0.3500270546503938,
      "flos": 22782699285120.0,
      "grad_norm": 1.8583959381481536,
      "language_loss": 0.80626738,
      "learning_rate": 3.018605769546385e-06,
      "loss": 0.82871801,
      "num_input_tokens_seen": 62534850,
      "step": 2911,
      "time_per_iteration": 2.7413394451141357
    },
    {
      "auxiliary_loss_clip": 0.01197007,
      "auxiliary_loss_mlp": 0.01032765,
      "balance_loss_clip": 1.05478191,
      "balance_loss_mlp": 1.02322185,
      "epoch": 0.3501472975410329,
      "flos": 22894058424960.0,
      "grad_norm": 2.194298959543356,
      "language_loss": 0.7950666,
      "learning_rate": 3.017935319163788e-06,
      "loss": 0.81736422,
      "num_input_tokens_seen": 62553810,
      "step": 2912,
      "time_per_iteration": 2.6532180309295654
    },
    {
      "auxiliary_loss_clip": 0.01200062,
      "auxiliary_loss_mlp": 0.01030042,
      "balance_loss_clip": 1.05771327,
      "balance_loss_mlp": 1.02004683,
      "epoch": 0.35026754043167196,
      "flos": 25446588658560.0,
      "grad_norm": 3.507729746252695,
      "language_loss": 0.70847708,
      "learning_rate": 3.017264714361213e-06,
      "loss": 0.7307781,
      "num_input_tokens_seen": 62573460,
      "step": 2913,
      "time_per_iteration": 2.718052864074707
    },
    {
      "auxiliary_loss_clip": 0.0119892,
      "auxiliary_loss_mlp": 0.01209273,
      "balance_loss_clip": 1.05451441,
      "balance_loss_mlp": 1.00117099,
      "epoch": 0.3503877833223111,
      "flos": 19573757959680.0,
      "grad_norm": 2.2225601039513907,
      "language_loss": 0.823645,
      "learning_rate": 3.016593955240389e-06,
      "loss": 0.84772694,
      "num_input_tokens_seen": 62592150,
      "step": 2914,
      "time_per_iteration": 2.6144206523895264
    },
    {
      "auxiliary_loss_clip": 0.01115918,
      "auxiliary_loss_mlp": 0.01000516,
      "balance_loss_clip": 1.03985989,
      "balance_loss_mlp": 0.99942535,
      "epoch": 0.3505080262129502,
      "flos": 65072075880960.0,
      "grad_norm": 0.822764495283062,
      "language_loss": 0.63696659,
      "learning_rate": 3.015923041903071e-06,
      "loss": 0.658131,
      "num_input_tokens_seen": 62658275,
      "step": 2915,
      "time_per_iteration": 3.2675323486328125
    },
    {
      "auxiliary_loss_clip": 0.01201548,
      "auxiliary_loss_mlp": 0.01028793,
      "balance_loss_clip": 1.06158507,
      "balance_loss_mlp": 1.01940525,
      "epoch": 0.35062826910358924,
      "flos": 29314562768640.0,
      "grad_norm": 2.0395332521565464,
      "language_loss": 0.83720255,
      "learning_rate": 3.0152519744510347e-06,
      "loss": 0.85950601,
      "num_input_tokens_seen": 62678075,
      "step": 2916,
      "time_per_iteration": 2.660799980163574
    },
    {
      "auxiliary_loss_clip": 0.0120451,
      "auxiliary_loss_mlp": 0.0102907,
      "balance_loss_clip": 1.05424607,
      "balance_loss_mlp": 1.01992106,
      "epoch": 0.35074851199422835,
      "flos": 23987717775360.0,
      "grad_norm": 1.9283692803007701,
      "language_loss": 0.83072126,
      "learning_rate": 3.014580752986081e-06,
      "loss": 0.85305709,
      "num_input_tokens_seen": 62696950,
      "step": 2917,
      "time_per_iteration": 2.7782320976257324
    },
    {
      "auxiliary_loss_clip": 0.0120713,
      "auxiliary_loss_mlp": 0.0103366,
      "balance_loss_clip": 1.05225086,
      "balance_loss_mlp": 1.02482629,
      "epoch": 0.3508687548848674,
      "flos": 15224436668160.0,
      "grad_norm": 2.0522533894719373,
      "language_loss": 0.79195833,
      "learning_rate": 3.0139093776100345e-06,
      "loss": 0.81436628,
      "num_input_tokens_seen": 62713540,
      "step": 2918,
      "time_per_iteration": 2.639798402786255
    },
    {
      "auxiliary_loss_clip": 0.01194649,
      "auxiliary_loss_mlp": 0.01026068,
      "balance_loss_clip": 1.05772448,
      "balance_loss_mlp": 1.01686478,
      "epoch": 0.3509889977755065,
      "flos": 21361750185600.0,
      "grad_norm": 2.020115895610875,
      "language_loss": 0.75234473,
      "learning_rate": 3.013237848424741e-06,
      "loss": 0.77455187,
      "num_input_tokens_seen": 62732925,
      "step": 2919,
      "time_per_iteration": 2.6067237854003906
    },
    {
      "auxiliary_loss_clip": 0.01201888,
      "auxiliary_loss_mlp": 0.01029147,
      "balance_loss_clip": 1.05701447,
      "balance_loss_mlp": 1.02014709,
      "epoch": 0.35110924066614563,
      "flos": 19135360465920.0,
      "grad_norm": 2.206372486785911,
      "language_loss": 0.75540507,
      "learning_rate": 3.012566165532072e-06,
      "loss": 0.77771533,
      "num_input_tokens_seen": 62751715,
      "step": 2920,
      "time_per_iteration": 2.593824863433838
    },
    {
      "auxiliary_loss_clip": 0.01213541,
      "auxiliary_loss_mlp": 0.01030135,
      "balance_loss_clip": 1.05043316,
      "balance_loss_mlp": 1.02096725,
      "epoch": 0.3512294835567847,
      "flos": 21980885938560.0,
      "grad_norm": 3.4928735798827564,
      "language_loss": 0.76410389,
      "learning_rate": 3.0118943290339207e-06,
      "loss": 0.78654063,
      "num_input_tokens_seen": 62771925,
      "step": 2921,
      "time_per_iteration": 2.886777400970459
    },
    {
      "auxiliary_loss_clip": 0.01193026,
      "auxiliary_loss_mlp": 0.01028066,
      "balance_loss_clip": 1.04863489,
      "balance_loss_mlp": 1.0185765,
      "epoch": 0.3513497264474238,
      "flos": 17817294896640.0,
      "grad_norm": 1.987494333156327,
      "language_loss": 0.68275404,
      "learning_rate": 3.011222339032204e-06,
      "loss": 0.704965,
      "num_input_tokens_seen": 62790075,
      "step": 2922,
      "time_per_iteration": 2.6673390865325928
    },
    {
      "auxiliary_loss_clip": 0.01197282,
      "auxiliary_loss_mlp": 0.01030426,
      "balance_loss_clip": 1.06028581,
      "balance_loss_mlp": 1.02105653,
      "epoch": 0.3514699693380629,
      "flos": 26943417239040.0,
      "grad_norm": 1.9841369776447866,
      "language_loss": 0.69531608,
      "learning_rate": 3.0105501956288626e-06,
      "loss": 0.71759313,
      "num_input_tokens_seen": 62810545,
      "step": 2923,
      "time_per_iteration": 2.6115288734436035
    },
    {
      "auxiliary_loss_clip": 0.01203453,
      "auxiliary_loss_mlp": 0.01029462,
      "balance_loss_clip": 1.05576634,
      "balance_loss_mlp": 1.01956773,
      "epoch": 0.35159021222870196,
      "flos": 15267565923840.0,
      "grad_norm": 2.6619111926940286,
      "language_loss": 0.72665101,
      "learning_rate": 3.0098778989258602e-06,
      "loss": 0.74898016,
      "num_input_tokens_seen": 62829155,
      "step": 2924,
      "time_per_iteration": 4.429685831069946
    },
    {
      "auxiliary_loss_clip": 0.01199487,
      "auxiliary_loss_mlp": 0.0103334,
      "balance_loss_clip": 1.05365801,
      "balance_loss_mlp": 1.02458382,
      "epoch": 0.35171045511934107,
      "flos": 13984154000640.0,
      "grad_norm": 1.9250119641354566,
      "language_loss": 0.88559985,
      "learning_rate": 3.009205449025183e-06,
      "loss": 0.90792811,
      "num_input_tokens_seen": 62845350,
      "step": 2925,
      "time_per_iteration": 2.6672956943511963
    },
    {
      "auxiliary_loss_clip": 0.01198783,
      "auxiliary_loss_mlp": 0.01028454,
      "balance_loss_clip": 1.05086136,
      "balance_loss_mlp": 1.0191797,
      "epoch": 0.3518306980099802,
      "flos": 14283434119680.0,
      "grad_norm": 1.8584951950838287,
      "language_loss": 0.63503331,
      "learning_rate": 3.008532846028842e-06,
      "loss": 0.65730572,
      "num_input_tokens_seen": 62862110,
      "step": 2926,
      "time_per_iteration": 2.6361522674560547
    },
    {
      "auxiliary_loss_clip": 0.01200316,
      "auxiliary_loss_mlp": 0.01031955,
      "balance_loss_clip": 1.05978811,
      "balance_loss_mlp": 1.02217412,
      "epoch": 0.35195094090061924,
      "flos": 27052872958080.0,
      "grad_norm": 2.5997238192968157,
      "language_loss": 0.72593576,
      "learning_rate": 3.0078600900388694e-06,
      "loss": 0.74825847,
      "num_input_tokens_seen": 62882415,
      "step": 2927,
      "time_per_iteration": 2.6589784622192383
    },
    {
      "auxiliary_loss_clip": 0.01190758,
      "auxiliary_loss_mlp": 0.01029395,
      "balance_loss_clip": 1.04875779,
      "balance_loss_mlp": 1.02007914,
      "epoch": 0.35207118379125835,
      "flos": 25629266252160.0,
      "grad_norm": 2.028390163787808,
      "language_loss": 0.73732412,
      "learning_rate": 3.007187181157323e-06,
      "loss": 0.75952566,
      "num_input_tokens_seen": 62902425,
      "step": 2928,
      "time_per_iteration": 3.363231897354126
    },
    {
      "auxiliary_loss_clip": 0.01197808,
      "auxiliary_loss_mlp": 0.0102697,
      "balance_loss_clip": 1.04746962,
      "balance_loss_mlp": 1.01795816,
      "epoch": 0.35219142668189746,
      "flos": 18004713085440.0,
      "grad_norm": 2.2670084958790917,
      "language_loss": 0.67989671,
      "learning_rate": 3.006514119486282e-06,
      "loss": 0.7021445,
      "num_input_tokens_seen": 62919255,
      "step": 2929,
      "time_per_iteration": 2.732189416885376
    },
    {
      "auxiliary_loss_clip": 0.0119819,
      "auxiliary_loss_mlp": 0.01028329,
      "balance_loss_clip": 1.05316591,
      "balance_loss_mlp": 1.01925099,
      "epoch": 0.3523116695725365,
      "flos": 14028109269120.0,
      "grad_norm": 1.6969598861519146,
      "language_loss": 0.69938731,
      "learning_rate": 3.005840905127849e-06,
      "loss": 0.72165251,
      "num_input_tokens_seen": 62936160,
      "step": 2930,
      "time_per_iteration": 2.7220258712768555
    },
    {
      "auxiliary_loss_clip": 0.01196446,
      "auxiliary_loss_mlp": 0.01028916,
      "balance_loss_clip": 1.06010914,
      "balance_loss_mlp": 1.01962376,
      "epoch": 0.3524319124631756,
      "flos": 21433966479360.0,
      "grad_norm": 2.189946335250198,
      "language_loss": 0.8744027,
      "learning_rate": 3.0051675381841516e-06,
      "loss": 0.89665633,
      "num_input_tokens_seen": 62953470,
      "step": 2931,
      "time_per_iteration": 3.7963626384735107
    },
    {
      "auxiliary_loss_clip": 0.01208414,
      "auxiliary_loss_mlp": 0.01209593,
      "balance_loss_clip": 1.04769516,
      "balance_loss_mlp": 1.00122452,
      "epoch": 0.3525521553538147,
      "flos": 26322773114880.0,
      "grad_norm": 1.9698889977126428,
      "language_loss": 0.77431524,
      "learning_rate": 3.0044940187573363e-06,
      "loss": 0.79849529,
      "num_input_tokens_seen": 62974480,
      "step": 2932,
      "time_per_iteration": 2.8169209957122803
    },
    {
      "auxiliary_loss_clip": 0.01202553,
      "auxiliary_loss_mlp": 0.0103166,
      "balance_loss_clip": 1.05650008,
      "balance_loss_mlp": 1.02237391,
      "epoch": 0.3526723982444538,
      "flos": 21543314457600.0,
      "grad_norm": 4.862746123580322,
      "language_loss": 0.65655339,
      "learning_rate": 3.003820346949578e-06,
      "loss": 0.67889547,
      "num_input_tokens_seen": 62992560,
      "step": 2933,
      "time_per_iteration": 2.659717082977295
    },
    {
      "auxiliary_loss_clip": 0.0119846,
      "auxiliary_loss_mlp": 0.01032879,
      "balance_loss_clip": 1.05931211,
      "balance_loss_mlp": 1.02295494,
      "epoch": 0.3527926411350929,
      "flos": 23733649900800.0,
      "grad_norm": 2.000010912858895,
      "language_loss": 0.7911526,
      "learning_rate": 3.003146522863071e-06,
      "loss": 0.81346595,
      "num_input_tokens_seen": 63013445,
      "step": 2934,
      "time_per_iteration": 2.664106607437134
    },
    {
      "auxiliary_loss_clip": 0.01203921,
      "auxiliary_loss_mlp": 0.01031681,
      "balance_loss_clip": 1.06003356,
      "balance_loss_mlp": 1.02278781,
      "epoch": 0.35291288402573195,
      "flos": 30445461544320.0,
      "grad_norm": 2.163856248528957,
      "language_loss": 0.86553681,
      "learning_rate": 3.0024725466000345e-06,
      "loss": 0.88789284,
      "num_input_tokens_seen": 63033400,
      "step": 2935,
      "time_per_iteration": 2.7337465286254883
    },
    {
      "auxiliary_loss_clip": 0.01200593,
      "auxiliary_loss_mlp": 0.01026104,
      "balance_loss_clip": 1.0598824,
      "balance_loss_mlp": 1.01766407,
      "epoch": 0.35303312691637107,
      "flos": 23112179763840.0,
      "grad_norm": 1.9240094430569,
      "language_loss": 0.7931267,
      "learning_rate": 3.0017984182627087e-06,
      "loss": 0.81539369,
      "num_input_tokens_seen": 63052725,
      "step": 2936,
      "time_per_iteration": 2.6905288696289062
    },
    {
      "auxiliary_loss_clip": 0.01203388,
      "auxiliary_loss_mlp": 0.0120978,
      "balance_loss_clip": 1.05263209,
      "balance_loss_mlp": 1.00122499,
      "epoch": 0.3531533698070102,
      "flos": 21835699165440.0,
      "grad_norm": 2.025220859897167,
      "language_loss": 0.82849729,
      "learning_rate": 3.00112413795336e-06,
      "loss": 0.85262895,
      "num_input_tokens_seen": 63072560,
      "step": 2937,
      "time_per_iteration": 2.6735477447509766
    },
    {
      "auxiliary_loss_clip": 0.0119873,
      "auxiliary_loss_mlp": 0.01031637,
      "balance_loss_clip": 1.05241787,
      "balance_loss_mlp": 1.02237439,
      "epoch": 0.35327361269764923,
      "flos": 15778969810560.0,
      "grad_norm": 2.0959129144015716,
      "language_loss": 0.8005259,
      "learning_rate": 3.000449705774275e-06,
      "loss": 0.82282954,
      "num_input_tokens_seen": 63090800,
      "step": 2938,
      "time_per_iteration": 2.6626458168029785
    },
    {
      "auxiliary_loss_clip": 0.01202273,
      "auxiliary_loss_mlp": 0.0102683,
      "balance_loss_clip": 1.06028676,
      "balance_loss_mlp": 1.0174427,
      "epoch": 0.35339385558828834,
      "flos": 22090413484800.0,
      "grad_norm": 2.0334315101051814,
      "language_loss": 0.71501625,
      "learning_rate": 2.9997751218277654e-06,
      "loss": 0.73730731,
      "num_input_tokens_seen": 63108955,
      "step": 2939,
      "time_per_iteration": 2.6465022563934326
    },
    {
      "auxiliary_loss_clip": 0.01199311,
      "auxiliary_loss_mlp": 0.0102981,
      "balance_loss_clip": 1.060251,
      "balance_loss_mlp": 1.02017164,
      "epoch": 0.35351409847892745,
      "flos": 24165008328960.0,
      "grad_norm": 2.2951548395658796,
      "language_loss": 0.77916235,
      "learning_rate": 2.999100386216166e-06,
      "loss": 0.80145353,
      "num_input_tokens_seen": 63127895,
      "step": 2940,
      "time_per_iteration": 2.6027543544769287
    },
    {
      "auxiliary_loss_clip": 0.01202911,
      "auxiliary_loss_mlp": 0.01027681,
      "balance_loss_clip": 1.05707741,
      "balance_loss_mlp": 1.01865745,
      "epoch": 0.3536343413695665,
      "flos": 27052298340480.0,
      "grad_norm": 2.4179545272828786,
      "language_loss": 0.74167281,
      "learning_rate": 2.998425499041831e-06,
      "loss": 0.76397872,
      "num_input_tokens_seen": 63148410,
      "step": 2941,
      "time_per_iteration": 2.6875927448272705
    },
    {
      "auxiliary_loss_clip": 0.01117322,
      "auxiliary_loss_mlp": 0.00999907,
      "balance_loss_clip": 1.04153991,
      "balance_loss_mlp": 0.99889964,
      "epoch": 0.3537545842602056,
      "flos": 65991066370560.0,
      "grad_norm": 1.261101870879829,
      "language_loss": 0.64470947,
      "learning_rate": 2.997750460407142e-06,
      "loss": 0.66588175,
      "num_input_tokens_seen": 63209765,
      "step": 2942,
      "time_per_iteration": 3.355818033218384
    },
    {
      "auxiliary_loss_clip": 0.01207046,
      "auxiliary_loss_mlp": 0.01027571,
      "balance_loss_clip": 1.05205846,
      "balance_loss_mlp": 1.01801062,
      "epoch": 0.35387482715084473,
      "flos": 18436897526400.0,
      "grad_norm": 2.135057911491607,
      "language_loss": 0.70560437,
      "learning_rate": 2.997075270414501e-06,
      "loss": 0.72795045,
      "num_input_tokens_seen": 63226980,
      "step": 2943,
      "time_per_iteration": 2.754969358444214
    },
    {
      "auxiliary_loss_clip": 0.01118949,
      "auxiliary_loss_mlp": 0.00999644,
      "balance_loss_clip": 1.03918648,
      "balance_loss_mlp": 0.998541,
      "epoch": 0.3539950700414838,
      "flos": 65588579498880.0,
      "grad_norm": 0.7039975709133965,
      "language_loss": 0.57679152,
      "learning_rate": 2.9963999291663347e-06,
      "loss": 0.5979774,
      "num_input_tokens_seen": 63292760,
      "step": 2944,
      "time_per_iteration": 3.2483508586883545
    },
    {
      "auxiliary_loss_clip": 0.01208747,
      "auxiliary_loss_mlp": 0.01033834,
      "balance_loss_clip": 1.05541229,
      "balance_loss_mlp": 1.02531099,
      "epoch": 0.3541153129321229,
      "flos": 20521655919360.0,
      "grad_norm": 2.51548208633407,
      "language_loss": 0.73851371,
      "learning_rate": 2.9957244367650915e-06,
      "loss": 0.76093954,
      "num_input_tokens_seen": 63309005,
      "step": 2945,
      "time_per_iteration": 2.6799116134643555
    },
    {
      "auxiliary_loss_clip": 0.01200864,
      "auxiliary_loss_mlp": 0.01031353,
      "balance_loss_clip": 1.05288768,
      "balance_loss_mlp": 1.02164888,
      "epoch": 0.354235555822762,
      "flos": 19573578391680.0,
      "grad_norm": 1.8578970527375824,
      "language_loss": 0.83898276,
      "learning_rate": 2.9950487933132425e-06,
      "loss": 0.86130488,
      "num_input_tokens_seen": 63326420,
      "step": 2946,
      "time_per_iteration": 2.6512820720672607
    },
    {
      "auxiliary_loss_clip": 0.01204443,
      "auxiliary_loss_mlp": 0.01032824,
      "balance_loss_clip": 1.05747485,
      "balance_loss_mlp": 1.02381146,
      "epoch": 0.35435579871340106,
      "flos": 20777268078720.0,
      "grad_norm": 2.1120233615864366,
      "language_loss": 0.71357155,
      "learning_rate": 2.994372998913283e-06,
      "loss": 0.73594415,
      "num_input_tokens_seen": 63344925,
      "step": 2947,
      "time_per_iteration": 2.6975739002227783
    },
    {
      "auxiliary_loss_clip": 0.01206375,
      "auxiliary_loss_mlp": 0.01029797,
      "balance_loss_clip": 1.06152952,
      "balance_loss_mlp": 1.02090967,
      "epoch": 0.35447604160404017,
      "flos": 23951807153280.0,
      "grad_norm": 2.182847081586397,
      "language_loss": 0.62706423,
      "learning_rate": 2.99369705366773e-06,
      "loss": 0.64942598,
      "num_input_tokens_seen": 63365170,
      "step": 2948,
      "time_per_iteration": 2.653318166732788
    },
    {
      "auxiliary_loss_clip": 0.01201021,
      "auxiliary_loss_mlp": 0.01025982,
      "balance_loss_clip": 1.05781937,
      "balance_loss_mlp": 1.01725626,
      "epoch": 0.3545962844946792,
      "flos": 23435662671360.0,
      "grad_norm": 2.148707754890999,
      "language_loss": 0.82469773,
      "learning_rate": 2.9930209576791244e-06,
      "loss": 0.8469677,
      "num_input_tokens_seen": 63383645,
      "step": 2949,
      "time_per_iteration": 2.688488006591797
    },
    {
      "auxiliary_loss_clip": 0.011974,
      "auxiliary_loss_mlp": 0.01029763,
      "balance_loss_clip": 1.05708671,
      "balance_loss_mlp": 1.02129889,
      "epoch": 0.35471652738531834,
      "flos": 22085134185600.0,
      "grad_norm": 2.183544002464086,
      "language_loss": 0.63872713,
      "learning_rate": 2.9923447110500285e-06,
      "loss": 0.66099876,
      "num_input_tokens_seen": 63402390,
      "step": 2950,
      "time_per_iteration": 2.6304399967193604
    },
    {
      "auxiliary_loss_clip": 0.01187873,
      "auxiliary_loss_mlp": 0.01032471,
      "balance_loss_clip": 1.05545282,
      "balance_loss_mlp": 1.02353621,
      "epoch": 0.35483677027595745,
      "flos": 27341881787520.0,
      "grad_norm": 1.7436617265545538,
      "language_loss": 0.756648,
      "learning_rate": 2.9916683138830295e-06,
      "loss": 0.77885145,
      "num_input_tokens_seen": 63423055,
      "step": 2951,
      "time_per_iteration": 4.546418905258179
    },
    {
      "auxiliary_loss_clip": 0.01198531,
      "auxiliary_loss_mlp": 0.01033022,
      "balance_loss_clip": 1.05654311,
      "balance_loss_mlp": 1.02312183,
      "epoch": 0.3549570131665965,
      "flos": 13516166678400.0,
      "grad_norm": 3.383272787512526,
      "language_loss": 0.81033224,
      "learning_rate": 2.9909917662807353e-06,
      "loss": 0.8326478,
      "num_input_tokens_seen": 63440855,
      "step": 2952,
      "time_per_iteration": 2.774928569793701
    },
    {
      "auxiliary_loss_clip": 0.01196433,
      "auxiliary_loss_mlp": 0.01034074,
      "balance_loss_clip": 1.05564642,
      "balance_loss_mlp": 1.02450109,
      "epoch": 0.3550772560572356,
      "flos": 20887549810560.0,
      "grad_norm": 2.6004108806183646,
      "language_loss": 0.6944809,
      "learning_rate": 2.9903150683457783e-06,
      "loss": 0.71678591,
      "num_input_tokens_seen": 63459400,
      "step": 2953,
      "time_per_iteration": 2.7134175300598145
    },
    {
      "auxiliary_loss_clip": 0.01201878,
      "auxiliary_loss_mlp": 0.0102401,
      "balance_loss_clip": 1.05429423,
      "balance_loss_mlp": 1.01551652,
      "epoch": 0.3551974989478747,
      "flos": 20194042947840.0,
      "grad_norm": 1.8943258360039248,
      "language_loss": 0.6494354,
      "learning_rate": 2.9896382201808126e-06,
      "loss": 0.67169434,
      "num_input_tokens_seen": 63476800,
      "step": 2954,
      "time_per_iteration": 2.6629786491394043
    },
    {
      "auxiliary_loss_clip": 0.01199249,
      "auxiliary_loss_mlp": 0.01031003,
      "balance_loss_clip": 1.05931723,
      "balance_loss_mlp": 1.02167463,
      "epoch": 0.3553177418385138,
      "flos": 19828831415040.0,
      "grad_norm": 2.2393579789315474,
      "language_loss": 0.81015146,
      "learning_rate": 2.988961221888516e-06,
      "loss": 0.83245397,
      "num_input_tokens_seen": 63493475,
      "step": 2955,
      "time_per_iteration": 3.5072410106658936
    },
    {
      "auxiliary_loss_clip": 0.01190708,
      "auxiliary_loss_mlp": 0.01028056,
      "balance_loss_clip": 1.05003726,
      "balance_loss_mlp": 1.01891828,
      "epoch": 0.3554379847291529,
      "flos": 14829132516480.0,
      "grad_norm": 3.8757604870879923,
      "language_loss": 0.79118013,
      "learning_rate": 2.988284073571589e-06,
      "loss": 0.81336778,
      "num_input_tokens_seen": 63509560,
      "step": 2956,
      "time_per_iteration": 2.8232192993164062
    },
    {
      "auxiliary_loss_clip": 0.01203726,
      "auxiliary_loss_mlp": 0.01208993,
      "balance_loss_clip": 1.060004,
      "balance_loss_mlp": 1.00107265,
      "epoch": 0.355558227619792,
      "flos": 20485350247680.0,
      "grad_norm": 6.733630669528753,
      "language_loss": 0.73227131,
      "learning_rate": 2.9876067753327528e-06,
      "loss": 0.75639856,
      "num_input_tokens_seen": 63527290,
      "step": 2957,
      "time_per_iteration": 2.657180070877075
    },
    {
      "auxiliary_loss_clip": 0.01201963,
      "auxiliary_loss_mlp": 0.01035861,
      "balance_loss_clip": 1.05660939,
      "balance_loss_mlp": 1.02578831,
      "epoch": 0.35567847051043106,
      "flos": 37663613256960.0,
      "grad_norm": 2.0521746257026714,
      "language_loss": 0.80686146,
      "learning_rate": 2.986929327274754e-06,
      "loss": 0.82923973,
      "num_input_tokens_seen": 63547870,
      "step": 2958,
      "time_per_iteration": 3.739422559738159
    },
    {
      "auxiliary_loss_clip": 0.01198925,
      "auxiliary_loss_mlp": 0.01032216,
      "balance_loss_clip": 1.05859113,
      "balance_loss_mlp": 1.02307272,
      "epoch": 0.35579871340107017,
      "flos": 26943058103040.0,
      "grad_norm": 1.5985133181680722,
      "language_loss": 0.79130816,
      "learning_rate": 2.9862517295003617e-06,
      "loss": 0.81361961,
      "num_input_tokens_seen": 63568285,
      "step": 2959,
      "time_per_iteration": 2.666299343109131
    },
    {
      "auxiliary_loss_clip": 0.01200574,
      "auxiliary_loss_mlp": 0.01029456,
      "balance_loss_clip": 1.05121326,
      "balance_loss_mlp": 1.02037227,
      "epoch": 0.3559189562917093,
      "flos": 28293335193600.0,
      "grad_norm": 1.7376731782233508,
      "language_loss": 0.72732699,
      "learning_rate": 2.9855739821123654e-06,
      "loss": 0.74962723,
      "num_input_tokens_seen": 63589865,
      "step": 2960,
      "time_per_iteration": 2.749652147293091
    },
    {
      "auxiliary_loss_clip": 0.01196709,
      "auxiliary_loss_mlp": 0.0102886,
      "balance_loss_clip": 1.0585711,
      "balance_loss_mlp": 1.01921558,
      "epoch": 0.35603919918234833,
      "flos": 25664063552640.0,
      "grad_norm": 1.76653779360559,
      "language_loss": 0.82314891,
      "learning_rate": 2.98489608521358e-06,
      "loss": 0.84540462,
      "num_input_tokens_seen": 63609805,
      "step": 2961,
      "time_per_iteration": 2.677912712097168
    },
    {
      "auxiliary_loss_clip": 0.0120589,
      "auxiliary_loss_mlp": 0.01209083,
      "balance_loss_clip": 1.05881321,
      "balance_loss_mlp": 1.00111496,
      "epoch": 0.35615944207298744,
      "flos": 23000856537600.0,
      "grad_norm": 2.0135219503386246,
      "language_loss": 0.79444104,
      "learning_rate": 2.9842180389068425e-06,
      "loss": 0.81859076,
      "num_input_tokens_seen": 63627115,
      "step": 2962,
      "time_per_iteration": 2.683258056640625
    },
    {
      "auxiliary_loss_clip": 0.01126395,
      "auxiliary_loss_mlp": 0.0100352,
      "balance_loss_clip": 1.04456639,
      "balance_loss_mlp": 1.00203025,
      "epoch": 0.35627968496362655,
      "flos": 68251283723520.0,
      "grad_norm": 0.761492453583856,
      "language_loss": 0.59187877,
      "learning_rate": 2.98353984329501e-06,
      "loss": 0.6131779,
      "num_input_tokens_seen": 63691460,
      "step": 2963,
      "time_per_iteration": 3.2240331172943115
    },
    {
      "auxiliary_loss_clip": 0.0120362,
      "auxiliary_loss_mlp": 0.01030358,
      "balance_loss_clip": 1.0587095,
      "balance_loss_mlp": 1.02083349,
      "epoch": 0.3563999278542656,
      "flos": 22641714403200.0,
      "grad_norm": 1.7846489424905534,
      "language_loss": 0.71062118,
      "learning_rate": 2.982861498480965e-06,
      "loss": 0.73296094,
      "num_input_tokens_seen": 63713840,
      "step": 2964,
      "time_per_iteration": 2.7723028659820557
    },
    {
      "auxiliary_loss_clip": 0.01198243,
      "auxiliary_loss_mlp": 0.01030473,
      "balance_loss_clip": 1.05013514,
      "balance_loss_mlp": 1.02132344,
      "epoch": 0.3565201707449047,
      "flos": 25952533678080.0,
      "grad_norm": 1.8981849037247311,
      "language_loss": 0.82602394,
      "learning_rate": 2.9821830045676122e-06,
      "loss": 0.84831113,
      "num_input_tokens_seen": 63733540,
      "step": 2965,
      "time_per_iteration": 2.718961000442505
    },
    {
      "auxiliary_loss_clip": 0.01200673,
      "auxiliary_loss_mlp": 0.01032135,
      "balance_loss_clip": 1.06113434,
      "balance_loss_mlp": 1.02333164,
      "epoch": 0.3566404136355438,
      "flos": 28475725478400.0,
      "grad_norm": 2.1222083197347326,
      "language_loss": 0.72998536,
      "learning_rate": 2.9815043616578793e-06,
      "loss": 0.75231349,
      "num_input_tokens_seen": 63754335,
      "step": 2966,
      "time_per_iteration": 2.7070868015289307
    },
    {
      "auxiliary_loss_clip": 0.0119968,
      "auxiliary_loss_mlp": 0.01034936,
      "balance_loss_clip": 1.05129945,
      "balance_loss_mlp": 1.02590036,
      "epoch": 0.3567606565261829,
      "flos": 38363117690880.0,
      "grad_norm": 2.1769402080478026,
      "language_loss": 0.76962805,
      "learning_rate": 2.9808255698547145e-06,
      "loss": 0.79197431,
      "num_input_tokens_seen": 63777135,
      "step": 2967,
      "time_per_iteration": 2.8792364597320557
    },
    {
      "auxiliary_loss_clip": 0.01199942,
      "auxiliary_loss_mlp": 0.0102731,
      "balance_loss_clip": 1.05983639,
      "balance_loss_mlp": 1.018363,
      "epoch": 0.356880899416822,
      "flos": 21981029592960.0,
      "grad_norm": 3.608792929984078,
      "language_loss": 0.7898097,
      "learning_rate": 2.9801466292610913e-06,
      "loss": 0.81208229,
      "num_input_tokens_seen": 63797020,
      "step": 2968,
      "time_per_iteration": 2.667006492614746
    },
    {
      "auxiliary_loss_clip": 0.0119679,
      "auxiliary_loss_mlp": 0.01026424,
      "balance_loss_clip": 1.05626464,
      "balance_loss_mlp": 1.01766825,
      "epoch": 0.35700114230746105,
      "flos": 18989132198400.0,
      "grad_norm": 2.162740434217853,
      "language_loss": 0.81469798,
      "learning_rate": 2.979467539980003e-06,
      "loss": 0.8369301,
      "num_input_tokens_seen": 63813810,
      "step": 2969,
      "time_per_iteration": 2.6728224754333496
    },
    {
      "auxiliary_loss_clip": 0.01201596,
      "auxiliary_loss_mlp": 0.01032031,
      "balance_loss_clip": 1.05825043,
      "balance_loss_mlp": 1.02338219,
      "epoch": 0.35712138519810016,
      "flos": 19756112330880.0,
      "grad_norm": 4.436343592405346,
      "language_loss": 0.76717347,
      "learning_rate": 2.978788302114468e-06,
      "loss": 0.78950977,
      "num_input_tokens_seen": 63830925,
      "step": 2970,
      "time_per_iteration": 2.5918915271759033
    },
    {
      "auxiliary_loss_clip": 0.0119733,
      "auxiliary_loss_mlp": 0.01031871,
      "balance_loss_clip": 1.05708194,
      "balance_loss_mlp": 1.02229261,
      "epoch": 0.35724162808873927,
      "flos": 35183012008320.0,
      "grad_norm": 3.6453490371127666,
      "language_loss": 0.83280075,
      "learning_rate": 2.9781089157675255e-06,
      "loss": 0.85509276,
      "num_input_tokens_seen": 63849385,
      "step": 2971,
      "time_per_iteration": 2.6703743934631348
    },
    {
      "auxiliary_loss_clip": 0.01196,
      "auxiliary_loss_mlp": 0.01030037,
      "balance_loss_clip": 1.05978417,
      "balance_loss_mlp": 1.0209713,
      "epoch": 0.3573618709793783,
      "flos": 25556726736000.0,
      "grad_norm": 1.4763913277517196,
      "language_loss": 0.88509566,
      "learning_rate": 2.977429381042238e-06,
      "loss": 0.90735602,
      "num_input_tokens_seen": 63870060,
      "step": 2972,
      "time_per_iteration": 2.628615617752075
    },
    {
      "auxiliary_loss_clip": 0.01199414,
      "auxiliary_loss_mlp": 0.01029195,
      "balance_loss_clip": 1.0545373,
      "balance_loss_mlp": 1.0210228,
      "epoch": 0.35748211387001744,
      "flos": 29132352051840.0,
      "grad_norm": 2.6147421532315143,
      "language_loss": 0.89381862,
      "learning_rate": 2.9767496980416913e-06,
      "loss": 0.91610473,
      "num_input_tokens_seen": 63889355,
      "step": 2973,
      "time_per_iteration": 2.6917781829833984
    },
    {
      "auxiliary_loss_clip": 0.0119651,
      "auxiliary_loss_mlp": 0.0103038,
      "balance_loss_clip": 1.05336189,
      "balance_loss_mlp": 1.0207299,
      "epoch": 0.35760235676065655,
      "flos": 13954169122560.0,
      "grad_norm": 2.907328465164619,
      "language_loss": 0.82134891,
      "learning_rate": 2.9760698668689914e-06,
      "loss": 0.84361774,
      "num_input_tokens_seen": 63905580,
      "step": 2974,
      "time_per_iteration": 2.5707106590270996
    },
    {
      "auxiliary_loss_clip": 0.01199796,
      "auxiliary_loss_mlp": 0.01023991,
      "balance_loss_clip": 1.0565238,
      "balance_loss_mlp": 1.01540852,
      "epoch": 0.3577225996512956,
      "flos": 44018688977280.0,
      "grad_norm": 2.3766773168967563,
      "language_loss": 0.71535581,
      "learning_rate": 2.975389887627269e-06,
      "loss": 0.73759371,
      "num_input_tokens_seen": 63928180,
      "step": 2975,
      "time_per_iteration": 2.8579416275024414
    },
    {
      "auxiliary_loss_clip": 0.01207202,
      "auxiliary_loss_mlp": 0.01028744,
      "balance_loss_clip": 1.05527592,
      "balance_loss_mlp": 1.02043557,
      "epoch": 0.3578428425419347,
      "flos": 17055199013760.0,
      "grad_norm": 1.991763566397909,
      "language_loss": 0.89575988,
      "learning_rate": 2.9747097604196764e-06,
      "loss": 0.91811937,
      "num_input_tokens_seen": 63944825,
      "step": 2976,
      "time_per_iteration": 2.632570266723633
    },
    {
      "auxiliary_loss_clip": 0.01125854,
      "auxiliary_loss_mlp": 0.009992,
      "balance_loss_clip": 1.03162885,
      "balance_loss_mlp": 0.99825257,
      "epoch": 0.3579630854325738,
      "flos": 71676550707840.0,
      "grad_norm": 0.6733100637685502,
      "language_loss": 0.56578261,
      "learning_rate": 2.9740294853493875e-06,
      "loss": 0.58703315,
      "num_input_tokens_seen": 64016385,
      "step": 2977,
      "time_per_iteration": 4.459169864654541
    },
    {
      "auxiliary_loss_clip": 0.01208131,
      "auxiliary_loss_mlp": 0.01029347,
      "balance_loss_clip": 1.05165839,
      "balance_loss_mlp": 1.0208714,
      "epoch": 0.3580833283232129,
      "flos": 25046651652480.0,
      "grad_norm": 2.4115451737888627,
      "language_loss": 0.66945064,
      "learning_rate": 2.9733490625196008e-06,
      "loss": 0.69182545,
      "num_input_tokens_seen": 64036245,
      "step": 2978,
      "time_per_iteration": 3.6728084087371826
    },
    {
      "auxiliary_loss_clip": 0.01189642,
      "auxiliary_loss_mlp": 0.01027815,
      "balance_loss_clip": 1.05093467,
      "balance_loss_mlp": 1.01992905,
      "epoch": 0.358203571213852,
      "flos": 13953127628160.0,
      "grad_norm": 3.463483553794683,
      "language_loss": 0.76098609,
      "learning_rate": 2.9726684920335353e-06,
      "loss": 0.78316069,
      "num_input_tokens_seen": 64054110,
      "step": 2979,
      "time_per_iteration": 2.6626598834991455
    },
    {
      "auxiliary_loss_clip": 0.01200174,
      "auxiliary_loss_mlp": 0.0120948,
      "balance_loss_clip": 1.059057,
      "balance_loss_mlp": 1.00107622,
      "epoch": 0.35832381410449105,
      "flos": 20302457172480.0,
      "grad_norm": 2.172319257936834,
      "language_loss": 0.82099098,
      "learning_rate": 2.971987773994432e-06,
      "loss": 0.84508753,
      "num_input_tokens_seen": 64070295,
      "step": 2980,
      "time_per_iteration": 2.662576675415039
    },
    {
      "auxiliary_loss_clip": 0.01187787,
      "auxiliary_loss_mlp": 0.01024159,
      "balance_loss_clip": 1.05229402,
      "balance_loss_mlp": 1.01551676,
      "epoch": 0.35844405699513016,
      "flos": 16983234115200.0,
      "grad_norm": 1.956762745703048,
      "language_loss": 0.82999855,
      "learning_rate": 2.9713069085055566e-06,
      "loss": 0.85211802,
      "num_input_tokens_seen": 64088605,
      "step": 2981,
      "time_per_iteration": 3.534566879272461
    },
    {
      "auxiliary_loss_clip": 0.01199943,
      "auxiliary_loss_mlp": 0.01027269,
      "balance_loss_clip": 1.05233359,
      "balance_loss_mlp": 1.01857305,
      "epoch": 0.35856429988576927,
      "flos": 23216858974080.0,
      "grad_norm": 2.673463290426677,
      "language_loss": 0.79322869,
      "learning_rate": 2.9706258956701958e-06,
      "loss": 0.81550086,
      "num_input_tokens_seen": 64108595,
      "step": 2982,
      "time_per_iteration": 2.686495065689087
    },
    {
      "auxiliary_loss_clip": 0.01201226,
      "auxiliary_loss_mlp": 0.01033377,
      "balance_loss_clip": 1.05712938,
      "balance_loss_mlp": 1.02456689,
      "epoch": 0.3586845427764083,
      "flos": 23034576430080.0,
      "grad_norm": 2.6819266037553486,
      "language_loss": 0.77715254,
      "learning_rate": 2.9699447355916575e-06,
      "loss": 0.79949856,
      "num_input_tokens_seen": 64127405,
      "step": 2983,
      "time_per_iteration": 2.6647965908050537
    },
    {
      "auxiliary_loss_clip": 0.0119376,
      "auxiliary_loss_mlp": 0.01208659,
      "balance_loss_clip": 1.05682325,
      "balance_loss_mlp": 1.00093961,
      "epoch": 0.35880478566704743,
      "flos": 20010682995840.0,
      "grad_norm": 2.6801793804631457,
      "language_loss": 0.74564612,
      "learning_rate": 2.969263428373275e-06,
      "loss": 0.76967031,
      "num_input_tokens_seen": 64145755,
      "step": 2984,
      "time_per_iteration": 2.580875873565674
    },
    {
      "auxiliary_loss_clip": 0.01201096,
      "auxiliary_loss_mlp": 0.01029416,
      "balance_loss_clip": 1.05383241,
      "balance_loss_mlp": 1.02065396,
      "epoch": 0.35892502855768654,
      "flos": 13699095667200.0,
      "grad_norm": 1.9797333343853534,
      "language_loss": 0.79452872,
      "learning_rate": 2.9685819741184007e-06,
      "loss": 0.81683385,
      "num_input_tokens_seen": 64164195,
      "step": 2985,
      "time_per_iteration": 3.6297500133514404
    },
    {
      "auxiliary_loss_clip": 0.01195405,
      "auxiliary_loss_mlp": 0.01031445,
      "balance_loss_clip": 1.05268848,
      "balance_loss_mlp": 1.02286172,
      "epoch": 0.3590452714483256,
      "flos": 18114096977280.0,
      "grad_norm": 2.956771006226869,
      "language_loss": 0.68955004,
      "learning_rate": 2.967900372930411e-06,
      "loss": 0.71181858,
      "num_input_tokens_seen": 64182705,
      "step": 2986,
      "time_per_iteration": 2.6773619651794434
    },
    {
      "auxiliary_loss_clip": 0.01195464,
      "auxiliary_loss_mlp": 0.01033082,
      "balance_loss_clip": 1.0547266,
      "balance_loss_mlp": 1.02372396,
      "epoch": 0.3591655143389647,
      "flos": 17749352321280.0,
      "grad_norm": 2.7500466821308165,
      "language_loss": 0.79268903,
      "learning_rate": 2.9672186249127046e-06,
      "loss": 0.81497449,
      "num_input_tokens_seen": 64202170,
      "step": 2987,
      "time_per_iteration": 2.6108779907226562
    },
    {
      "auxiliary_loss_clip": 0.01199296,
      "auxiliary_loss_mlp": 0.01034272,
      "balance_loss_clip": 1.05567455,
      "balance_loss_mlp": 1.02545071,
      "epoch": 0.3592857572296038,
      "flos": 25224409082880.0,
      "grad_norm": 5.612386309290006,
      "language_loss": 0.79071575,
      "learning_rate": 2.9665367301687014e-06,
      "loss": 0.81305146,
      "num_input_tokens_seen": 64220415,
      "step": 2988,
      "time_per_iteration": 2.6908066272735596
    },
    {
      "auxiliary_loss_clip": 0.01191887,
      "auxiliary_loss_mlp": 0.0102908,
      "balance_loss_clip": 1.05408192,
      "balance_loss_mlp": 1.02031255,
      "epoch": 0.3594060001202429,
      "flos": 29384408764800.0,
      "grad_norm": 2.1169344874036335,
      "language_loss": 0.76768225,
      "learning_rate": 2.965854688801845e-06,
      "loss": 0.78989196,
      "num_input_tokens_seen": 64242475,
      "step": 2989,
      "time_per_iteration": 2.681300163269043
    },
    {
      "auxiliary_loss_clip": 0.01193315,
      "auxiliary_loss_mlp": 0.01027263,
      "balance_loss_clip": 1.05170751,
      "balance_loss_mlp": 1.01900148,
      "epoch": 0.359526243010882,
      "flos": 17052900543360.0,
      "grad_norm": 1.9702994205886522,
      "language_loss": 0.76929522,
      "learning_rate": 2.9651725009156005e-06,
      "loss": 0.79150099,
      "num_input_tokens_seen": 64260220,
      "step": 2990,
      "time_per_iteration": 2.6266860961914062
    },
    {
      "auxiliary_loss_clip": 0.01193095,
      "auxiliary_loss_mlp": 0.01032761,
      "balance_loss_clip": 1.05239129,
      "balance_loss_mlp": 1.02296758,
      "epoch": 0.3596464859015211,
      "flos": 22965089569920.0,
      "grad_norm": 2.507024893652408,
      "language_loss": 0.74710613,
      "learning_rate": 2.964490166613454e-06,
      "loss": 0.76936471,
      "num_input_tokens_seen": 64280145,
      "step": 2991,
      "time_per_iteration": 2.670149803161621
    },
    {
      "auxiliary_loss_clip": 0.01103717,
      "auxiliary_loss_mlp": 0.01004988,
      "balance_loss_clip": 1.03283787,
      "balance_loss_mlp": 1.00400436,
      "epoch": 0.35976672879216015,
      "flos": 54739462590720.0,
      "grad_norm": 0.7596050402301656,
      "language_loss": 0.57708454,
      "learning_rate": 2.963807685998917e-06,
      "loss": 0.59817159,
      "num_input_tokens_seen": 64336010,
      "step": 2992,
      "time_per_iteration": 3.028960704803467
    },
    {
      "auxiliary_loss_clip": 0.01208216,
      "auxiliary_loss_mlp": 0.01027422,
      "balance_loss_clip": 1.05048919,
      "balance_loss_mlp": 1.01860094,
      "epoch": 0.35988697168279926,
      "flos": 43139020901760.0,
      "grad_norm": 1.5358837904223006,
      "language_loss": 0.78086281,
      "learning_rate": 2.9631250591755196e-06,
      "loss": 0.8032192,
      "num_input_tokens_seen": 64358725,
      "step": 2993,
      "time_per_iteration": 2.9168484210968018
    },
    {
      "auxiliary_loss_clip": 0.01192944,
      "auxiliary_loss_mlp": 0.01032164,
      "balance_loss_clip": 1.05423021,
      "balance_loss_mlp": 1.02290177,
      "epoch": 0.36000721457343837,
      "flos": 35845600239360.0,
      "grad_norm": 1.8092899723429796,
      "language_loss": 0.57816303,
      "learning_rate": 2.962442286246817e-06,
      "loss": 0.6004141,
      "num_input_tokens_seen": 64381555,
      "step": 2994,
      "time_per_iteration": 2.8349781036376953
    },
    {
      "auxiliary_loss_clip": 0.01202426,
      "auxiliary_loss_mlp": 0.01026534,
      "balance_loss_clip": 1.05418837,
      "balance_loss_mlp": 1.01793337,
      "epoch": 0.3601274574640774,
      "flos": 18291100222080.0,
      "grad_norm": 1.6350568968710126,
      "language_loss": 0.69776988,
      "learning_rate": 2.9617593673163853e-06,
      "loss": 0.72005951,
      "num_input_tokens_seen": 64400375,
      "step": 2995,
      "time_per_iteration": 2.6770455837249756
    },
    {
      "auxiliary_loss_clip": 0.01200821,
      "auxiliary_loss_mlp": 0.01022755,
      "balance_loss_clip": 1.05120492,
      "balance_loss_mlp": 1.01489353,
      "epoch": 0.36024770035471654,
      "flos": 13333955961600.0,
      "grad_norm": 2.3954726036128844,
      "language_loss": 0.77471507,
      "learning_rate": 2.9610763024878216e-06,
      "loss": 0.79695082,
      "num_input_tokens_seen": 64415880,
      "step": 2996,
      "time_per_iteration": 2.6158881187438965
    },
    {
      "auxiliary_loss_clip": 0.01193204,
      "auxiliary_loss_mlp": 0.01034364,
      "balance_loss_clip": 1.05299151,
      "balance_loss_mlp": 1.02536988,
      "epoch": 0.3603679432453556,
      "flos": 20267013427200.0,
      "grad_norm": 1.8126097704163662,
      "language_loss": 0.9174546,
      "learning_rate": 2.960393091864747e-06,
      "loss": 0.93973029,
      "num_input_tokens_seen": 64434260,
      "step": 2997,
      "time_per_iteration": 2.665977954864502
    },
    {
      "auxiliary_loss_clip": 0.01200486,
      "auxiliary_loss_mlp": 0.01020744,
      "balance_loss_clip": 1.05551219,
      "balance_loss_mlp": 1.01228023,
      "epoch": 0.3604881861359947,
      "flos": 22451135817600.0,
      "grad_norm": 5.800990963246788,
      "language_loss": 0.74834377,
      "learning_rate": 2.959709735550804e-06,
      "loss": 0.77055609,
      "num_input_tokens_seen": 64453855,
      "step": 2998,
      "time_per_iteration": 2.6418187618255615
    },
    {
      "auxiliary_loss_clip": 0.01207114,
      "auxiliary_loss_mlp": 0.01022529,
      "balance_loss_clip": 1.05208111,
      "balance_loss_mlp": 1.01402974,
      "epoch": 0.3606084290266338,
      "flos": 22054251467520.0,
      "grad_norm": 2.1576617948176295,
      "language_loss": 0.75409436,
      "learning_rate": 2.9590262336496575e-06,
      "loss": 0.77639085,
      "num_input_tokens_seen": 64473585,
      "step": 2999,
      "time_per_iteration": 2.7375094890594482
    },
    {
      "auxiliary_loss_clip": 0.01199433,
      "auxiliary_loss_mlp": 0.01035779,
      "balance_loss_clip": 1.0554812,
      "balance_loss_mlp": 1.02642751,
      "epoch": 0.36072867191727287,
      "flos": 15632921111040.0,
      "grad_norm": 2.21004840684637,
      "language_loss": 0.85412526,
      "learning_rate": 2.9583425862649936e-06,
      "loss": 0.87647736,
      "num_input_tokens_seen": 64491720,
      "step": 3000,
      "time_per_iteration": 2.6098570823669434
    },
    {
      "auxiliary_loss_clip": 0.01197322,
      "auxiliary_loss_mlp": 0.01030025,
      "balance_loss_clip": 1.05883896,
      "balance_loss_mlp": 1.02125704,
      "epoch": 0.360848914807912,
      "flos": 19677000625920.0,
      "grad_norm": 2.5708287920208526,
      "language_loss": 0.73997784,
      "learning_rate": 2.9576587935005215e-06,
      "loss": 0.76225126,
      "num_input_tokens_seen": 64509800,
      "step": 3001,
      "time_per_iteration": 2.640587329864502
    },
    {
      "auxiliary_loss_clip": 0.01198396,
      "auxiliary_loss_mlp": 0.01024219,
      "balance_loss_clip": 1.05442441,
      "balance_loss_mlp": 1.01520061,
      "epoch": 0.3609691576985511,
      "flos": 18877808972160.0,
      "grad_norm": 4.127551815830875,
      "language_loss": 0.72168219,
      "learning_rate": 2.9569748554599713e-06,
      "loss": 0.74390835,
      "num_input_tokens_seen": 64525410,
      "step": 3002,
      "time_per_iteration": 2.6179656982421875
    },
    {
      "auxiliary_loss_clip": 0.01197904,
      "auxiliary_loss_mlp": 0.01032318,
      "balance_loss_clip": 1.05321765,
      "balance_loss_mlp": 1.02380085,
      "epoch": 0.36108940058919015,
      "flos": 42224088648960.0,
      "grad_norm": 1.9519568255069135,
      "language_loss": 0.73274499,
      "learning_rate": 2.956290772247097e-06,
      "loss": 0.7550472,
      "num_input_tokens_seen": 64544085,
      "step": 3003,
      "time_per_iteration": 2.848266363143921
    },
    {
      "auxiliary_loss_clip": 0.01194567,
      "auxiliary_loss_mlp": 0.0102917,
      "balance_loss_clip": 1.05007851,
      "balance_loss_mlp": 1.02086735,
      "epoch": 0.36120964347982926,
      "flos": 23185150243200.0,
      "grad_norm": 1.7659998344124184,
      "language_loss": 0.73106933,
      "learning_rate": 2.9556065439656724e-06,
      "loss": 0.75330669,
      "num_input_tokens_seen": 64563135,
      "step": 3004,
      "time_per_iteration": 3.668353319168091
    },
    {
      "auxiliary_loss_clip": 0.0119748,
      "auxiliary_loss_mlp": 0.01028487,
      "balance_loss_clip": 1.04552221,
      "balance_loss_mlp": 1.01974893,
      "epoch": 0.36132988637046837,
      "flos": 18113055482880.0,
      "grad_norm": 1.7242443708427495,
      "language_loss": 0.81789362,
      "learning_rate": 2.9549221707194952e-06,
      "loss": 0.84015334,
      "num_input_tokens_seen": 64581985,
      "step": 3005,
      "time_per_iteration": 2.76631498336792
    },
    {
      "auxiliary_loss_clip": 0.01197789,
      "auxiliary_loss_mlp": 0.01026534,
      "balance_loss_clip": 1.05614662,
      "balance_loss_mlp": 1.01805258,
      "epoch": 0.3614501292611074,
      "flos": 27813101333760.0,
      "grad_norm": 3.339797057370517,
      "language_loss": 0.7282306,
      "learning_rate": 2.954237652612384e-06,
      "loss": 0.75047386,
      "num_input_tokens_seen": 64601035,
      "step": 3006,
      "time_per_iteration": 2.781271457672119
    },
    {
      "auxiliary_loss_clip": 0.01194585,
      "auxiliary_loss_mlp": 0.01025541,
      "balance_loss_clip": 1.05207086,
      "balance_loss_mlp": 1.01711869,
      "epoch": 0.36157037215174653,
      "flos": 22634926732800.0,
      "grad_norm": 2.4650468682407114,
      "language_loss": 0.84993708,
      "learning_rate": 2.9535529897481796e-06,
      "loss": 0.87213838,
      "num_input_tokens_seen": 64618580,
      "step": 3007,
      "time_per_iteration": 2.636805295944214
    },
    {
      "auxiliary_loss_clip": 0.01194927,
      "auxiliary_loss_mlp": 0.01026264,
      "balance_loss_clip": 1.05594635,
      "balance_loss_mlp": 1.01766324,
      "epoch": 0.36169061504238564,
      "flos": 12600839376000.0,
      "grad_norm": 2.491413250017177,
      "language_loss": 0.7679975,
      "learning_rate": 2.9528681822307446e-06,
      "loss": 0.79020935,
      "num_input_tokens_seen": 64635430,
      "step": 3008,
      "time_per_iteration": 3.538148880004883
    },
    {
      "auxiliary_loss_clip": 0.01192066,
      "auxiliary_loss_mlp": 0.01208228,
      "balance_loss_clip": 1.05705547,
      "balance_loss_mlp": 1.00100338,
      "epoch": 0.3618108579330247,
      "flos": 26684644682880.0,
      "grad_norm": 2.4581736281362447,
      "language_loss": 0.82343632,
      "learning_rate": 2.952183230163964e-06,
      "loss": 0.84743929,
      "num_input_tokens_seen": 64655005,
      "step": 3009,
      "time_per_iteration": 2.7101409435272217
    },
    {
      "auxiliary_loss_clip": 0.01194527,
      "auxiliary_loss_mlp": 0.01024708,
      "balance_loss_clip": 1.04909849,
      "balance_loss_mlp": 1.01649499,
      "epoch": 0.3619311008236638,
      "flos": 22817029708800.0,
      "grad_norm": 1.991285945912161,
      "language_loss": 0.73129332,
      "learning_rate": 2.9514981336517448e-06,
      "loss": 0.75348568,
      "num_input_tokens_seen": 64674775,
      "step": 3010,
      "time_per_iteration": 2.665354013442993
    },
    {
      "auxiliary_loss_clip": 0.01193167,
      "auxiliary_loss_mlp": 0.01025936,
      "balance_loss_clip": 1.05561256,
      "balance_loss_mlp": 1.01724577,
      "epoch": 0.36205134371430286,
      "flos": 25919603884800.0,
      "grad_norm": 2.355591836826195,
      "language_loss": 0.81953681,
      "learning_rate": 2.950812892798015e-06,
      "loss": 0.84172779,
      "num_input_tokens_seen": 64695670,
      "step": 3011,
      "time_per_iteration": 2.6822938919067383
    },
    {
      "auxiliary_loss_clip": 0.01199389,
      "auxiliary_loss_mlp": 0.0120844,
      "balance_loss_clip": 1.05243719,
      "balance_loss_mlp": 1.00101662,
      "epoch": 0.362171586604942,
      "flos": 26139592730880.0,
      "grad_norm": 1.798347073982544,
      "language_loss": 0.87500703,
      "learning_rate": 2.9501275077067256e-06,
      "loss": 0.89908528,
      "num_input_tokens_seen": 64716290,
      "step": 3012,
      "time_per_iteration": 3.6240901947021484
    },
    {
      "auxiliary_loss_clip": 0.0118686,
      "auxiliary_loss_mlp": 0.01024006,
      "balance_loss_clip": 1.0442034,
      "balance_loss_mlp": 1.01609695,
      "epoch": 0.3622918294955811,
      "flos": 28074208273920.0,
      "grad_norm": 1.4216300630060523,
      "language_loss": 0.88464063,
      "learning_rate": 2.949441978481848e-06,
      "loss": 0.90674931,
      "num_input_tokens_seen": 64737190,
      "step": 3013,
      "time_per_iteration": 2.7921319007873535
    },
    {
      "auxiliary_loss_clip": 0.01204045,
      "auxiliary_loss_mlp": 0.01027103,
      "balance_loss_clip": 1.05119145,
      "balance_loss_mlp": 1.01805496,
      "epoch": 0.36241207238622014,
      "flos": 19828005402240.0,
      "grad_norm": 2.4502904612614356,
      "language_loss": 0.80175769,
      "learning_rate": 2.9487563052273778e-06,
      "loss": 0.82406914,
      "num_input_tokens_seen": 64753950,
      "step": 3014,
      "time_per_iteration": 2.6574513912200928
    },
    {
      "auxiliary_loss_clip": 0.01192515,
      "auxiliary_loss_mlp": 0.01030729,
      "balance_loss_clip": 1.05790174,
      "balance_loss_mlp": 1.02252185,
      "epoch": 0.36253231527685925,
      "flos": 21397158017280.0,
      "grad_norm": 1.9474458204634955,
      "language_loss": 0.8586902,
      "learning_rate": 2.94807048804733e-06,
      "loss": 0.88092262,
      "num_input_tokens_seen": 64773570,
      "step": 3015,
      "time_per_iteration": 2.622791290283203
    },
    {
      "auxiliary_loss_clip": 0.01201829,
      "auxiliary_loss_mlp": 0.01028688,
      "balance_loss_clip": 1.04808724,
      "balance_loss_mlp": 1.01938367,
      "epoch": 0.36265255816749836,
      "flos": 18362885552640.0,
      "grad_norm": 2.787984702604585,
      "language_loss": 0.9029277,
      "learning_rate": 2.9473845270457434e-06,
      "loss": 0.92523283,
      "num_input_tokens_seen": 64790385,
      "step": 3016,
      "time_per_iteration": 2.651036262512207
    },
    {
      "auxiliary_loss_clip": 0.01188865,
      "auxiliary_loss_mlp": 0.01026322,
      "balance_loss_clip": 1.05033708,
      "balance_loss_mlp": 1.01785254,
      "epoch": 0.3627728010581374,
      "flos": 18660046769280.0,
      "grad_norm": 2.412090707350273,
      "language_loss": 0.698596,
      "learning_rate": 2.946698422326677e-06,
      "loss": 0.72074795,
      "num_input_tokens_seen": 64807845,
      "step": 3017,
      "time_per_iteration": 2.6522064208984375
    },
    {
      "auxiliary_loss_clip": 0.01198794,
      "auxiliary_loss_mlp": 0.01028574,
      "balance_loss_clip": 1.04717493,
      "balance_loss_mlp": 1.01979423,
      "epoch": 0.36289304394877653,
      "flos": 27524272072320.0,
      "grad_norm": 2.0215317406850177,
      "language_loss": 0.79859829,
      "learning_rate": 2.946012173994213e-06,
      "loss": 0.82087195,
      "num_input_tokens_seen": 64827630,
      "step": 3018,
      "time_per_iteration": 2.7834534645080566
    },
    {
      "auxiliary_loss_clip": 0.01191678,
      "auxiliary_loss_mlp": 0.01025605,
      "balance_loss_clip": 1.05757868,
      "balance_loss_mlp": 1.01739192,
      "epoch": 0.36301328683941564,
      "flos": 34533244932480.0,
      "grad_norm": 1.4009507004535366,
      "language_loss": 0.67912251,
      "learning_rate": 2.945325782152454e-06,
      "loss": 0.70129538,
      "num_input_tokens_seen": 64850665,
      "step": 3019,
      "time_per_iteration": 2.7972168922424316
    },
    {
      "auxiliary_loss_clip": 0.0119783,
      "auxiliary_loss_mlp": 0.01026315,
      "balance_loss_clip": 1.05027604,
      "balance_loss_mlp": 1.01815557,
      "epoch": 0.3631335297300547,
      "flos": 19025976574080.0,
      "grad_norm": 2.550437814503233,
      "language_loss": 0.79059124,
      "learning_rate": 2.9446392469055257e-06,
      "loss": 0.81283259,
      "num_input_tokens_seen": 64868700,
      "step": 3020,
      "time_per_iteration": 2.6851203441619873
    },
    {
      "auxiliary_loss_clip": 0.01193709,
      "auxiliary_loss_mlp": 0.01025939,
      "balance_loss_clip": 1.05518985,
      "balance_loss_mlp": 1.01774931,
      "epoch": 0.3632537726206938,
      "flos": 19536769929600.0,
      "grad_norm": 1.8644738122388058,
      "language_loss": 0.80001509,
      "learning_rate": 2.9439525683575745e-06,
      "loss": 0.8222115,
      "num_input_tokens_seen": 64887620,
      "step": 3021,
      "time_per_iteration": 2.6264431476593018
    },
    {
      "auxiliary_loss_clip": 0.01196402,
      "auxiliary_loss_mlp": 0.01029889,
      "balance_loss_clip": 1.05835247,
      "balance_loss_mlp": 1.02059031,
      "epoch": 0.3633740155113329,
      "flos": 21068611292160.0,
      "grad_norm": 2.2742255038602677,
      "language_loss": 0.7494415,
      "learning_rate": 2.9432657466127694e-06,
      "loss": 0.77170444,
      "num_input_tokens_seen": 64907190,
      "step": 3022,
      "time_per_iteration": 2.6402220726013184
    },
    {
      "auxiliary_loss_clip": 0.01201462,
      "auxiliary_loss_mlp": 0.01022907,
      "balance_loss_clip": 1.05286324,
      "balance_loss_mlp": 1.01471484,
      "epoch": 0.36349425840197197,
      "flos": 20298722158080.0,
      "grad_norm": 1.9836943248647576,
      "language_loss": 0.76766527,
      "learning_rate": 2.9425787817753007e-06,
      "loss": 0.78990889,
      "num_input_tokens_seen": 64925850,
      "step": 3023,
      "time_per_iteration": 2.6555728912353516
    },
    {
      "auxiliary_loss_clip": 0.01201205,
      "auxiliary_loss_mlp": 0.01029042,
      "balance_loss_clip": 1.05264604,
      "balance_loss_mlp": 1.0205723,
      "epoch": 0.3636145012926111,
      "flos": 29716762331520.0,
      "grad_norm": 2.6033443678475034,
      "language_loss": 0.71484852,
      "learning_rate": 2.94189167394938e-06,
      "loss": 0.73715103,
      "num_input_tokens_seen": 64948285,
      "step": 3024,
      "time_per_iteration": 2.801466941833496
    },
    {
      "auxiliary_loss_clip": 0.01195404,
      "auxiliary_loss_mlp": 0.01029335,
      "balance_loss_clip": 1.05938816,
      "balance_loss_mlp": 1.02081776,
      "epoch": 0.3637347441832502,
      "flos": 21431847576960.0,
      "grad_norm": 1.9898314364135112,
      "language_loss": 0.8107059,
      "learning_rate": 2.941204423239241e-06,
      "loss": 0.83295333,
      "num_input_tokens_seen": 64967160,
      "step": 3025,
      "time_per_iteration": 2.558370590209961
    },
    {
      "auxiliary_loss_clip": 0.01192116,
      "auxiliary_loss_mlp": 0.01026271,
      "balance_loss_clip": 1.05501771,
      "balance_loss_mlp": 1.01769352,
      "epoch": 0.36385498707388925,
      "flos": 29533941083520.0,
      "grad_norm": 1.816214393823461,
      "language_loss": 0.76155561,
      "learning_rate": 2.9405170297491395e-06,
      "loss": 0.78373951,
      "num_input_tokens_seen": 64987155,
      "step": 3026,
      "time_per_iteration": 2.691011428833008
    },
    {
      "auxiliary_loss_clip": 0.0119422,
      "auxiliary_loss_mlp": 0.01208244,
      "balance_loss_clip": 1.0518527,
      "balance_loss_mlp": 1.00098526,
      "epoch": 0.36397522996452836,
      "flos": 22236569925120.0,
      "grad_norm": 1.817755883576687,
      "language_loss": 0.80036575,
      "learning_rate": 2.939829493583353e-06,
      "loss": 0.82439041,
      "num_input_tokens_seen": 65003800,
      "step": 3027,
      "time_per_iteration": 2.70931077003479
    },
    {
      "auxiliary_loss_clip": 0.01190592,
      "auxiliary_loss_mlp": 0.01025088,
      "balance_loss_clip": 1.0473932,
      "balance_loss_mlp": 1.01670158,
      "epoch": 0.3640954728551674,
      "flos": 21506505995520.0,
      "grad_norm": 2.6054555461495568,
      "language_loss": 0.82790041,
      "learning_rate": 2.939141814846179e-06,
      "loss": 0.85005718,
      "num_input_tokens_seen": 65021215,
      "step": 3028,
      "time_per_iteration": 2.670727014541626
    },
    {
      "auxiliary_loss_clip": 0.01196454,
      "auxiliary_loss_mlp": 0.01025236,
      "balance_loss_clip": 1.05186236,
      "balance_loss_mlp": 1.01625371,
      "epoch": 0.3642157157458065,
      "flos": 17712867081600.0,
      "grad_norm": 1.5706018683350293,
      "language_loss": 0.82440186,
      "learning_rate": 2.938453993641938e-06,
      "loss": 0.84661877,
      "num_input_tokens_seen": 65039590,
      "step": 3029,
      "time_per_iteration": 2.6951870918273926
    },
    {
      "auxiliary_loss_clip": 0.01197728,
      "auxiliary_loss_mlp": 0.01028083,
      "balance_loss_clip": 1.05763745,
      "balance_loss_mlp": 1.01902342,
      "epoch": 0.36433595863644563,
      "flos": 17639537466240.0,
      "grad_norm": 2.2047066998729115,
      "language_loss": 0.70460522,
      "learning_rate": 2.937766030074973e-06,
      "loss": 0.72686338,
      "num_input_tokens_seen": 65056845,
      "step": 3030,
      "time_per_iteration": 3.575820207595825
    },
    {
      "auxiliary_loss_clip": 0.01203041,
      "auxiliary_loss_mlp": 0.01028315,
      "balance_loss_clip": 1.05276036,
      "balance_loss_mlp": 1.01997626,
      "epoch": 0.3644562015270847,
      "flos": 26833279161600.0,
      "grad_norm": 2.2698135706127456,
      "language_loss": 0.83109963,
      "learning_rate": 2.937077924249646e-06,
      "loss": 0.85341322,
      "num_input_tokens_seen": 65079435,
      "step": 3031,
      "time_per_iteration": 3.769631862640381
    },
    {
      "auxiliary_loss_clip": 0.01202038,
      "auxiliary_loss_mlp": 0.01026743,
      "balance_loss_clip": 1.05304945,
      "balance_loss_mlp": 1.01816642,
      "epoch": 0.3645764444177238,
      "flos": 14282715847680.0,
      "grad_norm": 2.4988935714003393,
      "language_loss": 0.75850779,
      "learning_rate": 2.9363896762703443e-06,
      "loss": 0.78079563,
      "num_input_tokens_seen": 65096500,
      "step": 3032,
      "time_per_iteration": 2.6045994758605957
    },
    {
      "auxiliary_loss_clip": 0.01195364,
      "auxiliary_loss_mlp": 0.01030252,
      "balance_loss_clip": 1.05937338,
      "balance_loss_mlp": 1.02087641,
      "epoch": 0.3646966873083629,
      "flos": 20667489137280.0,
      "grad_norm": 1.834217580284325,
      "language_loss": 0.84310842,
      "learning_rate": 2.9357012862414725e-06,
      "loss": 0.86536455,
      "num_input_tokens_seen": 65115860,
      "step": 3033,
      "time_per_iteration": 2.571577548980713
    },
    {
      "auxiliary_loss_clip": 0.01194496,
      "auxiliary_loss_mlp": 0.01026351,
      "balance_loss_clip": 1.05642438,
      "balance_loss_mlp": 1.0179112,
      "epoch": 0.36481693019900197,
      "flos": 27782613665280.0,
      "grad_norm": 1.933533077682195,
      "language_loss": 0.71389723,
      "learning_rate": 2.9350127542674593e-06,
      "loss": 0.73610574,
      "num_input_tokens_seen": 65138070,
      "step": 3034,
      "time_per_iteration": 3.595534563064575
    },
    {
      "auxiliary_loss_clip": 0.0120228,
      "auxiliary_loss_mlp": 0.01029937,
      "balance_loss_clip": 1.05522478,
      "balance_loss_mlp": 1.02151513,
      "epoch": 0.3649371730896411,
      "flos": 19712588025600.0,
      "grad_norm": 2.1622259829118105,
      "language_loss": 0.76288271,
      "learning_rate": 2.934324080452755e-06,
      "loss": 0.78520489,
      "num_input_tokens_seen": 65155860,
      "step": 3035,
      "time_per_iteration": 2.6340672969818115
    },
    {
      "auxiliary_loss_clip": 0.01188854,
      "auxiliary_loss_mlp": 0.01208868,
      "balance_loss_clip": 1.04812527,
      "balance_loss_mlp": 1.00098681,
      "epoch": 0.3650574159802802,
      "flos": 24750496016640.0,
      "grad_norm": 1.7787352440702215,
      "language_loss": 0.78094071,
      "learning_rate": 2.9336352649018307e-06,
      "loss": 0.80491793,
      "num_input_tokens_seen": 65175930,
      "step": 3036,
      "time_per_iteration": 2.7560887336730957
    },
    {
      "auxiliary_loss_clip": 0.01198773,
      "auxiliary_loss_mlp": 0.0102756,
      "balance_loss_clip": 1.05421519,
      "balance_loss_mlp": 1.01925099,
      "epoch": 0.36517765887091924,
      "flos": 32853487363200.0,
      "grad_norm": 1.8618630067782291,
      "language_loss": 0.70216399,
      "learning_rate": 2.9329463077191783e-06,
      "loss": 0.72442734,
      "num_input_tokens_seen": 65199305,
      "step": 3037,
      "time_per_iteration": 2.7341721057891846
    },
    {
      "auxiliary_loss_clip": 0.0120099,
      "auxiliary_loss_mlp": 0.01022908,
      "balance_loss_clip": 1.05092645,
      "balance_loss_mlp": 1.01456952,
      "epoch": 0.36529790176155835,
      "flos": 20120318282880.0,
      "grad_norm": 2.814806866650201,
      "language_loss": 0.63634044,
      "learning_rate": 2.9322572090093135e-06,
      "loss": 0.65857935,
      "num_input_tokens_seen": 65218010,
      "step": 3038,
      "time_per_iteration": 3.573789119720459
    },
    {
      "auxiliary_loss_clip": 0.01200809,
      "auxiliary_loss_mlp": 0.01029325,
      "balance_loss_clip": 1.04770648,
      "balance_loss_mlp": 1.0206939,
      "epoch": 0.36541814465219746,
      "flos": 17639573379840.0,
      "grad_norm": 4.237262830194408,
      "language_loss": 0.77181232,
      "learning_rate": 2.9315679688767713e-06,
      "loss": 0.79411364,
      "num_input_tokens_seen": 65236020,
      "step": 3039,
      "time_per_iteration": 2.6827731132507324
    },
    {
      "auxiliary_loss_clip": 0.01191835,
      "auxiliary_loss_mlp": 0.01026614,
      "balance_loss_clip": 1.05214357,
      "balance_loss_mlp": 1.01837659,
      "epoch": 0.3655383875428365,
      "flos": 22674356887680.0,
      "grad_norm": 1.7736183810888926,
      "language_loss": 0.66775626,
      "learning_rate": 2.9308785874261085e-06,
      "loss": 0.68994075,
      "num_input_tokens_seen": 65256210,
      "step": 3040,
      "time_per_iteration": 2.6638567447662354
    },
    {
      "auxiliary_loss_clip": 0.01193603,
      "auxiliary_loss_mlp": 0.01028949,
      "balance_loss_clip": 1.05831385,
      "balance_loss_mlp": 1.02084839,
      "epoch": 0.36565863043347563,
      "flos": 21981173247360.0,
      "grad_norm": 4.53531949090335,
      "language_loss": 0.81925344,
      "learning_rate": 2.9301890647619045e-06,
      "loss": 0.84147894,
      "num_input_tokens_seen": 65275505,
      "step": 3041,
      "time_per_iteration": 2.6601336002349854
    },
    {
      "auxiliary_loss_clip": 0.01202786,
      "auxiliary_loss_mlp": 0.01031097,
      "balance_loss_clip": 1.05484068,
      "balance_loss_mlp": 1.02179861,
      "epoch": 0.36577887332411474,
      "flos": 24827632473600.0,
      "grad_norm": 2.35531806631609,
      "language_loss": 0.8009975,
      "learning_rate": 2.929499400988759e-06,
      "loss": 0.8233363,
      "num_input_tokens_seen": 65296665,
      "step": 3042,
      "time_per_iteration": 2.642246723175049
    },
    {
      "auxiliary_loss_clip": 0.01193091,
      "auxiliary_loss_mlp": 0.01029201,
      "balance_loss_clip": 1.05633724,
      "balance_loss_mlp": 1.02020109,
      "epoch": 0.3658991162147538,
      "flos": 28293191539200.0,
      "grad_norm": 1.9481065686814343,
      "language_loss": 0.65833306,
      "learning_rate": 2.9288095962112927e-06,
      "loss": 0.680556,
      "num_input_tokens_seen": 65317370,
      "step": 3043,
      "time_per_iteration": 2.728959560394287
    },
    {
      "auxiliary_loss_clip": 0.01191699,
      "auxiliary_loss_mlp": 0.01026415,
      "balance_loss_clip": 1.05689442,
      "balance_loss_mlp": 1.01783156,
      "epoch": 0.3660193591053929,
      "flos": 17785550252160.0,
      "grad_norm": 2.169511510889268,
      "language_loss": 0.85371923,
      "learning_rate": 2.9281196505341503e-06,
      "loss": 0.87590039,
      "num_input_tokens_seen": 65334540,
      "step": 3044,
      "time_per_iteration": 2.6044068336486816
    },
    {
      "auxiliary_loss_clip": 0.01195558,
      "auxiliary_loss_mlp": 0.01208059,
      "balance_loss_clip": 1.05206573,
      "balance_loss_mlp": 1.00083268,
      "epoch": 0.36613960199603196,
      "flos": 10342776839040.0,
      "grad_norm": 2.4984391393263428,
      "language_loss": 0.78738606,
      "learning_rate": 2.9274295640619946e-06,
      "loss": 0.81142223,
      "num_input_tokens_seen": 65351670,
      "step": 3045,
      "time_per_iteration": 2.713568687438965
    },
    {
      "auxiliary_loss_clip": 0.01197177,
      "auxiliary_loss_mlp": 0.01023414,
      "balance_loss_clip": 1.04941356,
      "balance_loss_mlp": 1.01577926,
      "epoch": 0.36625984488667107,
      "flos": 19755609540480.0,
      "grad_norm": 1.7469187867710014,
      "language_loss": 0.78895217,
      "learning_rate": 2.9267393368995103e-06,
      "loss": 0.81115806,
      "num_input_tokens_seen": 65370900,
      "step": 3046,
      "time_per_iteration": 2.625152587890625
    },
    {
      "auxiliary_loss_clip": 0.01193667,
      "auxiliary_loss_mlp": 0.0103243,
      "balance_loss_clip": 1.05811584,
      "balance_loss_mlp": 1.02434206,
      "epoch": 0.3663800877773102,
      "flos": 17674262939520.0,
      "grad_norm": 2.2571482853706297,
      "language_loss": 0.74096912,
      "learning_rate": 2.926048969151407e-06,
      "loss": 0.76323009,
      "num_input_tokens_seen": 65388185,
      "step": 3047,
      "time_per_iteration": 2.6187984943389893
    },
    {
      "auxiliary_loss_clip": 0.01199783,
      "auxiliary_loss_mlp": 0.01023855,
      "balance_loss_clip": 1.05444562,
      "balance_loss_mlp": 1.0150516,
      "epoch": 0.36650033066794924,
      "flos": 20303606407680.0,
      "grad_norm": 4.127675267706631,
      "language_loss": 0.68749243,
      "learning_rate": 2.92535846092241e-06,
      "loss": 0.70972878,
      "num_input_tokens_seen": 65407200,
      "step": 3048,
      "time_per_iteration": 2.6847808361053467
    },
    {
      "auxiliary_loss_clip": 0.01201296,
      "auxiliary_loss_mlp": 0.01028451,
      "balance_loss_clip": 1.05783761,
      "balance_loss_mlp": 1.02025557,
      "epoch": 0.36662057355858835,
      "flos": 24716237420160.0,
      "grad_norm": 1.7359009940322043,
      "language_loss": 0.82899237,
      "learning_rate": 2.9246678123172704e-06,
      "loss": 0.85128987,
      "num_input_tokens_seen": 65427290,
      "step": 3049,
      "time_per_iteration": 2.698131799697876
    },
    {
      "auxiliary_loss_clip": 0.01194838,
      "auxiliary_loss_mlp": 0.01031994,
      "balance_loss_clip": 1.05857408,
      "balance_loss_mlp": 1.02308905,
      "epoch": 0.36674081644922746,
      "flos": 12385267902720.0,
      "grad_norm": 4.063976607923815,
      "language_loss": 0.74384558,
      "learning_rate": 2.9239770234407596e-06,
      "loss": 0.76611388,
      "num_input_tokens_seen": 65445595,
      "step": 3050,
      "time_per_iteration": 2.6150922775268555
    },
    {
      "auxiliary_loss_clip": 0.01196737,
      "auxiliary_loss_mlp": 0.01028307,
      "balance_loss_clip": 1.05634975,
      "balance_loss_mlp": 1.01976013,
      "epoch": 0.3668610593398665,
      "flos": 21105922544640.0,
      "grad_norm": 1.7121960267031302,
      "language_loss": 0.6855818,
      "learning_rate": 2.9232860943976686e-06,
      "loss": 0.70783228,
      "num_input_tokens_seen": 65466330,
      "step": 3051,
      "time_per_iteration": 2.669603109359741
    },
    {
      "auxiliary_loss_clip": 0.01195168,
      "auxiliary_loss_mlp": 0.01023225,
      "balance_loss_clip": 1.05587614,
      "balance_loss_mlp": 1.01501763,
      "epoch": 0.3669813022305056,
      "flos": 26758082039040.0,
      "grad_norm": 1.655524997634487,
      "language_loss": 0.84641033,
      "learning_rate": 2.9225950252928115e-06,
      "loss": 0.86859429,
      "num_input_tokens_seen": 65487180,
      "step": 3052,
      "time_per_iteration": 2.6958224773406982
    },
    {
      "auxiliary_loss_clip": 0.0119405,
      "auxiliary_loss_mlp": 0.01029851,
      "balance_loss_clip": 1.05779624,
      "balance_loss_mlp": 1.02106571,
      "epoch": 0.36710154512114473,
      "flos": 19099521671040.0,
      "grad_norm": 2.1465282746941456,
      "language_loss": 0.81948167,
      "learning_rate": 2.9219038162310217e-06,
      "loss": 0.8417207,
      "num_input_tokens_seen": 65505380,
      "step": 3053,
      "time_per_iteration": 2.653135061264038
    },
    {
      "auxiliary_loss_clip": 0.01207204,
      "auxiliary_loss_mlp": 0.0120833,
      "balance_loss_clip": 1.04852521,
      "balance_loss_mlp": 1.00080252,
      "epoch": 0.3672217880117838,
      "flos": 20812029465600.0,
      "grad_norm": 1.7685393100206588,
      "language_loss": 0.82671231,
      "learning_rate": 2.921212467317157e-06,
      "loss": 0.85086775,
      "num_input_tokens_seen": 65524825,
      "step": 3054,
      "time_per_iteration": 2.770124673843384
    },
    {
      "auxiliary_loss_clip": 0.01183739,
      "auxiliary_loss_mlp": 0.01032214,
      "balance_loss_clip": 1.05081975,
      "balance_loss_mlp": 1.02327323,
      "epoch": 0.3673420309024229,
      "flos": 13590394133760.0,
      "grad_norm": 2.480283273498108,
      "language_loss": 0.79998076,
      "learning_rate": 2.920520978656093e-06,
      "loss": 0.82214034,
      "num_input_tokens_seen": 65541790,
      "step": 3055,
      "time_per_iteration": 2.599457025527954
    },
    {
      "auxiliary_loss_clip": 0.01188208,
      "auxiliary_loss_mlp": 0.01208276,
      "balance_loss_clip": 1.05621636,
      "balance_loss_mlp": 1.00081134,
      "epoch": 0.367462273793062,
      "flos": 28986877969920.0,
      "grad_norm": 2.6161732844941956,
      "language_loss": 0.77097285,
      "learning_rate": 2.919829350352729e-06,
      "loss": 0.79493767,
      "num_input_tokens_seen": 65563395,
      "step": 3056,
      "time_per_iteration": 2.7537357807159424
    },
    {
      "auxiliary_loss_clip": 0.01119227,
      "auxiliary_loss_mlp": 0.01002011,
      "balance_loss_clip": 1.04862309,
      "balance_loss_mlp": 1.00111723,
      "epoch": 0.36758251668370107,
      "flos": 62643148346880.0,
      "grad_norm": 0.7597799626635197,
      "language_loss": 0.59944236,
      "learning_rate": 2.919137582511983e-06,
      "loss": 0.62065476,
      "num_input_tokens_seen": 65619835,
      "step": 3057,
      "time_per_iteration": 4.011793613433838
    },
    {
      "auxiliary_loss_clip": 0.01206241,
      "auxiliary_loss_mlp": 0.01029222,
      "balance_loss_clip": 1.05826259,
      "balance_loss_mlp": 1.0211699,
      "epoch": 0.3677027595743402,
      "flos": 12713886455040.0,
      "grad_norm": 2.110369183318507,
      "language_loss": 0.63493872,
      "learning_rate": 2.918445675238797e-06,
      "loss": 0.65729332,
      "num_input_tokens_seen": 65636760,
      "step": 3058,
      "time_per_iteration": 2.6864564418792725
    },
    {
      "auxiliary_loss_clip": 0.01193443,
      "auxiliary_loss_mlp": 0.01025303,
      "balance_loss_clip": 1.05767381,
      "balance_loss_mlp": 1.01689243,
      "epoch": 0.36782300246497923,
      "flos": 25046579825280.0,
      "grad_norm": 1.9301544070583863,
      "language_loss": 0.69557655,
      "learning_rate": 2.917753628638132e-06,
      "loss": 0.71776402,
      "num_input_tokens_seen": 65657065,
      "step": 3059,
      "time_per_iteration": 2.7225589752197266
    },
    {
      "auxiliary_loss_clip": 0.01197745,
      "auxiliary_loss_mlp": 0.01023882,
      "balance_loss_clip": 1.0565052,
      "balance_loss_mlp": 1.01522112,
      "epoch": 0.36794324535561834,
      "flos": 17419512706560.0,
      "grad_norm": 3.995855350583021,
      "language_loss": 0.70233971,
      "learning_rate": 2.9170614428149716e-06,
      "loss": 0.72455597,
      "num_input_tokens_seen": 65675400,
      "step": 3060,
      "time_per_iteration": 2.6097679138183594
    },
    {
      "auxiliary_loss_clip": 0.01196167,
      "auxiliary_loss_mlp": 0.01033544,
      "balance_loss_clip": 1.0533613,
      "balance_loss_mlp": 1.02426338,
      "epoch": 0.36806348824625745,
      "flos": 24089128848000.0,
      "grad_norm": 2.674315806767818,
      "language_loss": 0.87198353,
      "learning_rate": 2.9163691178743195e-06,
      "loss": 0.89428067,
      "num_input_tokens_seen": 65694050,
      "step": 3061,
      "time_per_iteration": 3.566336154937744
    },
    {
      "auxiliary_loss_clip": 0.01190268,
      "auxiliary_loss_mlp": 0.01028271,
      "balance_loss_clip": 1.0550983,
      "balance_loss_mlp": 1.01971757,
      "epoch": 0.3681837311368965,
      "flos": 20521871400960.0,
      "grad_norm": 1.8167740348751935,
      "language_loss": 0.77370667,
      "learning_rate": 2.9156766539212006e-06,
      "loss": 0.79589212,
      "num_input_tokens_seen": 65711695,
      "step": 3062,
      "time_per_iteration": 2.544088840484619
    },
    {
      "auxiliary_loss_clip": 0.01196542,
      "auxiliary_loss_mlp": 0.01034776,
      "balance_loss_clip": 1.05356884,
      "balance_loss_mlp": 1.02606142,
      "epoch": 0.3683039740275356,
      "flos": 21466644877440.0,
      "grad_norm": 2.345558957660508,
      "language_loss": 0.72003388,
      "learning_rate": 2.9149840510606614e-06,
      "loss": 0.742347,
      "num_input_tokens_seen": 65730350,
      "step": 3063,
      "time_per_iteration": 2.6509087085723877
    },
    {
      "auxiliary_loss_clip": 0.01115976,
      "auxiliary_loss_mlp": 0.01199687,
      "balance_loss_clip": 1.04172564,
      "balance_loss_mlp": 1.00028944,
      "epoch": 0.36842421691817473,
      "flos": 70380999987840.0,
      "grad_norm": 3.4604757594896194,
      "language_loss": 0.64135951,
      "learning_rate": 2.914291309397769e-06,
      "loss": 0.66451615,
      "num_input_tokens_seen": 65787820,
      "step": 3064,
      "time_per_iteration": 4.181278705596924
    },
    {
      "auxiliary_loss_clip": 0.01190708,
      "auxiliary_loss_mlp": 0.01025026,
      "balance_loss_clip": 1.04624355,
      "balance_loss_mlp": 1.0158112,
      "epoch": 0.3685444598088138,
      "flos": 23331378510720.0,
      "grad_norm": 2.0542515236267778,
      "language_loss": 0.78633767,
      "learning_rate": 2.9135984290376117e-06,
      "loss": 0.80849504,
      "num_input_tokens_seen": 65806685,
      "step": 3065,
      "time_per_iteration": 2.697672128677368
    },
    {
      "auxiliary_loss_clip": 0.01200015,
      "auxiliary_loss_mlp": 0.01029882,
      "balance_loss_clip": 1.04737043,
      "balance_loss_mlp": 1.02171564,
      "epoch": 0.3686647026994529,
      "flos": 23070271570560.0,
      "grad_norm": 1.95345782545935,
      "language_loss": 0.82777154,
      "learning_rate": 2.9129054100853e-06,
      "loss": 0.85007042,
      "num_input_tokens_seen": 65825525,
      "step": 3066,
      "time_per_iteration": 2.7480194568634033
    },
    {
      "auxiliary_loss_clip": 0.01198825,
      "auxiliary_loss_mlp": 0.01027275,
      "balance_loss_clip": 1.05445874,
      "balance_loss_mlp": 1.01822066,
      "epoch": 0.368784945590092,
      "flos": 25119909440640.0,
      "grad_norm": 1.9350385595723285,
      "language_loss": 0.75964057,
      "learning_rate": 2.912212252645963e-06,
      "loss": 0.78190154,
      "num_input_tokens_seen": 65848110,
      "step": 3067,
      "time_per_iteration": 2.6408610343933105
    },
    {
      "auxiliary_loss_clip": 0.01199681,
      "auxiliary_loss_mlp": 0.01028645,
      "balance_loss_clip": 1.0552882,
      "balance_loss_mlp": 1.01966894,
      "epoch": 0.36890518848073106,
      "flos": 18442284566400.0,
      "grad_norm": 26.20649172100686,
      "language_loss": 0.76934063,
      "learning_rate": 2.9115189568247523e-06,
      "loss": 0.79162389,
      "num_input_tokens_seen": 65865670,
      "step": 3068,
      "time_per_iteration": 2.6783270835876465
    },
    {
      "auxiliary_loss_clip": 0.01190066,
      "auxiliary_loss_mlp": 0.01027857,
      "balance_loss_clip": 1.05402982,
      "balance_loss_mlp": 1.01907122,
      "epoch": 0.36902543137137017,
      "flos": 16362446336640.0,
      "grad_norm": 3.648677682739401,
      "language_loss": 0.92505181,
      "learning_rate": 2.910825522726841e-06,
      "loss": 0.94723105,
      "num_input_tokens_seen": 65883195,
      "step": 3069,
      "time_per_iteration": 2.6547117233276367
    },
    {
      "auxiliary_loss_clip": 0.01193775,
      "auxiliary_loss_mlp": 0.0103065,
      "balance_loss_clip": 1.04694796,
      "balance_loss_mlp": 1.02196574,
      "epoch": 0.3691456742620093,
      "flos": 12275596702080.0,
      "grad_norm": 2.6354974661367114,
      "language_loss": 0.7734412,
      "learning_rate": 2.9101319504574215e-06,
      "loss": 0.79568541,
      "num_input_tokens_seen": 65899635,
      "step": 3070,
      "time_per_iteration": 2.6902546882629395
    },
    {
      "auxiliary_loss_clip": 0.01185079,
      "auxiliary_loss_mlp": 0.01027922,
      "balance_loss_clip": 1.05238748,
      "balance_loss_mlp": 1.01886761,
      "epoch": 0.36926591715264834,
      "flos": 17786412178560.0,
      "grad_norm": 1.9813934763927978,
      "language_loss": 0.76450706,
      "learning_rate": 2.909438240121709e-06,
      "loss": 0.78663707,
      "num_input_tokens_seen": 65919910,
      "step": 3071,
      "time_per_iteration": 2.6605188846588135
    },
    {
      "auxiliary_loss_clip": 0.01190397,
      "auxiliary_loss_mlp": 0.01023377,
      "balance_loss_clip": 1.05406475,
      "balance_loss_mlp": 1.01480591,
      "epoch": 0.36938616004328745,
      "flos": 28948309741440.0,
      "grad_norm": 1.8335793869208845,
      "language_loss": 0.70433331,
      "learning_rate": 2.908744391824939e-06,
      "loss": 0.72647101,
      "num_input_tokens_seen": 65940930,
      "step": 3072,
      "time_per_iteration": 2.7254154682159424
    },
    {
      "auxiliary_loss_clip": 0.01202775,
      "auxiliary_loss_mlp": 0.01024975,
      "balance_loss_clip": 1.04717803,
      "balance_loss_mlp": 1.0163976,
      "epoch": 0.36950640293392656,
      "flos": 29205394358400.0,
      "grad_norm": 1.9795186454733378,
      "language_loss": 0.79343671,
      "learning_rate": 2.908050405672367e-06,
      "loss": 0.81571424,
      "num_input_tokens_seen": 65960475,
      "step": 3073,
      "time_per_iteration": 2.7946739196777344
    },
    {
      "auxiliary_loss_clip": 0.01201791,
      "auxiliary_loss_mlp": 0.01030812,
      "balance_loss_clip": 1.05042171,
      "balance_loss_mlp": 1.02217531,
      "epoch": 0.3696266458245656,
      "flos": 24827776128000.0,
      "grad_norm": 1.7357231299122229,
      "language_loss": 0.7920469,
      "learning_rate": 2.9073562817692703e-06,
      "loss": 0.81437296,
      "num_input_tokens_seen": 65979160,
      "step": 3074,
      "time_per_iteration": 2.7103898525238037
    },
    {
      "auxiliary_loss_clip": 0.01138636,
      "auxiliary_loss_mlp": 0.01004358,
      "balance_loss_clip": 1.03966308,
      "balance_loss_mlp": 1.00336313,
      "epoch": 0.3697468887152047,
      "flos": 59887257264000.0,
      "grad_norm": 0.7232517053315664,
      "language_loss": 0.56517428,
      "learning_rate": 2.9066620202209468e-06,
      "loss": 0.58660418,
      "num_input_tokens_seen": 66041650,
      "step": 3075,
      "time_per_iteration": 3.2144434452056885
    },
    {
      "auxiliary_loss_clip": 0.0119136,
      "auxiliary_loss_mlp": 0.01029967,
      "balance_loss_clip": 1.05239415,
      "balance_loss_mlp": 1.02137172,
      "epoch": 0.3698671316058438,
      "flos": 26137581569280.0,
      "grad_norm": 3.4904836307370357,
      "language_loss": 0.77548933,
      "learning_rate": 2.905967621132716e-06,
      "loss": 0.79770255,
      "num_input_tokens_seen": 66059260,
      "step": 3076,
      "time_per_iteration": 2.90080189704895
    },
    {
      "auxiliary_loss_clip": 0.01201568,
      "auxiliary_loss_mlp": 0.01029291,
      "balance_loss_clip": 1.05379558,
      "balance_loss_mlp": 1.02070165,
      "epoch": 0.3699873744964829,
      "flos": 24607464059520.0,
      "grad_norm": 2.119765964154256,
      "language_loss": 0.75278765,
      "learning_rate": 2.9052730846099172e-06,
      "loss": 0.7750963,
      "num_input_tokens_seen": 66080605,
      "step": 3077,
      "time_per_iteration": 2.7137348651885986
    },
    {
      "auxiliary_loss_clip": 0.01116695,
      "auxiliary_loss_mlp": 0.01000922,
      "balance_loss_clip": 1.03586781,
      "balance_loss_mlp": 0.99995071,
      "epoch": 0.370107617387122,
      "flos": 64885340050560.0,
      "grad_norm": 0.871768203347208,
      "language_loss": 0.60820633,
      "learning_rate": 2.9045784107579123e-06,
      "loss": 0.62938255,
      "num_input_tokens_seen": 66140710,
      "step": 3078,
      "time_per_iteration": 3.259683847427368
    },
    {
      "auxiliary_loss_clip": 0.01192599,
      "auxiliary_loss_mlp": 0.0102662,
      "balance_loss_clip": 1.05841887,
      "balance_loss_mlp": 1.01777458,
      "epoch": 0.37022786027776106,
      "flos": 15961683317760.0,
      "grad_norm": 2.2160260833988588,
      "language_loss": 0.66890001,
      "learning_rate": 2.9038835996820807e-06,
      "loss": 0.69109225,
      "num_input_tokens_seen": 66158320,
      "step": 3079,
      "time_per_iteration": 2.5920634269714355
    },
    {
      "auxiliary_loss_clip": 0.01201441,
      "auxiliary_loss_mlp": 0.01028113,
      "balance_loss_clip": 1.04816437,
      "balance_loss_mlp": 1.01920187,
      "epoch": 0.37034810316840017,
      "flos": 18546927863040.0,
      "grad_norm": 2.5575281876544196,
      "language_loss": 0.79628217,
      "learning_rate": 2.903188651487826e-06,
      "loss": 0.81857765,
      "num_input_tokens_seen": 66176875,
      "step": 3080,
      "time_per_iteration": 2.7014925479888916
    },
    {
      "auxiliary_loss_clip": 0.01199475,
      "auxiliary_loss_mlp": 0.01027362,
      "balance_loss_clip": 1.05847454,
      "balance_loss_mlp": 1.01848662,
      "epoch": 0.3704683460590393,
      "flos": 17821927751040.0,
      "grad_norm": 2.1092308319476873,
      "language_loss": 0.86613542,
      "learning_rate": 2.902493566280571e-06,
      "loss": 0.88840377,
      "num_input_tokens_seen": 66194980,
      "step": 3081,
      "time_per_iteration": 2.6694939136505127
    },
    {
      "auxiliary_loss_clip": 0.01194519,
      "auxiliary_loss_mlp": 0.01026209,
      "balance_loss_clip": 1.05337572,
      "balance_loss_mlp": 1.01722014,
      "epoch": 0.37058858894967833,
      "flos": 14134081368960.0,
      "grad_norm": 15.53922090883016,
      "language_loss": 0.81217009,
      "learning_rate": 2.9017983441657595e-06,
      "loss": 0.83437741,
      "num_input_tokens_seen": 66212310,
      "step": 3082,
      "time_per_iteration": 2.6443493366241455
    },
    {
      "auxiliary_loss_clip": 0.01205387,
      "auxiliary_loss_mlp": 0.01026011,
      "balance_loss_clip": 1.04999125,
      "balance_loss_mlp": 1.01723075,
      "epoch": 0.37070883184031744,
      "flos": 13954492344960.0,
      "grad_norm": 2.3651642954941994,
      "language_loss": 0.7504099,
      "learning_rate": 2.9011029852488564e-06,
      "loss": 0.77272391,
      "num_input_tokens_seen": 66229545,
      "step": 3083,
      "time_per_iteration": 2.7152161598205566
    },
    {
      "auxiliary_loss_clip": 0.01105387,
      "auxiliary_loss_mlp": 0.01000995,
      "balance_loss_clip": 1.03650153,
      "balance_loss_mlp": 1.00000513,
      "epoch": 0.37082907473095655,
      "flos": 52315419306240.0,
      "grad_norm": 0.9760391544176348,
      "language_loss": 0.62513578,
      "learning_rate": 2.9004074896353465e-06,
      "loss": 0.64619964,
      "num_input_tokens_seen": 66283545,
      "step": 3084,
      "time_per_iteration": 3.9906342029571533
    },
    {
      "auxiliary_loss_clip": 0.01193462,
      "auxiliary_loss_mlp": 0.01026003,
      "balance_loss_clip": 1.06138897,
      "balance_loss_mlp": 1.01819789,
      "epoch": 0.3709493176215956,
      "flos": 15998096730240.0,
      "grad_norm": 4.634900828575179,
      "language_loss": 0.81479526,
      "learning_rate": 2.8997118574307362e-06,
      "loss": 0.83698988,
      "num_input_tokens_seen": 66300500,
      "step": 3085,
      "time_per_iteration": 2.5810303688049316
    },
    {
      "auxiliary_loss_clip": 0.01205659,
      "auxiliary_loss_mlp": 0.01029661,
      "balance_loss_clip": 1.05437732,
      "balance_loss_mlp": 1.02087522,
      "epoch": 0.3710695605122347,
      "flos": 20959837931520.0,
      "grad_norm": 2.129386950016678,
      "language_loss": 0.74245441,
      "learning_rate": 2.899016088740553e-06,
      "loss": 0.76480758,
      "num_input_tokens_seen": 66318610,
      "step": 3086,
      "time_per_iteration": 2.7189831733703613
    },
    {
      "auxiliary_loss_clip": 0.01196774,
      "auxiliary_loss_mlp": 0.01024926,
      "balance_loss_clip": 1.04885614,
      "balance_loss_mlp": 1.01725459,
      "epoch": 0.37118980340287383,
      "flos": 14355578586240.0,
      "grad_norm": 2.6504062437793654,
      "language_loss": 0.79290301,
      "learning_rate": 2.898320183670344e-06,
      "loss": 0.81511998,
      "num_input_tokens_seen": 66336025,
      "step": 3087,
      "time_per_iteration": 2.648331642150879
    },
    {
      "auxiliary_loss_clip": 0.01201721,
      "auxiliary_loss_mlp": 0.01024172,
      "balance_loss_clip": 1.05536199,
      "balance_loss_mlp": 1.01548147,
      "epoch": 0.3713100462935129,
      "flos": 25885381201920.0,
      "grad_norm": 1.7286507940853961,
      "language_loss": 0.89457923,
      "learning_rate": 2.8976241423256767e-06,
      "loss": 0.91683817,
      "num_input_tokens_seen": 66356120,
      "step": 3088,
      "time_per_iteration": 3.639479637145996
    },
    {
      "auxiliary_loss_clip": 0.01191457,
      "auxiliary_loss_mlp": 0.01030035,
      "balance_loss_clip": 1.05249429,
      "balance_loss_mlp": 1.02196419,
      "epoch": 0.371430289184152,
      "flos": 30518934814080.0,
      "grad_norm": 2.8640060652699635,
      "language_loss": 0.6820817,
      "learning_rate": 2.896927964812142e-06,
      "loss": 0.70429665,
      "num_input_tokens_seen": 66376685,
      "step": 3089,
      "time_per_iteration": 2.7560524940490723
    },
    {
      "auxiliary_loss_clip": 0.0119513,
      "auxiliary_loss_mlp": 0.01027445,
      "balance_loss_clip": 1.05701876,
      "balance_loss_mlp": 1.01868868,
      "epoch": 0.37155053207479105,
      "flos": 15742233175680.0,
      "grad_norm": 2.3504783449786992,
      "language_loss": 0.75235403,
      "learning_rate": 2.8962316512353465e-06,
      "loss": 0.77457976,
      "num_input_tokens_seen": 66394230,
      "step": 3090,
      "time_per_iteration": 2.624627113342285
    },
    {
      "auxiliary_loss_clip": 0.01195045,
      "auxiliary_loss_mlp": 0.01029193,
      "balance_loss_clip": 1.04392338,
      "balance_loss_mlp": 1.02080035,
      "epoch": 0.37167077496543016,
      "flos": 23404061681280.0,
      "grad_norm": 1.7057876688465785,
      "language_loss": 0.75225788,
      "learning_rate": 2.8955352017009233e-06,
      "loss": 0.77450025,
      "num_input_tokens_seen": 66413475,
      "step": 3091,
      "time_per_iteration": 3.7400834560394287
    },
    {
      "auxiliary_loss_clip": 0.01195222,
      "auxiliary_loss_mlp": 0.01029336,
      "balance_loss_clip": 1.05657732,
      "balance_loss_mlp": 1.0204196,
      "epoch": 0.3717910178560693,
      "flos": 22088653718400.0,
      "grad_norm": 2.289198419939049,
      "language_loss": 0.77214158,
      "learning_rate": 2.8948386163145212e-06,
      "loss": 0.79438722,
      "num_input_tokens_seen": 66432685,
      "step": 3092,
      "time_per_iteration": 2.70924711227417
    },
    {
      "auxiliary_loss_clip": 0.0119904,
      "auxiliary_loss_mlp": 0.01027811,
      "balance_loss_clip": 1.05784464,
      "balance_loss_mlp": 1.01944876,
      "epoch": 0.3719112607467083,
      "flos": 26939969533440.0,
      "grad_norm": 3.08822486600604,
      "language_loss": 0.80046284,
      "learning_rate": 2.8941418951818135e-06,
      "loss": 0.82273132,
      "num_input_tokens_seen": 66452245,
      "step": 3093,
      "time_per_iteration": 2.7176127433776855
    },
    {
      "auxiliary_loss_clip": 0.01200626,
      "auxiliary_loss_mlp": 0.01026565,
      "balance_loss_clip": 1.05305791,
      "balance_loss_mlp": 1.01836979,
      "epoch": 0.37203150363734744,
      "flos": 12166500119040.0,
      "grad_norm": 2.1644627379020673,
      "language_loss": 0.718027,
      "learning_rate": 2.8934450384084903e-06,
      "loss": 0.74029893,
      "num_input_tokens_seen": 66469760,
      "step": 3094,
      "time_per_iteration": 2.7170069217681885
    },
    {
      "auxiliary_loss_clip": 0.01190813,
      "auxiliary_loss_mlp": 0.01026033,
      "balance_loss_clip": 1.05418611,
      "balance_loss_mlp": 1.01728308,
      "epoch": 0.37215174652798655,
      "flos": 23697595624320.0,
      "grad_norm": 2.1615949746954377,
      "language_loss": 0.69927663,
      "learning_rate": 2.8927480461002653e-06,
      "loss": 0.72144508,
      "num_input_tokens_seen": 66489730,
      "step": 3095,
      "time_per_iteration": 2.654261350631714
    },
    {
      "auxiliary_loss_clip": 0.01196932,
      "auxiliary_loss_mlp": 0.0103226,
      "balance_loss_clip": 1.05358648,
      "balance_loss_mlp": 1.02303886,
      "epoch": 0.3722719894186256,
      "flos": 17887751424000.0,
      "grad_norm": 6.436759461763789,
      "language_loss": 0.86092925,
      "learning_rate": 2.892050918362872e-06,
      "loss": 0.88322121,
      "num_input_tokens_seen": 66504785,
      "step": 3096,
      "time_per_iteration": 2.674198627471924
    },
    {
      "auxiliary_loss_clip": 0.01132157,
      "auxiliary_loss_mlp": 0.0100221,
      "balance_loss_clip": 1.02891827,
      "balance_loss_mlp": 1.00111294,
      "epoch": 0.3723922323092647,
      "flos": 62419891363200.0,
      "grad_norm": 0.8558340038258414,
      "language_loss": 0.5588944,
      "learning_rate": 2.8913536553020626e-06,
      "loss": 0.58023804,
      "num_input_tokens_seen": 66558840,
      "step": 3097,
      "time_per_iteration": 3.3310141563415527
    },
    {
      "auxiliary_loss_clip": 0.01194671,
      "auxiliary_loss_mlp": 0.01025239,
      "balance_loss_clip": 1.05083585,
      "balance_loss_mlp": 1.01700163,
      "epoch": 0.3725124751999038,
      "flos": 23039747988480.0,
      "grad_norm": 2.0494224833003756,
      "language_loss": 0.85270667,
      "learning_rate": 2.8906562570236137e-06,
      "loss": 0.87490582,
      "num_input_tokens_seen": 66576750,
      "step": 3098,
      "time_per_iteration": 2.921712636947632
    },
    {
      "auxiliary_loss_clip": 0.01201449,
      "auxiliary_loss_mlp": 0.0102875,
      "balance_loss_clip": 1.04904187,
      "balance_loss_mlp": 1.02095962,
      "epoch": 0.3726327180905429,
      "flos": 20920551431040.0,
      "grad_norm": 1.6254233268581546,
      "language_loss": 0.76419544,
      "learning_rate": 2.889958723633318e-06,
      "loss": 0.78649741,
      "num_input_tokens_seen": 66595690,
      "step": 3099,
      "time_per_iteration": 2.763690710067749
    },
    {
      "auxiliary_loss_clip": 0.01199698,
      "auxiliary_loss_mlp": 0.01025474,
      "balance_loss_clip": 1.05510604,
      "balance_loss_mlp": 1.01754022,
      "epoch": 0.372752960981182,
      "flos": 30592156688640.0,
      "grad_norm": 2.084786643633297,
      "language_loss": 0.73783612,
      "learning_rate": 2.889261055236992e-06,
      "loss": 0.76008785,
      "num_input_tokens_seen": 66617905,
      "step": 3100,
      "time_per_iteration": 2.763310432434082
    },
    {
      "auxiliary_loss_clip": 0.01194461,
      "auxiliary_loss_mlp": 0.01024922,
      "balance_loss_clip": 1.05648518,
      "balance_loss_mlp": 1.0167619,
      "epoch": 0.3728732038718211,
      "flos": 25116749043840.0,
      "grad_norm": 1.8177403269782708,
      "language_loss": 0.82521999,
      "learning_rate": 2.8885632519404704e-06,
      "loss": 0.84741384,
      "num_input_tokens_seen": 66638175,
      "step": 3101,
      "time_per_iteration": 2.7434942722320557
    },
    {
      "auxiliary_loss_clip": 0.01195147,
      "auxiliary_loss_mlp": 0.01025955,
      "balance_loss_clip": 1.0556879,
      "balance_loss_mlp": 1.01731241,
      "epoch": 0.37299344676246016,
      "flos": 25302048330240.0,
      "grad_norm": 1.999587520073078,
      "language_loss": 0.75530279,
      "learning_rate": 2.8878653138496107e-06,
      "loss": 0.77751386,
      "num_input_tokens_seen": 66658670,
      "step": 3102,
      "time_per_iteration": 2.7040798664093018
    },
    {
      "auxiliary_loss_clip": 0.01199385,
      "auxiliary_loss_mlp": 0.01027377,
      "balance_loss_clip": 1.04537606,
      "balance_loss_mlp": 1.0188117,
      "epoch": 0.37311368965309927,
      "flos": 23842531002240.0,
      "grad_norm": 2.441975811775521,
      "language_loss": 0.76823878,
      "learning_rate": 2.8871672410702878e-06,
      "loss": 0.79050636,
      "num_input_tokens_seen": 66676030,
      "step": 3103,
      "time_per_iteration": 2.795614242553711
    },
    {
      "auxiliary_loss_clip": 0.01208246,
      "auxiliary_loss_mlp": 0.01027603,
      "balance_loss_clip": 1.05293393,
      "balance_loss_mlp": 1.01835823,
      "epoch": 0.3732339325437384,
      "flos": 25811943845760.0,
      "grad_norm": 1.8622009736766023,
      "language_loss": 0.82370383,
      "learning_rate": 2.8864690337084008e-06,
      "loss": 0.8460623,
      "num_input_tokens_seen": 66695305,
      "step": 3104,
      "time_per_iteration": 2.7107934951782227
    },
    {
      "auxiliary_loss_clip": 0.01189787,
      "auxiliary_loss_mlp": 0.01027623,
      "balance_loss_clip": 1.05472994,
      "balance_loss_mlp": 1.0186224,
      "epoch": 0.37335417543437743,
      "flos": 26208433146240.0,
      "grad_norm": 2.2975446515482005,
      "language_loss": 0.77999008,
      "learning_rate": 2.885770691869866e-06,
      "loss": 0.80216408,
      "num_input_tokens_seen": 66716185,
      "step": 3105,
      "time_per_iteration": 2.7270443439483643
    },
    {
      "auxiliary_loss_clip": 0.01191851,
      "auxiliary_loss_mlp": 0.01029572,
      "balance_loss_clip": 1.05690956,
      "balance_loss_mlp": 1.02156758,
      "epoch": 0.37347441832501654,
      "flos": 24023879792640.0,
      "grad_norm": 2.538250846260206,
      "language_loss": 0.73815024,
      "learning_rate": 2.8850722156606207e-06,
      "loss": 0.76036447,
      "num_input_tokens_seen": 66734575,
      "step": 3106,
      "time_per_iteration": 2.6873891353607178
    },
    {
      "auxiliary_loss_clip": 0.01189467,
      "auxiliary_loss_mlp": 0.01028473,
      "balance_loss_clip": 1.05554581,
      "balance_loss_mlp": 1.01990175,
      "epoch": 0.3735946612156556,
      "flos": 19714922409600.0,
      "grad_norm": 2.1164963018030303,
      "language_loss": 0.67282653,
      "learning_rate": 2.8843736051866252e-06,
      "loss": 0.69500601,
      "num_input_tokens_seen": 66753500,
      "step": 3107,
      "time_per_iteration": 2.73230242729187
    },
    {
      "auxiliary_loss_clip": 0.01199359,
      "auxiliary_loss_mlp": 0.01208695,
      "balance_loss_clip": 1.05090594,
      "balance_loss_mlp": 1.00091064,
      "epoch": 0.3737149041062947,
      "flos": 23039604334080.0,
      "grad_norm": 1.8739045880910479,
      "language_loss": 0.69221675,
      "learning_rate": 2.8836748605538557e-06,
      "loss": 0.71629733,
      "num_input_tokens_seen": 66775140,
      "step": 3108,
      "time_per_iteration": 2.760538101196289
    },
    {
      "auxiliary_loss_clip": 0.0120369,
      "auxiliary_loss_mlp": 0.01025314,
      "balance_loss_clip": 1.0547967,
      "balance_loss_mlp": 1.01674843,
      "epoch": 0.3738351469969338,
      "flos": 34678108483200.0,
      "grad_norm": 15.529073610659731,
      "language_loss": 0.63626963,
      "learning_rate": 2.882975981868313e-06,
      "loss": 0.65855968,
      "num_input_tokens_seen": 66795525,
      "step": 3109,
      "time_per_iteration": 2.757822036743164
    },
    {
      "auxiliary_loss_clip": 0.01196285,
      "auxiliary_loss_mlp": 0.01026743,
      "balance_loss_clip": 1.05966234,
      "balance_loss_mlp": 1.0180881,
      "epoch": 0.3739553898875729,
      "flos": 43507967448960.0,
      "grad_norm": 21.54864934536681,
      "language_loss": 0.6872651,
      "learning_rate": 2.882276969236016e-06,
      "loss": 0.70949537,
      "num_input_tokens_seen": 66816885,
      "step": 3110,
      "time_per_iteration": 4.732790946960449
    },
    {
      "auxiliary_loss_clip": 0.01195555,
      "auxiliary_loss_mlp": 0.01027062,
      "balance_loss_clip": 1.05446541,
      "balance_loss_mlp": 1.01815724,
      "epoch": 0.374075632778212,
      "flos": 12856487448960.0,
      "grad_norm": 2.5508370437963936,
      "language_loss": 0.76999652,
      "learning_rate": 2.881577822763005e-06,
      "loss": 0.79222274,
      "num_input_tokens_seen": 66834835,
      "step": 3111,
      "time_per_iteration": 2.608919143676758
    },
    {
      "auxiliary_loss_clip": 0.01194387,
      "auxiliary_loss_mlp": 0.01023897,
      "balance_loss_clip": 1.05679536,
      "balance_loss_mlp": 1.01609468,
      "epoch": 0.3741958756688511,
      "flos": 26024031699840.0,
      "grad_norm": 2.030760604779055,
      "language_loss": 0.87628633,
      "learning_rate": 2.880878542555338e-06,
      "loss": 0.89846921,
      "num_input_tokens_seen": 66852600,
      "step": 3112,
      "time_per_iteration": 2.682940721511841
    },
    {
      "auxiliary_loss_clip": 0.01195663,
      "auxiliary_loss_mlp": 0.01027537,
      "balance_loss_clip": 1.05948162,
      "balance_loss_mlp": 1.01882291,
      "epoch": 0.37431611855949015,
      "flos": 21433894652160.0,
      "grad_norm": 2.0457887819361136,
      "language_loss": 0.80409837,
      "learning_rate": 2.8801791287190976e-06,
      "loss": 0.82633042,
      "num_input_tokens_seen": 66870595,
      "step": 3113,
      "time_per_iteration": 3.5223920345306396
    },
    {
      "auxiliary_loss_clip": 0.01197201,
      "auxiliary_loss_mlp": 0.01026007,
      "balance_loss_clip": 1.05635929,
      "balance_loss_mlp": 1.01731038,
      "epoch": 0.37443636145012926,
      "flos": 24207096090240.0,
      "grad_norm": 3.010070403699019,
      "language_loss": 0.86022091,
      "learning_rate": 2.8794795813603817e-06,
      "loss": 0.88245296,
      "num_input_tokens_seen": 66886060,
      "step": 3114,
      "time_per_iteration": 2.637702226638794
    },
    {
      "auxiliary_loss_clip": 0.01201577,
      "auxiliary_loss_mlp": 0.01029029,
      "balance_loss_clip": 1.05697715,
      "balance_loss_mlp": 1.0201416,
      "epoch": 0.3745566043407684,
      "flos": 15378601841280.0,
      "grad_norm": 1.854767093430064,
      "language_loss": 0.81366599,
      "learning_rate": 2.878779900585314e-06,
      "loss": 0.83597207,
      "num_input_tokens_seen": 66903900,
      "step": 3115,
      "time_per_iteration": 2.632458448410034
    },
    {
      "auxiliary_loss_clip": 0.01205082,
      "auxiliary_loss_mlp": 0.0102902,
      "balance_loss_clip": 1.05705643,
      "balance_loss_mlp": 1.02050221,
      "epoch": 0.37467684723140743,
      "flos": 24608218245120.0,
      "grad_norm": 2.2301468868610743,
      "language_loss": 0.7530992,
      "learning_rate": 2.8780800865000336e-06,
      "loss": 0.77544022,
      "num_input_tokens_seen": 66925210,
      "step": 3116,
      "time_per_iteration": 2.6272170543670654
    },
    {
      "auxiliary_loss_clip": 0.01110768,
      "auxiliary_loss_mlp": 0.01003931,
      "balance_loss_clip": 1.03565097,
      "balance_loss_mlp": 1.00264347,
      "epoch": 0.37479709012204654,
      "flos": 64377491610240.0,
      "grad_norm": 0.9820476399912408,
      "language_loss": 0.59225452,
      "learning_rate": 2.877380139210702e-06,
      "loss": 0.61340153,
      "num_input_tokens_seen": 66983880,
      "step": 3117,
      "time_per_iteration": 3.1107370853424072
    },
    {
      "auxiliary_loss_clip": 0.01202792,
      "auxiliary_loss_mlp": 0.01031285,
      "balance_loss_clip": 1.05501056,
      "balance_loss_mlp": 1.02152157,
      "epoch": 0.37491733301268565,
      "flos": 23803962773760.0,
      "grad_norm": 1.9785408455850508,
      "language_loss": 0.75986958,
      "learning_rate": 2.876680058823501e-06,
      "loss": 0.78221035,
      "num_input_tokens_seen": 67004280,
      "step": 3118,
      "time_per_iteration": 3.802812337875366
    },
    {
      "auxiliary_loss_clip": 0.01190589,
      "auxiliary_loss_mlp": 0.01026155,
      "balance_loss_clip": 1.05462527,
      "balance_loss_mlp": 1.01702929,
      "epoch": 0.3750375759033247,
      "flos": 32160950167680.0,
      "grad_norm": 1.9747112575781496,
      "language_loss": 0.66029525,
      "learning_rate": 2.8759798454446314e-06,
      "loss": 0.68246269,
      "num_input_tokens_seen": 67027445,
      "step": 3119,
      "time_per_iteration": 2.7844772338867188
    },
    {
      "auxiliary_loss_clip": 0.0119888,
      "auxiliary_loss_mlp": 0.01035027,
      "balance_loss_clip": 1.05686688,
      "balance_loss_mlp": 1.02677774,
      "epoch": 0.3751578187939638,
      "flos": 23367791923200.0,
      "grad_norm": 1.9877408131652896,
      "language_loss": 0.81374955,
      "learning_rate": 2.8752794991803173e-06,
      "loss": 0.8360886,
      "num_input_tokens_seen": 67045130,
      "step": 3120,
      "time_per_iteration": 2.6889779567718506
    },
    {
      "auxiliary_loss_clip": 0.01198661,
      "auxiliary_loss_mlp": 0.01034697,
      "balance_loss_clip": 1.05635953,
      "balance_loss_mlp": 1.02589297,
      "epoch": 0.37527806168460287,
      "flos": 14605731878400.0,
      "grad_norm": 5.186340062720625,
      "language_loss": 0.74930775,
      "learning_rate": 2.8745790201367976e-06,
      "loss": 0.77164137,
      "num_input_tokens_seen": 67060885,
      "step": 3121,
      "time_per_iteration": 2.638423442840576
    },
    {
      "auxiliary_loss_clip": 0.01197562,
      "auxiliary_loss_mlp": 0.01031885,
      "balance_loss_clip": 1.05963087,
      "balance_loss_mlp": 1.02265787,
      "epoch": 0.375398304575242,
      "flos": 26390823431040.0,
      "grad_norm": 4.8765065631407865,
      "language_loss": 0.84117758,
      "learning_rate": 2.8738784084203373e-06,
      "loss": 0.8634721,
      "num_input_tokens_seen": 67080960,
      "step": 3122,
      "time_per_iteration": 2.714568853378296
    },
    {
      "auxiliary_loss_clip": 0.01190731,
      "auxiliary_loss_mlp": 0.01027337,
      "balance_loss_clip": 1.05158782,
      "balance_loss_mlp": 1.01915896,
      "epoch": 0.3755185474658811,
      "flos": 22236605838720.0,
      "grad_norm": 1.7228578448753342,
      "language_loss": 0.78923452,
      "learning_rate": 2.873177664137216e-06,
      "loss": 0.8114152,
      "num_input_tokens_seen": 67101890,
      "step": 3123,
      "time_per_iteration": 2.6983871459960938
    },
    {
      "auxiliary_loss_clip": 0.01196908,
      "auxiliary_loss_mlp": 0.01022849,
      "balance_loss_clip": 1.05567825,
      "balance_loss_mlp": 1.01421857,
      "epoch": 0.37563879035652015,
      "flos": 30812935633920.0,
      "grad_norm": 2.0294584943068714,
      "language_loss": 0.69490457,
      "learning_rate": 2.8724767873937384e-06,
      "loss": 0.71710217,
      "num_input_tokens_seen": 67126010,
      "step": 3124,
      "time_per_iteration": 2.8038952350616455
    },
    {
      "auxiliary_loss_clip": 0.01200781,
      "auxiliary_loss_mlp": 0.01034673,
      "balance_loss_clip": 1.05656719,
      "balance_loss_mlp": 1.02613127,
      "epoch": 0.37575903324715926,
      "flos": 20773533064320.0,
      "grad_norm": 2.886279562444464,
      "language_loss": 0.87750608,
      "learning_rate": 2.871775778296225e-06,
      "loss": 0.89986062,
      "num_input_tokens_seen": 67143100,
      "step": 3125,
      "time_per_iteration": 2.675503730773926
    },
    {
      "auxiliary_loss_clip": 0.01199876,
      "auxiliary_loss_mlp": 0.01032845,
      "balance_loss_clip": 1.0611856,
      "balance_loss_mlp": 1.02293885,
      "epoch": 0.37587927613779837,
      "flos": 18697681244160.0,
      "grad_norm": 2.2245960897870876,
      "language_loss": 0.77917451,
      "learning_rate": 2.8710746369510196e-06,
      "loss": 0.80150169,
      "num_input_tokens_seen": 67161085,
      "step": 3126,
      "time_per_iteration": 2.6330044269561768
    },
    {
      "auxiliary_loss_clip": 0.01195197,
      "auxiliary_loss_mlp": 0.01028664,
      "balance_loss_clip": 1.05719757,
      "balance_loss_mlp": 1.01985419,
      "epoch": 0.3759995190284374,
      "flos": 13624796384640.0,
      "grad_norm": 2.472250628596378,
      "language_loss": 0.83567226,
      "learning_rate": 2.8703733634644846e-06,
      "loss": 0.85791087,
      "num_input_tokens_seen": 67175840,
      "step": 3127,
      "time_per_iteration": 2.675363063812256
    },
    {
      "auxiliary_loss_clip": 0.0119329,
      "auxiliary_loss_mlp": 0.01026207,
      "balance_loss_clip": 1.05924773,
      "balance_loss_mlp": 1.0176177,
      "epoch": 0.37611976191907653,
      "flos": 20484847457280.0,
      "grad_norm": 2.4033661223997576,
      "language_loss": 0.79156566,
      "learning_rate": 2.869671957943002e-06,
      "loss": 0.81376064,
      "num_input_tokens_seen": 67194995,
      "step": 3128,
      "time_per_iteration": 2.6222617626190186
    },
    {
      "auxiliary_loss_clip": 0.01192565,
      "auxiliary_loss_mlp": 0.01027084,
      "balance_loss_clip": 1.05907023,
      "balance_loss_mlp": 1.01849532,
      "epoch": 0.37624000480971564,
      "flos": 21141797253120.0,
      "grad_norm": 3.401560468955437,
      "language_loss": 0.74113709,
      "learning_rate": 2.8689704204929747e-06,
      "loss": 0.76333362,
      "num_input_tokens_seen": 67214175,
      "step": 3129,
      "time_per_iteration": 2.658482551574707
    },
    {
      "auxiliary_loss_clip": 0.01195177,
      "auxiliary_loss_mlp": 0.01026736,
      "balance_loss_clip": 1.05936098,
      "balance_loss_mlp": 1.01796484,
      "epoch": 0.3763602477003547,
      "flos": 22564470205440.0,
      "grad_norm": 1.8141152987275808,
      "language_loss": 0.81103665,
      "learning_rate": 2.8682687512208253e-06,
      "loss": 0.83325577,
      "num_input_tokens_seen": 67233185,
      "step": 3130,
      "time_per_iteration": 2.6427037715911865
    },
    {
      "auxiliary_loss_clip": 0.01200241,
      "auxiliary_loss_mlp": 0.01030553,
      "balance_loss_clip": 1.05551434,
      "balance_loss_mlp": 1.02146316,
      "epoch": 0.3764804905909938,
      "flos": 27526857851520.0,
      "grad_norm": 2.411629486199357,
      "language_loss": 0.80772495,
      "learning_rate": 2.8675669502329972e-06,
      "loss": 0.83003283,
      "num_input_tokens_seen": 67254715,
      "step": 3131,
      "time_per_iteration": 2.734145402908325
    },
    {
      "auxiliary_loss_clip": 0.01198031,
      "auxiliary_loss_mlp": 0.01209223,
      "balance_loss_clip": 1.05820823,
      "balance_loss_mlp": 1.00080109,
      "epoch": 0.3766007334816329,
      "flos": 22528092706560.0,
      "grad_norm": 6.312372680667023,
      "language_loss": 0.85407424,
      "learning_rate": 2.866865017635952e-06,
      "loss": 0.87814677,
      "num_input_tokens_seen": 67272535,
      "step": 3132,
      "time_per_iteration": 2.6861319541931152
    },
    {
      "auxiliary_loss_clip": 0.011964,
      "auxiliary_loss_mlp": 0.01024587,
      "balance_loss_clip": 1.05430746,
      "balance_loss_mlp": 1.01589108,
      "epoch": 0.376720976372272,
      "flos": 25957166532480.0,
      "grad_norm": 1.732516343218208,
      "language_loss": 0.79430461,
      "learning_rate": 2.866162953536174e-06,
      "loss": 0.81651449,
      "num_input_tokens_seen": 67293505,
      "step": 3133,
      "time_per_iteration": 2.770503282546997
    },
    {
      "auxiliary_loss_clip": 0.01196416,
      "auxiliary_loss_mlp": 0.01208958,
      "balance_loss_clip": 1.05412424,
      "balance_loss_mlp": 1.00087905,
      "epoch": 0.3768412192629111,
      "flos": 18041162411520.0,
      "grad_norm": 1.624423989246287,
      "language_loss": 0.75168216,
      "learning_rate": 2.8654607580401634e-06,
      "loss": 0.77573586,
      "num_input_tokens_seen": 67313240,
      "step": 3134,
      "time_per_iteration": 2.6615383625030518
    },
    {
      "auxiliary_loss_clip": 0.01110063,
      "auxiliary_loss_mlp": 0.01002834,
      "balance_loss_clip": 1.03659439,
      "balance_loss_mlp": 1.00148666,
      "epoch": 0.3769614621535502,
      "flos": 62989472304000.0,
      "grad_norm": 0.8978941707245649,
      "language_loss": 0.65186203,
      "learning_rate": 2.8647584312544446e-06,
      "loss": 0.67299104,
      "num_input_tokens_seen": 67378445,
      "step": 3135,
      "time_per_iteration": 3.201817750930786
    },
    {
      "auxiliary_loss_clip": 0.01194571,
      "auxiliary_loss_mlp": 0.012089,
      "balance_loss_clip": 1.0512104,
      "balance_loss_mlp": 1.00089669,
      "epoch": 0.37708170504418925,
      "flos": 23661685002240.0,
      "grad_norm": 1.6065122433114465,
      "language_loss": 0.85379195,
      "learning_rate": 2.864055973285559e-06,
      "loss": 0.87782669,
      "num_input_tokens_seen": 67400445,
      "step": 3136,
      "time_per_iteration": 2.7534377574920654
    },
    {
      "auxiliary_loss_clip": 0.0118789,
      "auxiliary_loss_mlp": 0.01031787,
      "balance_loss_clip": 1.05318475,
      "balance_loss_mlp": 1.02265525,
      "epoch": 0.37720194793482836,
      "flos": 24423170353920.0,
      "grad_norm": 2.0095764092533805,
      "language_loss": 0.86547607,
      "learning_rate": 2.8633533842400698e-06,
      "loss": 0.88767278,
      "num_input_tokens_seen": 67420645,
      "step": 3137,
      "time_per_iteration": 4.548182249069214
    },
    {
      "auxiliary_loss_clip": 0.01199027,
      "auxiliary_loss_mlp": 0.01209362,
      "balance_loss_clip": 1.058568,
      "balance_loss_mlp": 1.00088406,
      "epoch": 0.3773221908254674,
      "flos": 20996502739200.0,
      "grad_norm": 2.707649815508903,
      "language_loss": 0.77311748,
      "learning_rate": 2.862650664224558e-06,
      "loss": 0.7972014,
      "num_input_tokens_seen": 67439495,
      "step": 3138,
      "time_per_iteration": 2.6305370330810547
    },
    {
      "auxiliary_loss_clip": 0.01192112,
      "auxiliary_loss_mlp": 0.01027038,
      "balance_loss_clip": 1.0589354,
      "balance_loss_mlp": 1.0188663,
      "epoch": 0.37744243371610653,
      "flos": 37631724958080.0,
      "grad_norm": 1.470497522538082,
      "language_loss": 0.69924223,
      "learning_rate": 2.861947813345627e-06,
      "loss": 0.7214337,
      "num_input_tokens_seen": 67462195,
      "step": 3139,
      "time_per_iteration": 2.7524468898773193
    },
    {
      "auxiliary_loss_clip": 0.01197702,
      "auxiliary_loss_mlp": 0.01209046,
      "balance_loss_clip": 1.06102705,
      "balance_loss_mlp": 1.00068998,
      "epoch": 0.37756267660674564,
      "flos": 26140526484480.0,
      "grad_norm": 2.299652581492006,
      "language_loss": 0.72590661,
      "learning_rate": 2.8612448317098974e-06,
      "loss": 0.74997401,
      "num_input_tokens_seen": 67482530,
      "step": 3140,
      "time_per_iteration": 3.5835788249969482
    },
    {
      "auxiliary_loss_clip": 0.01203158,
      "auxiliary_loss_mlp": 0.01209316,
      "balance_loss_clip": 1.05189157,
      "balance_loss_mlp": 1.00079942,
      "epoch": 0.3776829194973847,
      "flos": 19427888828160.0,
      "grad_norm": 5.6684915320897336,
      "language_loss": 0.83508039,
      "learning_rate": 2.8605417194240114e-06,
      "loss": 0.85920513,
      "num_input_tokens_seen": 67500890,
      "step": 3141,
      "time_per_iteration": 2.7177464962005615
    },
    {
      "auxiliary_loss_clip": 0.01190432,
      "auxiliary_loss_mlp": 0.01026529,
      "balance_loss_clip": 1.05438185,
      "balance_loss_mlp": 1.01826751,
      "epoch": 0.3778031623880238,
      "flos": 17382309194880.0,
      "grad_norm": 1.8559566066219664,
      "language_loss": 0.78912902,
      "learning_rate": 2.8598384765946315e-06,
      "loss": 0.81129861,
      "num_input_tokens_seen": 67519545,
      "step": 3142,
      "time_per_iteration": 2.619947910308838
    },
    {
      "auxiliary_loss_clip": 0.01192249,
      "auxiliary_loss_mlp": 0.01029118,
      "balance_loss_clip": 1.05552006,
      "balance_loss_mlp": 1.0203321,
      "epoch": 0.3779234052786629,
      "flos": 27125843437440.0,
      "grad_norm": 2.1273547278422953,
      "language_loss": 0.71606112,
      "learning_rate": 2.8591351033284377e-06,
      "loss": 0.73827475,
      "num_input_tokens_seen": 67539275,
      "step": 3143,
      "time_per_iteration": 2.6699252128601074
    },
    {
      "auxiliary_loss_clip": 0.0119931,
      "auxiliary_loss_mlp": 0.01027129,
      "balance_loss_clip": 1.05625296,
      "balance_loss_mlp": 1.01839066,
      "epoch": 0.37804364816930197,
      "flos": 19682639061120.0,
      "grad_norm": 2.092660796983505,
      "language_loss": 0.8384822,
      "learning_rate": 2.8584315997321325e-06,
      "loss": 0.8607465,
      "num_input_tokens_seen": 67558280,
      "step": 3144,
      "time_per_iteration": 3.542606830596924
    },
    {
      "auxiliary_loss_clip": 0.01193613,
      "auxiliary_loss_mlp": 0.01026359,
      "balance_loss_clip": 1.05710053,
      "balance_loss_mlp": 1.0172987,
      "epoch": 0.3781638910599411,
      "flos": 22702905221760.0,
      "grad_norm": 3.4883802639390007,
      "language_loss": 0.77603042,
      "learning_rate": 2.8577279659124356e-06,
      "loss": 0.79823011,
      "num_input_tokens_seen": 67575955,
      "step": 3145,
      "time_per_iteration": 2.623178482055664
    },
    {
      "auxiliary_loss_clip": 0.01190659,
      "auxiliary_loss_mlp": 0.01026852,
      "balance_loss_clip": 1.05365157,
      "balance_loss_mlp": 1.01923418,
      "epoch": 0.3782841339505802,
      "flos": 14647604158080.0,
      "grad_norm": 3.311812286811543,
      "language_loss": 0.83640516,
      "learning_rate": 2.857024201976089e-06,
      "loss": 0.85858023,
      "num_input_tokens_seen": 67593515,
      "step": 3146,
      "time_per_iteration": 2.6333136558532715
    },
    {
      "auxiliary_loss_clip": 0.01197956,
      "auxiliary_loss_mlp": 0.010253,
      "balance_loss_clip": 1.05846119,
      "balance_loss_mlp": 1.01606703,
      "epoch": 0.37840437684121925,
      "flos": 32818223185920.0,
      "grad_norm": 2.303838028181616,
      "language_loss": 0.73190296,
      "learning_rate": 2.8563203080298516e-06,
      "loss": 0.75413549,
      "num_input_tokens_seen": 67614290,
      "step": 3147,
      "time_per_iteration": 2.764551877975464
    },
    {
      "auxiliary_loss_clip": 0.01199291,
      "auxiliary_loss_mlp": 0.01209244,
      "balance_loss_clip": 1.05646229,
      "balance_loss_mlp": 1.00073147,
      "epoch": 0.37852461973185836,
      "flos": 18369206346240.0,
      "grad_norm": 2.4074101742486262,
      "language_loss": 0.88966274,
      "learning_rate": 2.855616284180505e-06,
      "loss": 0.91374803,
      "num_input_tokens_seen": 67631340,
      "step": 3148,
      "time_per_iteration": 2.676301956176758
    },
    {
      "auxiliary_loss_clip": 0.01110556,
      "auxiliary_loss_mlp": 0.01002797,
      "balance_loss_clip": 1.03418994,
      "balance_loss_mlp": 1.00155759,
      "epoch": 0.37864486262249747,
      "flos": 59500680117120.0,
      "grad_norm": 0.8911349994857323,
      "language_loss": 0.66081834,
      "learning_rate": 2.8549121305348477e-06,
      "loss": 0.68195188,
      "num_input_tokens_seen": 67691125,
      "step": 3149,
      "time_per_iteration": 3.1207282543182373
    },
    {
      "auxiliary_loss_clip": 0.01196048,
      "auxiliary_loss_mlp": 0.01027554,
      "balance_loss_clip": 1.05583096,
      "balance_loss_mlp": 1.01938796,
      "epoch": 0.3787651055131365,
      "flos": 23363015414400.0,
      "grad_norm": 4.397301260507727,
      "language_loss": 0.83370459,
      "learning_rate": 2.8542078471997006e-06,
      "loss": 0.85594058,
      "num_input_tokens_seen": 67708740,
      "step": 3150,
      "time_per_iteration": 2.6793744564056396
    },
    {
      "auxiliary_loss_clip": 0.0119654,
      "auxiliary_loss_mlp": 0.01025584,
      "balance_loss_clip": 1.05690122,
      "balance_loss_mlp": 1.01719129,
      "epoch": 0.37888534840377563,
      "flos": 24601394661120.0,
      "grad_norm": 1.776651393820303,
      "language_loss": 0.7590096,
      "learning_rate": 2.8535034342819013e-06,
      "loss": 0.78123087,
      "num_input_tokens_seen": 67726150,
      "step": 3151,
      "time_per_iteration": 2.6428110599517822
    },
    {
      "auxiliary_loss_clip": 0.01191746,
      "auxiliary_loss_mlp": 0.01032807,
      "balance_loss_clip": 1.05775428,
      "balance_loss_mlp": 1.02370501,
      "epoch": 0.37900559129441475,
      "flos": 23986891762560.0,
      "grad_norm": 1.6135032148620834,
      "language_loss": 0.72685993,
      "learning_rate": 2.85279889188831e-06,
      "loss": 0.74910545,
      "num_input_tokens_seen": 67746525,
      "step": 3152,
      "time_per_iteration": 2.695194721221924
    },
    {
      "auxiliary_loss_clip": 0.01202291,
      "auxiliary_loss_mlp": 0.01027134,
      "balance_loss_clip": 1.05233467,
      "balance_loss_mlp": 1.01794887,
      "epoch": 0.3791258341850538,
      "flos": 24644667571200.0,
      "grad_norm": 2.0492910852109407,
      "language_loss": 0.8106575,
      "learning_rate": 2.852094220125805e-06,
      "loss": 0.83295172,
      "num_input_tokens_seen": 67766035,
      "step": 3153,
      "time_per_iteration": 2.790374279022217
    },
    {
      "auxiliary_loss_clip": 0.01200316,
      "auxiliary_loss_mlp": 0.01035284,
      "balance_loss_clip": 1.05916488,
      "balance_loss_mlp": 1.0260396,
      "epoch": 0.3792460770756929,
      "flos": 17420841509760.0,
      "grad_norm": 2.5597244355027344,
      "language_loss": 0.71110243,
      "learning_rate": 2.8513894191012846e-06,
      "loss": 0.7334584,
      "num_input_tokens_seen": 67785015,
      "step": 3154,
      "time_per_iteration": 2.6008312702178955
    },
    {
      "auxiliary_loss_clip": 0.01197757,
      "auxiliary_loss_mlp": 0.01030904,
      "balance_loss_clip": 1.06078458,
      "balance_loss_mlp": 1.02171326,
      "epoch": 0.37936631996633197,
      "flos": 24206557386240.0,
      "grad_norm": 2.1702063005288643,
      "language_loss": 0.78798735,
      "learning_rate": 2.8506844889216664e-06,
      "loss": 0.810274,
      "num_input_tokens_seen": 67804400,
      "step": 3155,
      "time_per_iteration": 2.6359126567840576
    },
    {
      "auxiliary_loss_clip": 0.01104723,
      "auxiliary_loss_mlp": 0.01001418,
      "balance_loss_clip": 1.03603005,
      "balance_loss_mlp": 1.00036907,
      "epoch": 0.3794865628569711,
      "flos": 70297114752000.0,
      "grad_norm": 0.8638874743627012,
      "language_loss": 0.62779391,
      "learning_rate": 2.849979429693887e-06,
      "loss": 0.64885533,
      "num_input_tokens_seen": 67865385,
      "step": 3156,
      "time_per_iteration": 3.2408018112182617
    },
    {
      "auxiliary_loss_clip": 0.01193337,
      "auxiliary_loss_mlp": 0.01027815,
      "balance_loss_clip": 1.0587697,
      "balance_loss_mlp": 1.01896942,
      "epoch": 0.3796068057476102,
      "flos": 15779364860160.0,
      "grad_norm": 5.040934315468317,
      "language_loss": 0.74250543,
      "learning_rate": 2.8492742415249042e-06,
      "loss": 0.76471686,
      "num_input_tokens_seen": 67883030,
      "step": 3157,
      "time_per_iteration": 2.659749746322632
    },
    {
      "auxiliary_loss_clip": 0.0119282,
      "auxiliary_loss_mlp": 0.01024801,
      "balance_loss_clip": 1.05692017,
      "balance_loss_mlp": 1.01611114,
      "epoch": 0.37972704863824924,
      "flos": 25191694771200.0,
      "grad_norm": 1.7446970075199475,
      "language_loss": 0.76233405,
      "learning_rate": 2.848568924521694e-06,
      "loss": 0.78451025,
      "num_input_tokens_seen": 67903810,
      "step": 3158,
      "time_per_iteration": 2.6311049461364746
    },
    {
      "auxiliary_loss_clip": 0.01188985,
      "auxiliary_loss_mlp": 0.01024239,
      "balance_loss_clip": 1.05280852,
      "balance_loss_mlp": 1.0149585,
      "epoch": 0.37984729152888835,
      "flos": 26210372480640.0,
      "grad_norm": 2.601593087458344,
      "language_loss": 0.73520172,
      "learning_rate": 2.8478634787912526e-06,
      "loss": 0.75733393,
      "num_input_tokens_seen": 67921865,
      "step": 3159,
      "time_per_iteration": 2.713628053665161
    },
    {
      "auxiliary_loss_clip": 0.01195748,
      "auxiliary_loss_mlp": 0.01025647,
      "balance_loss_clip": 1.0542804,
      "balance_loss_mlp": 1.01676011,
      "epoch": 0.37996753441952746,
      "flos": 25629302165760.0,
      "grad_norm": 2.032257942374533,
      "language_loss": 0.76560247,
      "learning_rate": 2.847157904440596e-06,
      "loss": 0.78781641,
      "num_input_tokens_seen": 67941595,
      "step": 3160,
      "time_per_iteration": 2.648646593093872
    },
    {
      "auxiliary_loss_clip": 0.01197123,
      "auxiliary_loss_mlp": 0.01026841,
      "balance_loss_clip": 1.05684149,
      "balance_loss_mlp": 1.01825774,
      "epoch": 0.3800877773101665,
      "flos": 20118414862080.0,
      "grad_norm": 1.5467269945267308,
      "language_loss": 0.73610389,
      "learning_rate": 2.846452201576759e-06,
      "loss": 0.75834352,
      "num_input_tokens_seen": 67960970,
      "step": 3161,
      "time_per_iteration": 2.6490366458892822
    },
    {
      "auxiliary_loss_clip": 0.01117285,
      "auxiliary_loss_mlp": 0.01003203,
      "balance_loss_clip": 1.03342319,
      "balance_loss_mlp": 1.00186205,
      "epoch": 0.38020802020080563,
      "flos": 63053608037760.0,
      "grad_norm": 0.8551529015778377,
      "language_loss": 0.62712777,
      "learning_rate": 2.845746370306795e-06,
      "loss": 0.64833266,
      "num_input_tokens_seen": 68026160,
      "step": 3162,
      "time_per_iteration": 3.3392717838287354
    },
    {
      "auxiliary_loss_clip": 0.01199875,
      "auxiliary_loss_mlp": 0.01026639,
      "balance_loss_clip": 1.05877566,
      "balance_loss_mlp": 1.01791906,
      "epoch": 0.38032826309144474,
      "flos": 21288420570240.0,
      "grad_norm": 2.541211299628048,
      "language_loss": 0.78514957,
      "learning_rate": 2.84504041073778e-06,
      "loss": 0.80741477,
      "num_input_tokens_seen": 68044575,
      "step": 3163,
      "time_per_iteration": 3.678184986114502
    },
    {
      "auxiliary_loss_clip": 0.01194397,
      "auxiliary_loss_mlp": 0.01034729,
      "balance_loss_clip": 1.05591035,
      "balance_loss_mlp": 1.02482843,
      "epoch": 0.3804485059820838,
      "flos": 18954119416320.0,
      "grad_norm": 3.0862844078752945,
      "language_loss": 0.79296899,
      "learning_rate": 2.844334322976806e-06,
      "loss": 0.81526029,
      "num_input_tokens_seen": 68064790,
      "step": 3164,
      "time_per_iteration": 3.7182695865631104
    },
    {
      "auxiliary_loss_clip": 0.01203518,
      "auxiliary_loss_mlp": 0.01032416,
      "balance_loss_clip": 1.05028486,
      "balance_loss_mlp": 1.02305257,
      "epoch": 0.3805687488727229,
      "flos": 21833759831040.0,
      "grad_norm": 2.263241279966576,
      "language_loss": 0.83570576,
      "learning_rate": 2.8436281071309866e-06,
      "loss": 0.85806513,
      "num_input_tokens_seen": 68083330,
      "step": 3165,
      "time_per_iteration": 2.7552599906921387
    },
    {
      "auxiliary_loss_clip": 0.01126952,
      "auxiliary_loss_mlp": 0.01003072,
      "balance_loss_clip": 1.02952588,
      "balance_loss_mlp": 1.00164735,
      "epoch": 0.380688991763362,
      "flos": 58546209968640.0,
      "grad_norm": 0.7307500190228895,
      "language_loss": 0.52955323,
      "learning_rate": 2.842921763307455e-06,
      "loss": 0.55085343,
      "num_input_tokens_seen": 68146140,
      "step": 3166,
      "time_per_iteration": 3.1916110515594482
    },
    {
      "auxiliary_loss_clip": 0.01193255,
      "auxiliary_loss_mlp": 0.01025825,
      "balance_loss_clip": 1.05445814,
      "balance_loss_mlp": 1.01705706,
      "epoch": 0.38080923465400107,
      "flos": 23799509487360.0,
      "grad_norm": 1.8020290137755688,
      "language_loss": 0.82615012,
      "learning_rate": 2.842215291613361e-06,
      "loss": 0.84834087,
      "num_input_tokens_seen": 68164520,
      "step": 3167,
      "time_per_iteration": 3.6528031826019287
    },
    {
      "auxiliary_loss_clip": 0.01120671,
      "auxiliary_loss_mlp": 0.0100475,
      "balance_loss_clip": 1.02619994,
      "balance_loss_mlp": 1.00317609,
      "epoch": 0.3809294775446402,
      "flos": 54969866380800.0,
      "grad_norm": 0.8369614713530463,
      "language_loss": 0.59167069,
      "learning_rate": 2.8415086921558774e-06,
      "loss": 0.61292487,
      "num_input_tokens_seen": 68227945,
      "step": 3168,
      "time_per_iteration": 3.391669750213623
    },
    {
      "auxiliary_loss_clip": 0.01181203,
      "auxiliary_loss_mlp": 0.01025954,
      "balance_loss_clip": 1.0469656,
      "balance_loss_mlp": 1.01715016,
      "epoch": 0.38104972043527924,
      "flos": 24643697904000.0,
      "grad_norm": 1.5313106260473486,
      "language_loss": 0.78861701,
      "learning_rate": 2.840801965042194e-06,
      "loss": 0.8106885,
      "num_input_tokens_seen": 68247405,
      "step": 3169,
      "time_per_iteration": 2.9844985008239746
    },
    {
      "auxiliary_loss_clip": 0.01188343,
      "auxiliary_loss_mlp": 0.01025973,
      "balance_loss_clip": 1.05035889,
      "balance_loss_mlp": 1.01635873,
      "epoch": 0.38116996332591835,
      "flos": 22856783086080.0,
      "grad_norm": 2.5471290580475965,
      "language_loss": 0.84261149,
      "learning_rate": 2.840095110379521e-06,
      "loss": 0.86475468,
      "num_input_tokens_seen": 68266925,
      "step": 3170,
      "time_per_iteration": 2.643575429916382
    },
    {
      "auxiliary_loss_clip": 0.01114756,
      "auxiliary_loss_mlp": 0.01000637,
      "balance_loss_clip": 1.02391076,
      "balance_loss_mlp": 0.99933147,
      "epoch": 0.38129020621655746,
      "flos": 60836160804480.0,
      "grad_norm": 0.7432407593990914,
      "language_loss": 0.53910279,
      "learning_rate": 2.8393881282750884e-06,
      "loss": 0.56025672,
      "num_input_tokens_seen": 68329755,
      "step": 3171,
      "time_per_iteration": 4.153020858764648
    },
    {
      "auxiliary_loss_clip": 0.01198342,
      "auxiliary_loss_mlp": 0.01025776,
      "balance_loss_clip": 1.05709028,
      "balance_loss_mlp": 1.01638865,
      "epoch": 0.3814104491071965,
      "flos": 21648101408640.0,
      "grad_norm": 2.997886323011289,
      "language_loss": 0.78887159,
      "learning_rate": 2.838681018836144e-06,
      "loss": 0.81111276,
      "num_input_tokens_seen": 68347075,
      "step": 3172,
      "time_per_iteration": 2.6753132343292236
    },
    {
      "auxiliary_loss_clip": 0.01200489,
      "auxiliary_loss_mlp": 0.0120894,
      "balance_loss_clip": 1.04990697,
      "balance_loss_mlp": 1.00067568,
      "epoch": 0.3815306919978356,
      "flos": 19099090707840.0,
      "grad_norm": 2.3922783014192213,
      "language_loss": 0.78595138,
      "learning_rate": 2.837973782169955e-06,
      "loss": 0.8100456,
      "num_input_tokens_seen": 68365450,
      "step": 3173,
      "time_per_iteration": 2.6806766986846924
    },
    {
      "auxiliary_loss_clip": 0.01097276,
      "auxiliary_loss_mlp": 0.01002611,
      "balance_loss_clip": 1.02828348,
      "balance_loss_mlp": 1.00139511,
      "epoch": 0.38165093488847474,
      "flos": 67067918156160.0,
      "grad_norm": 0.813863331922513,
      "language_loss": 0.59154993,
      "learning_rate": 2.8372664183838096e-06,
      "loss": 0.61254877,
      "num_input_tokens_seen": 68428470,
      "step": 3174,
      "time_per_iteration": 3.231975793838501
    },
    {
      "auxiliary_loss_clip": 0.0119338,
      "auxiliary_loss_mlp": 0.01033298,
      "balance_loss_clip": 1.05798709,
      "balance_loss_mlp": 1.02371359,
      "epoch": 0.3817711777791138,
      "flos": 22341105480960.0,
      "grad_norm": 2.445223409753671,
      "language_loss": 0.68521655,
      "learning_rate": 2.836558927585015e-06,
      "loss": 0.70748329,
      "num_input_tokens_seen": 68445440,
      "step": 3175,
      "time_per_iteration": 2.6768276691436768
    },
    {
      "auxiliary_loss_clip": 0.01199271,
      "auxiliary_loss_mlp": 0.01036015,
      "balance_loss_clip": 1.05703592,
      "balance_loss_mlp": 1.02755094,
      "epoch": 0.3818914206697529,
      "flos": 22820621068800.0,
      "grad_norm": 1.9431404314260126,
      "language_loss": 0.82157469,
      "learning_rate": 2.8358513098808957e-06,
      "loss": 0.8439275,
      "num_input_tokens_seen": 68465755,
      "step": 3176,
      "time_per_iteration": 2.627587080001831
    },
    {
      "auxiliary_loss_clip": 0.0119477,
      "auxiliary_loss_mlp": 0.01028219,
      "balance_loss_clip": 1.05064297,
      "balance_loss_mlp": 1.018718,
      "epoch": 0.382011663560392,
      "flos": 24386074583040.0,
      "grad_norm": 1.6879248831491498,
      "language_loss": 0.77091396,
      "learning_rate": 2.835143565378798e-06,
      "loss": 0.79314387,
      "num_input_tokens_seen": 68486220,
      "step": 3177,
      "time_per_iteration": 2.722898483276367
    },
    {
      "auxiliary_loss_clip": 0.01201615,
      "auxiliary_loss_mlp": 0.01022327,
      "balance_loss_clip": 1.04875994,
      "balance_loss_mlp": 1.01337409,
      "epoch": 0.38213190645103107,
      "flos": 21981568296960.0,
      "grad_norm": 1.9413875379561427,
      "language_loss": 0.78777301,
      "learning_rate": 2.8344356941860847e-06,
      "loss": 0.81001246,
      "num_input_tokens_seen": 68505850,
      "step": 3178,
      "time_per_iteration": 2.7203545570373535
    },
    {
      "auxiliary_loss_clip": 0.01197,
      "auxiliary_loss_mlp": 0.01028968,
      "balance_loss_clip": 1.05344439,
      "balance_loss_mlp": 1.01986647,
      "epoch": 0.3822521493416702,
      "flos": 35516945773440.0,
      "grad_norm": 2.2096474866432065,
      "language_loss": 0.66295469,
      "learning_rate": 2.8337276964101403e-06,
      "loss": 0.6852144,
      "num_input_tokens_seen": 68526290,
      "step": 3179,
      "time_per_iteration": 2.7872588634490967
    },
    {
      "auxiliary_loss_clip": 0.0119722,
      "auxiliary_loss_mlp": 0.01033503,
      "balance_loss_clip": 1.05542803,
      "balance_loss_mlp": 1.02438378,
      "epoch": 0.3823723922323093,
      "flos": 21069904181760.0,
      "grad_norm": 1.868761583517442,
      "language_loss": 0.76249409,
      "learning_rate": 2.833019572158367e-06,
      "loss": 0.7848013,
      "num_input_tokens_seen": 68544725,
      "step": 3180,
      "time_per_iteration": 2.679438352584839
    },
    {
      "auxiliary_loss_clip": 0.01197535,
      "auxiliary_loss_mlp": 0.0102907,
      "balance_loss_clip": 1.0539434,
      "balance_loss_mlp": 1.02008128,
      "epoch": 0.38249263512294834,
      "flos": 19789149864960.0,
      "grad_norm": 2.418928733011627,
      "language_loss": 0.80241597,
      "learning_rate": 2.8323113215381872e-06,
      "loss": 0.82468194,
      "num_input_tokens_seen": 68563070,
      "step": 3181,
      "time_per_iteration": 2.770672559738159
    },
    {
      "auxiliary_loss_clip": 0.01196462,
      "auxiliary_loss_mlp": 0.01034149,
      "balance_loss_clip": 1.05130744,
      "balance_loss_mlp": 1.0245471,
      "epoch": 0.38261287801358745,
      "flos": 21433930565760.0,
      "grad_norm": 1.9884238419269995,
      "language_loss": 0.75974327,
      "learning_rate": 2.831602944657042e-06,
      "loss": 0.78204936,
      "num_input_tokens_seen": 68581150,
      "step": 3182,
      "time_per_iteration": 2.728296995162964
    },
    {
      "auxiliary_loss_clip": 0.0120451,
      "auxiliary_loss_mlp": 0.01028028,
      "balance_loss_clip": 1.05315995,
      "balance_loss_mlp": 1.01890218,
      "epoch": 0.38273312090422656,
      "flos": 21981568296960.0,
      "grad_norm": 3.012005005064789,
      "language_loss": 0.74110436,
      "learning_rate": 2.830894441622391e-06,
      "loss": 0.76342976,
      "num_input_tokens_seen": 68597800,
      "step": 3183,
      "time_per_iteration": 2.6545841693878174
    },
    {
      "auxiliary_loss_clip": 0.01198348,
      "auxiliary_loss_mlp": 0.01209137,
      "balance_loss_clip": 1.04922986,
      "balance_loss_mlp": 1.0006845,
      "epoch": 0.3828533637948656,
      "flos": 24790895838720.0,
      "grad_norm": 1.8717879017421342,
      "language_loss": 0.80272222,
      "learning_rate": 2.8301858125417134e-06,
      "loss": 0.82679713,
      "num_input_tokens_seen": 68617640,
      "step": 3184,
      "time_per_iteration": 2.735278367996216
    },
    {
      "auxiliary_loss_clip": 0.01198627,
      "auxiliary_loss_mlp": 0.01026509,
      "balance_loss_clip": 1.05637765,
      "balance_loss_mlp": 1.01813471,
      "epoch": 0.38297360668550473,
      "flos": 22455445449600.0,
      "grad_norm": 2.077155853202071,
      "language_loss": 0.74163264,
      "learning_rate": 2.8294770575225082e-06,
      "loss": 0.76388395,
      "num_input_tokens_seen": 68637770,
      "step": 3185,
      "time_per_iteration": 2.656095266342163
    },
    {
      "auxiliary_loss_clip": 0.01199526,
      "auxiliary_loss_mlp": 0.01032432,
      "balance_loss_clip": 1.05912995,
      "balance_loss_mlp": 1.02308011,
      "epoch": 0.3830938495761438,
      "flos": 24896903852160.0,
      "grad_norm": 1.8684950405825504,
      "language_loss": 0.83867991,
      "learning_rate": 2.828768176672293e-06,
      "loss": 0.86099946,
      "num_input_tokens_seen": 68656885,
      "step": 3186,
      "time_per_iteration": 2.7130980491638184
    },
    {
      "auxiliary_loss_clip": 0.01197313,
      "auxiliary_loss_mlp": 0.01029698,
      "balance_loss_clip": 1.05157876,
      "balance_loss_mlp": 1.02041197,
      "epoch": 0.3832140924667829,
      "flos": 33036236784000.0,
      "grad_norm": 2.3930764189527953,
      "language_loss": 0.71464175,
      "learning_rate": 2.8280591700986044e-06,
      "loss": 0.73691189,
      "num_input_tokens_seen": 68678750,
      "step": 3187,
      "time_per_iteration": 2.7491190433502197
    },
    {
      "auxiliary_loss_clip": 0.01200628,
      "auxiliary_loss_mlp": 0.0103589,
      "balance_loss_clip": 1.05265331,
      "balance_loss_mlp": 1.02669859,
      "epoch": 0.383334335357422,
      "flos": 31903721896320.0,
      "grad_norm": 6.211262025158446,
      "language_loss": 0.74895787,
      "learning_rate": 2.827350037908999e-06,
      "loss": 0.77132303,
      "num_input_tokens_seen": 68698190,
      "step": 3188,
      "time_per_iteration": 2.7613723278045654
    },
    {
      "auxiliary_loss_clip": 0.01206729,
      "auxiliary_loss_mlp": 0.01034982,
      "balance_loss_clip": 1.05277479,
      "balance_loss_mlp": 1.02515876,
      "epoch": 0.38345457824806106,
      "flos": 19791915212160.0,
      "grad_norm": 7.541975250053557,
      "language_loss": 0.79648232,
      "learning_rate": 2.8266407802110496e-06,
      "loss": 0.81889939,
      "num_input_tokens_seen": 68716445,
      "step": 3189,
      "time_per_iteration": 2.687598705291748
    },
    {
      "auxiliary_loss_clip": 0.01210803,
      "auxiliary_loss_mlp": 0.01031769,
      "balance_loss_clip": 1.04592717,
      "balance_loss_mlp": 1.02163649,
      "epoch": 0.3835748211387002,
      "flos": 22419391173120.0,
      "grad_norm": 1.9316620062790442,
      "language_loss": 0.75888038,
      "learning_rate": 2.8259313971123515e-06,
      "loss": 0.78130615,
      "num_input_tokens_seen": 68737565,
      "step": 3190,
      "time_per_iteration": 4.640646934509277
    },
    {
      "auxiliary_loss_clip": 0.01194523,
      "auxiliary_loss_mlp": 0.01032255,
      "balance_loss_clip": 1.05794394,
      "balance_loss_mlp": 1.02363014,
      "epoch": 0.3836950640293393,
      "flos": 25118436983040.0,
      "grad_norm": 1.5445067415866962,
      "language_loss": 0.78501618,
      "learning_rate": 2.8252218887205166e-06,
      "loss": 0.807284,
      "num_input_tokens_seen": 68758255,
      "step": 3191,
      "time_per_iteration": 2.7591452598571777
    },
    {
      "auxiliary_loss_clip": 0.01205627,
      "auxiliary_loss_mlp": 0.01030572,
      "balance_loss_clip": 1.0502876,
      "balance_loss_mlp": 1.02179849,
      "epoch": 0.38381530691997834,
      "flos": 21799213925760.0,
      "grad_norm": 1.759278183866423,
      "language_loss": 0.80667019,
      "learning_rate": 2.824512255143178e-06,
      "loss": 0.82903218,
      "num_input_tokens_seen": 68777490,
      "step": 3192,
      "time_per_iteration": 2.8123927116394043
    },
    {
      "auxiliary_loss_clip": 0.01202718,
      "auxiliary_loss_mlp": 0.01028892,
      "balance_loss_clip": 1.05177665,
      "balance_loss_mlp": 1.02055371,
      "epoch": 0.38393554981061745,
      "flos": 21252689516160.0,
      "grad_norm": 2.5406745218244176,
      "language_loss": 0.79245543,
      "learning_rate": 2.8238024964879855e-06,
      "loss": 0.81477153,
      "num_input_tokens_seen": 68798385,
      "step": 3193,
      "time_per_iteration": 3.6534547805786133
    },
    {
      "auxiliary_loss_clip": 0.01196826,
      "auxiliary_loss_mlp": 0.01031739,
      "balance_loss_clip": 1.05892968,
      "balance_loss_mlp": 1.02214873,
      "epoch": 0.38405579270125656,
      "flos": 17019360218880.0,
      "grad_norm": 2.0495491453425614,
      "language_loss": 0.77235591,
      "learning_rate": 2.8230926128626095e-06,
      "loss": 0.79464161,
      "num_input_tokens_seen": 68816880,
      "step": 3194,
      "time_per_iteration": 2.7096128463745117
    },
    {
      "auxiliary_loss_clip": 0.01192626,
      "auxiliary_loss_mlp": 0.01032637,
      "balance_loss_clip": 1.0520153,
      "balance_loss_mlp": 1.02274847,
      "epoch": 0.3841760355918956,
      "flos": 21835375943040.0,
      "grad_norm": 1.9066636949710272,
      "language_loss": 0.79237306,
      "learning_rate": 2.822382604374738e-06,
      "loss": 0.81462574,
      "num_input_tokens_seen": 68835805,
      "step": 3195,
      "time_per_iteration": 2.6693687438964844
    },
    {
      "auxiliary_loss_clip": 0.01199843,
      "auxiliary_loss_mlp": 0.01034227,
      "balance_loss_clip": 1.05823588,
      "balance_loss_mlp": 1.02470839,
      "epoch": 0.3842962784825347,
      "flos": 25915114684800.0,
      "grad_norm": 2.8897833179368,
      "language_loss": 0.65512514,
      "learning_rate": 2.8216724711320793e-06,
      "loss": 0.67746586,
      "num_input_tokens_seen": 68854930,
      "step": 3196,
      "time_per_iteration": 2.7849605083465576
    },
    {
      "auxiliary_loss_clip": 0.01192784,
      "auxiliary_loss_mlp": 0.01208642,
      "balance_loss_clip": 1.05682349,
      "balance_loss_mlp": 1.00064969,
      "epoch": 0.38441652137317384,
      "flos": 25337492075520.0,
      "grad_norm": 2.065346608377059,
      "language_loss": 0.79782641,
      "learning_rate": 2.820962213242361e-06,
      "loss": 0.82184064,
      "num_input_tokens_seen": 68874260,
      "step": 3197,
      "time_per_iteration": 2.6340742111206055
    },
    {
      "auxiliary_loss_clip": 0.01190567,
      "auxiliary_loss_mlp": 0.01038934,
      "balance_loss_clip": 1.05607986,
      "balance_loss_mlp": 1.02966559,
      "epoch": 0.3845367642638129,
      "flos": 18113486446080.0,
      "grad_norm": 2.276492516523179,
      "language_loss": 0.8454743,
      "learning_rate": 2.8202518308133264e-06,
      "loss": 0.86776936,
      "num_input_tokens_seen": 68891535,
      "step": 3198,
      "time_per_iteration": 3.570380926132202
    },
    {
      "auxiliary_loss_clip": 0.01195195,
      "auxiliary_loss_mlp": 0.01031755,
      "balance_loss_clip": 1.05692124,
      "balance_loss_mlp": 1.02193248,
      "epoch": 0.384657007154452,
      "flos": 25228395492480.0,
      "grad_norm": 2.1543834687223935,
      "language_loss": 0.73428988,
      "learning_rate": 2.8195413239527426e-06,
      "loss": 0.75655937,
      "num_input_tokens_seen": 68911275,
      "step": 3199,
      "time_per_iteration": 2.6896960735321045
    },
    {
      "auxiliary_loss_clip": 0.0119336,
      "auxiliary_loss_mlp": 0.01031704,
      "balance_loss_clip": 1.05517137,
      "balance_loss_mlp": 1.02270961,
      "epoch": 0.38477725004509106,
      "flos": 19865855358720.0,
      "grad_norm": 2.176707898739848,
      "language_loss": 0.8091855,
      "learning_rate": 2.8188306927683906e-06,
      "loss": 0.83143616,
      "num_input_tokens_seen": 68930745,
      "step": 3200,
      "time_per_iteration": 2.715359687805176
    },
    {
      "auxiliary_loss_clip": 0.01200395,
      "auxiliary_loss_mlp": 0.01028302,
      "balance_loss_clip": 1.05571985,
      "balance_loss_mlp": 1.01977301,
      "epoch": 0.38489749293573017,
      "flos": 18259391491200.0,
      "grad_norm": 2.2413366777382007,
      "language_loss": 0.74326962,
      "learning_rate": 2.818119937368074e-06,
      "loss": 0.76555669,
      "num_input_tokens_seen": 68949380,
      "step": 3201,
      "time_per_iteration": 2.802722692489624
    },
    {
      "auxiliary_loss_clip": 0.01202515,
      "auxiliary_loss_mlp": 0.01026392,
      "balance_loss_clip": 1.05599689,
      "balance_loss_mlp": 1.01692045,
      "epoch": 0.3850177358263693,
      "flos": 24389163152640.0,
      "grad_norm": 1.9788591136769382,
      "language_loss": 0.6576364,
      "learning_rate": 2.817409057859613e-06,
      "loss": 0.6799255,
      "num_input_tokens_seen": 68968370,
      "step": 3202,
      "time_per_iteration": 2.713102340698242
    },
    {
      "auxiliary_loss_clip": 0.01195951,
      "auxiliary_loss_mlp": 0.01030922,
      "balance_loss_clip": 1.04974782,
      "balance_loss_mlp": 1.02097404,
      "epoch": 0.38513797871700833,
      "flos": 17671533505920.0,
      "grad_norm": 2.3368027227363943,
      "language_loss": 0.79255342,
      "learning_rate": 2.8166980543508482e-06,
      "loss": 0.81482214,
      "num_input_tokens_seen": 68984260,
      "step": 3203,
      "time_per_iteration": 2.7343997955322266
    },
    {
      "auxiliary_loss_clip": 0.01196368,
      "auxiliary_loss_mlp": 0.01026532,
      "balance_loss_clip": 1.05907297,
      "balance_loss_mlp": 1.01729345,
      "epoch": 0.38525822160764744,
      "flos": 25739583897600.0,
      "grad_norm": 1.9538325022973608,
      "language_loss": 0.80124974,
      "learning_rate": 2.815986926949638e-06,
      "loss": 0.8234787,
      "num_input_tokens_seen": 69002760,
      "step": 3204,
      "time_per_iteration": 2.645268201828003
    },
    {
      "auxiliary_loss_clip": 0.01197072,
      "auxiliary_loss_mlp": 0.01026954,
      "balance_loss_clip": 1.0575918,
      "balance_loss_mlp": 1.0182333,
      "epoch": 0.38537846449828655,
      "flos": 20193647898240.0,
      "grad_norm": 1.8764457849963108,
      "language_loss": 0.80281639,
      "learning_rate": 2.8152756757638597e-06,
      "loss": 0.82505667,
      "num_input_tokens_seen": 69021260,
      "step": 3205,
      "time_per_iteration": 2.6809186935424805
    },
    {
      "auxiliary_loss_clip": 0.01195032,
      "auxiliary_loss_mlp": 0.01026923,
      "balance_loss_clip": 1.05724669,
      "balance_loss_mlp": 1.01838195,
      "epoch": 0.3854987073889256,
      "flos": 23039352938880.0,
      "grad_norm": 2.044721221877591,
      "language_loss": 0.846573,
      "learning_rate": 2.8145643009014093e-06,
      "loss": 0.86879253,
      "num_input_tokens_seen": 69039755,
      "step": 3206,
      "time_per_iteration": 2.618070125579834
    },
    {
      "auxiliary_loss_clip": 0.01197175,
      "auxiliary_loss_mlp": 0.01029788,
      "balance_loss_clip": 1.0569551,
      "balance_loss_mlp": 1.02130067,
      "epoch": 0.3856189502795647,
      "flos": 20190631155840.0,
      "grad_norm": 2.0046945448282703,
      "language_loss": 0.79209697,
      "learning_rate": 2.813852802470202e-06,
      "loss": 0.81436658,
      "num_input_tokens_seen": 69057650,
      "step": 3207,
      "time_per_iteration": 2.815725326538086
    },
    {
      "auxiliary_loss_clip": 0.01193785,
      "auxiliary_loss_mlp": 0.01029342,
      "balance_loss_clip": 1.05451,
      "balance_loss_mlp": 1.01961422,
      "epoch": 0.38573919317020383,
      "flos": 25702631781120.0,
      "grad_norm": 1.8253600346605234,
      "language_loss": 0.7240656,
      "learning_rate": 2.8131411805781717e-06,
      "loss": 0.74629688,
      "num_input_tokens_seen": 69077775,
      "step": 3208,
      "time_per_iteration": 2.6955039501190186
    },
    {
      "auxiliary_loss_clip": 0.01201644,
      "auxiliary_loss_mlp": 0.01031915,
      "balance_loss_clip": 1.05814528,
      "balance_loss_mlp": 1.0223186,
      "epoch": 0.3858594360608429,
      "flos": 29821405628160.0,
      "grad_norm": 2.1574218811087937,
      "language_loss": 0.63974649,
      "learning_rate": 2.8124294353332707e-06,
      "loss": 0.66208208,
      "num_input_tokens_seen": 69096450,
      "step": 3209,
      "time_per_iteration": 2.717623233795166
    },
    {
      "auxiliary_loss_clip": 0.01206909,
      "auxiliary_loss_mlp": 0.01033057,
      "balance_loss_clip": 1.05386233,
      "balance_loss_mlp": 1.02424717,
      "epoch": 0.385979678951482,
      "flos": 24790428961920.0,
      "grad_norm": 3.8809864276569717,
      "language_loss": 0.77736014,
      "learning_rate": 2.8117175668434713e-06,
      "loss": 0.79975975,
      "num_input_tokens_seen": 69116110,
      "step": 3210,
      "time_per_iteration": 2.8459348678588867
    },
    {
      "auxiliary_loss_clip": 0.01194783,
      "auxiliary_loss_mlp": 0.01025863,
      "balance_loss_clip": 1.05776703,
      "balance_loss_mlp": 1.01695204,
      "epoch": 0.3860999218421211,
      "flos": 21287881866240.0,
      "grad_norm": 2.858241742313565,
      "language_loss": 0.70106304,
      "learning_rate": 2.811005575216762e-06,
      "loss": 0.72326946,
      "num_input_tokens_seen": 69134825,
      "step": 3211,
      "time_per_iteration": 2.636626720428467
    },
    {
      "auxiliary_loss_clip": 0.01193455,
      "auxiliary_loss_mlp": 0.01031719,
      "balance_loss_clip": 1.05165136,
      "balance_loss_mlp": 1.02298725,
      "epoch": 0.38622016473276016,
      "flos": 24536720223360.0,
      "grad_norm": 1.4528425398663871,
      "language_loss": 0.79009104,
      "learning_rate": 2.8102934605611513e-06,
      "loss": 0.81234276,
      "num_input_tokens_seen": 69156460,
      "step": 3212,
      "time_per_iteration": 2.769033908843994
    },
    {
      "auxiliary_loss_clip": 0.01206246,
      "auxiliary_loss_mlp": 0.01029537,
      "balance_loss_clip": 1.05808425,
      "balance_loss_mlp": 1.02083433,
      "epoch": 0.3863404076233993,
      "flos": 20558212986240.0,
      "grad_norm": 2.6557562821009086,
      "language_loss": 0.67400026,
      "learning_rate": 2.8095812229846665e-06,
      "loss": 0.69635808,
      "num_input_tokens_seen": 69176420,
      "step": 3213,
      "time_per_iteration": 2.681030750274658
    },
    {
      "auxiliary_loss_clip": 0.01200251,
      "auxiliary_loss_mlp": 0.01031681,
      "balance_loss_clip": 1.05414152,
      "balance_loss_mlp": 1.02255583,
      "epoch": 0.3864606505140384,
      "flos": 22346277039360.0,
      "grad_norm": 3.393269546966571,
      "language_loss": 0.68689716,
      "learning_rate": 2.808868862595355e-06,
      "loss": 0.70921648,
      "num_input_tokens_seen": 69196665,
      "step": 3214,
      "time_per_iteration": 2.7126388549804688
    },
    {
      "auxiliary_loss_clip": 0.01199412,
      "auxiliary_loss_mlp": 0.01033254,
      "balance_loss_clip": 1.05628324,
      "balance_loss_mlp": 1.02427125,
      "epoch": 0.38658089340467744,
      "flos": 25703601448320.0,
      "grad_norm": 2.3209299864763113,
      "language_loss": 0.79355276,
      "learning_rate": 2.8081563795012795e-06,
      "loss": 0.81587934,
      "num_input_tokens_seen": 69216290,
      "step": 3215,
      "time_per_iteration": 2.6552908420562744
    },
    {
      "auxiliary_loss_clip": 0.01205644,
      "auxiliary_loss_mlp": 0.01030045,
      "balance_loss_clip": 1.05407822,
      "balance_loss_mlp": 1.02107418,
      "epoch": 0.38670113629531655,
      "flos": 33802534558080.0,
      "grad_norm": 1.8128445646046956,
      "language_loss": 0.73768568,
      "learning_rate": 2.807443773810524e-06,
      "loss": 0.76004261,
      "num_input_tokens_seen": 69237550,
      "step": 3216,
      "time_per_iteration": 3.661541223526001
    },
    {
      "auxiliary_loss_clip": 0.01201226,
      "auxiliary_loss_mlp": 0.01032198,
      "balance_loss_clip": 1.05527496,
      "balance_loss_mlp": 1.02374649,
      "epoch": 0.3868213791859556,
      "flos": 23331522165120.0,
      "grad_norm": 1.9818660704691167,
      "language_loss": 0.89795792,
      "learning_rate": 2.80673104563119e-06,
      "loss": 0.9202922,
      "num_input_tokens_seen": 69258175,
      "step": 3217,
      "time_per_iteration": 3.6636078357696533
    },
    {
      "auxiliary_loss_clip": 0.011931,
      "auxiliary_loss_mlp": 0.01024067,
      "balance_loss_clip": 1.05698156,
      "balance_loss_mlp": 1.01569891,
      "epoch": 0.3869416220765947,
      "flos": 18441530380800.0,
      "grad_norm": 1.8107971501381797,
      "language_loss": 0.78888035,
      "learning_rate": 2.8060181950713976e-06,
      "loss": 0.81105196,
      "num_input_tokens_seen": 69274965,
      "step": 3218,
      "time_per_iteration": 2.6405649185180664
    },
    {
      "auxiliary_loss_clip": 0.01199553,
      "auxiliary_loss_mlp": 0.01030061,
      "balance_loss_clip": 1.05306542,
      "balance_loss_mlp": 1.02049422,
      "epoch": 0.3870618649672338,
      "flos": 15632992938240.0,
      "grad_norm": 2.184405073924568,
      "language_loss": 0.80560327,
      "learning_rate": 2.805305222239286e-06,
      "loss": 0.82789934,
      "num_input_tokens_seen": 69292220,
      "step": 3219,
      "time_per_iteration": 2.6972386837005615
    },
    {
      "auxiliary_loss_clip": 0.0119453,
      "auxiliary_loss_mlp": 0.0102867,
      "balance_loss_clip": 1.05468762,
      "balance_loss_mlp": 1.01983094,
      "epoch": 0.3871821078578729,
      "flos": 23513804709120.0,
      "grad_norm": 2.061549467116816,
      "language_loss": 0.73843545,
      "learning_rate": 2.8045921272430118e-06,
      "loss": 0.76066744,
      "num_input_tokens_seen": 69311900,
      "step": 3220,
      "time_per_iteration": 3.7707436084747314
    },
    {
      "auxiliary_loss_clip": 0.01204266,
      "auxiliary_loss_mlp": 0.01030979,
      "balance_loss_clip": 1.05662143,
      "balance_loss_mlp": 1.02135873,
      "epoch": 0.387302350748512,
      "flos": 17778259791360.0,
      "grad_norm": 2.193782446279296,
      "language_loss": 0.76538283,
      "learning_rate": 2.803878910190753e-06,
      "loss": 0.78773528,
      "num_input_tokens_seen": 69328820,
      "step": 3221,
      "time_per_iteration": 2.6191275119781494
    },
    {
      "auxiliary_loss_clip": 0.0119901,
      "auxiliary_loss_mlp": 0.01028896,
      "balance_loss_clip": 1.05504334,
      "balance_loss_mlp": 1.01949,
      "epoch": 0.3874225936391511,
      "flos": 11503409097600.0,
      "grad_norm": 2.4785646712517693,
      "language_loss": 0.81840146,
      "learning_rate": 2.8031655711907017e-06,
      "loss": 0.84068054,
      "num_input_tokens_seen": 69342525,
      "step": 3222,
      "time_per_iteration": 2.5678844451904297
    },
    {
      "auxiliary_loss_clip": 0.01199677,
      "auxiliary_loss_mlp": 0.01030634,
      "balance_loss_clip": 1.05743504,
      "balance_loss_mlp": 1.02175319,
      "epoch": 0.38754283652979016,
      "flos": 21945154884480.0,
      "grad_norm": 2.159407426645904,
      "language_loss": 0.80962569,
      "learning_rate": 2.8024521103510723e-06,
      "loss": 0.83192879,
      "num_input_tokens_seen": 69359295,
      "step": 3223,
      "time_per_iteration": 2.63606858253479
    },
    {
      "auxiliary_loss_clip": 0.01194747,
      "auxiliary_loss_mlp": 0.01029041,
      "balance_loss_clip": 1.05377614,
      "balance_loss_mlp": 1.02070189,
      "epoch": 0.38766307942042927,
      "flos": 21175984022400.0,
      "grad_norm": 2.2315046714158813,
      "language_loss": 0.75769067,
      "learning_rate": 2.8017385277800952e-06,
      "loss": 0.77992857,
      "num_input_tokens_seen": 69377650,
      "step": 3224,
      "time_per_iteration": 3.5409529209136963
    },
    {
      "auxiliary_loss_clip": 0.01205068,
      "auxiliary_loss_mlp": 0.01033754,
      "balance_loss_clip": 1.05401051,
      "balance_loss_mlp": 1.02445006,
      "epoch": 0.3877833223110684,
      "flos": 27417294391680.0,
      "grad_norm": 2.079076210392622,
      "language_loss": 0.74840641,
      "learning_rate": 2.8010248235860213e-06,
      "loss": 0.77079463,
      "num_input_tokens_seen": 69397765,
      "step": 3225,
      "time_per_iteration": 2.6979410648345947
    },
    {
      "auxiliary_loss_clip": 0.01115817,
      "auxiliary_loss_mlp": 0.01199613,
      "balance_loss_clip": 1.03203619,
      "balance_loss_mlp": 1.00008929,
      "epoch": 0.38790356520170743,
      "flos": 64500019879680.0,
      "grad_norm": 0.8298292136352589,
      "language_loss": 0.62831438,
      "learning_rate": 2.8003109978771192e-06,
      "loss": 0.65146863,
      "num_input_tokens_seen": 69458930,
      "step": 3226,
      "time_per_iteration": 3.25058913230896
    },
    {
      "auxiliary_loss_clip": 0.01191532,
      "auxiliary_loss_mlp": 0.0103259,
      "balance_loss_clip": 1.04723239,
      "balance_loss_mlp": 1.02337539,
      "epoch": 0.38802380809234654,
      "flos": 22345415112960.0,
      "grad_norm": 2.0454510032078033,
      "language_loss": 0.79290819,
      "learning_rate": 2.799597050761674e-06,
      "loss": 0.81514943,
      "num_input_tokens_seen": 69475135,
      "step": 3227,
      "time_per_iteration": 2.733947277069092
    },
    {
      "auxiliary_loss_clip": 0.01195193,
      "auxiliary_loss_mlp": 0.01028778,
      "balance_loss_clip": 1.05725563,
      "balance_loss_mlp": 1.01966476,
      "epoch": 0.38814405098298566,
      "flos": 25261361199360.0,
      "grad_norm": 2.1172673450248003,
      "language_loss": 0.79246575,
      "learning_rate": 2.7988829823479924e-06,
      "loss": 0.81470543,
      "num_input_tokens_seen": 69493525,
      "step": 3228,
      "time_per_iteration": 2.640578031539917
    },
    {
      "auxiliary_loss_clip": 0.01193633,
      "auxiliary_loss_mlp": 0.0103805,
      "balance_loss_clip": 1.05244815,
      "balance_loss_mlp": 1.02791142,
      "epoch": 0.3882642938736247,
      "flos": 18841180078080.0,
      "grad_norm": 2.0151607239024854,
      "language_loss": 0.64108872,
      "learning_rate": 2.7981687927443976e-06,
      "loss": 0.66340554,
      "num_input_tokens_seen": 69510325,
      "step": 3229,
      "time_per_iteration": 2.6328372955322266
    },
    {
      "auxiliary_loss_clip": 0.01193893,
      "auxiliary_loss_mlp": 0.01026894,
      "balance_loss_clip": 1.05261123,
      "balance_loss_mlp": 1.01857877,
      "epoch": 0.3883845367642638,
      "flos": 21652806090240.0,
      "grad_norm": 2.2563053064207446,
      "language_loss": 0.85703778,
      "learning_rate": 2.797454482059231e-06,
      "loss": 0.87924564,
      "num_input_tokens_seen": 69530480,
      "step": 3230,
      "time_per_iteration": 2.6536989212036133
    },
    {
      "auxiliary_loss_clip": 0.01197411,
      "auxiliary_loss_mlp": 0.01024284,
      "balance_loss_clip": 1.05810201,
      "balance_loss_mlp": 1.0157721,
      "epoch": 0.3885047796549029,
      "flos": 20557530627840.0,
      "grad_norm": 1.6585719334358862,
      "language_loss": 0.84542197,
      "learning_rate": 2.7967400504008537e-06,
      "loss": 0.86763895,
      "num_input_tokens_seen": 69549780,
      "step": 3231,
      "time_per_iteration": 2.6392714977264404
    },
    {
      "auxiliary_loss_clip": 0.01121435,
      "auxiliary_loss_mlp": 0.01001418,
      "balance_loss_clip": 1.03211069,
      "balance_loss_mlp": 1.00032139,
      "epoch": 0.388625022545542,
      "flos": 64325491695360.0,
      "grad_norm": 0.8181431628252249,
      "language_loss": 0.5742321,
      "learning_rate": 2.7960254978776456e-06,
      "loss": 0.59546059,
      "num_input_tokens_seen": 69611870,
      "step": 3232,
      "time_per_iteration": 3.229909896850586
    },
    {
      "auxiliary_loss_clip": 0.01198651,
      "auxiliary_loss_mlp": 0.0103704,
      "balance_loss_clip": 1.06036854,
      "balance_loss_mlp": 1.02732491,
      "epoch": 0.3887452654361811,
      "flos": 18113881495680.0,
      "grad_norm": 2.634600508926509,
      "language_loss": 0.81469142,
      "learning_rate": 2.7953108245980006e-06,
      "loss": 0.83704829,
      "num_input_tokens_seen": 69630385,
      "step": 3233,
      "time_per_iteration": 2.6584043502807617
    },
    {
      "auxiliary_loss_clip": 0.0119195,
      "auxiliary_loss_mlp": 0.01026697,
      "balance_loss_clip": 1.05488503,
      "balance_loss_mlp": 1.01842427,
      "epoch": 0.38886550832682015,
      "flos": 24975261371520.0,
      "grad_norm": 1.6612294018275433,
      "language_loss": 0.73502636,
      "learning_rate": 2.7945960306703365e-06,
      "loss": 0.75721282,
      "num_input_tokens_seen": 69653370,
      "step": 3234,
      "time_per_iteration": 2.743602752685547
    },
    {
      "auxiliary_loss_clip": 0.01200544,
      "auxiliary_loss_mlp": 0.01028316,
      "balance_loss_clip": 1.05593741,
      "balance_loss_mlp": 1.01887488,
      "epoch": 0.38898575121745926,
      "flos": 27199496275200.0,
      "grad_norm": 2.494840246652254,
      "language_loss": 0.65856993,
      "learning_rate": 2.7938811162030865e-06,
      "loss": 0.68085855,
      "num_input_tokens_seen": 69673635,
      "step": 3235,
      "time_per_iteration": 2.7019295692443848
    },
    {
      "auxiliary_loss_clip": 0.01194381,
      "auxiliary_loss_mlp": 0.01034815,
      "balance_loss_clip": 1.05630636,
      "balance_loss_mlp": 1.02623224,
      "epoch": 0.3891059941080984,
      "flos": 28763728727040.0,
      "grad_norm": 1.9527329536887867,
      "language_loss": 0.82496858,
      "learning_rate": 2.793166081304702e-06,
      "loss": 0.84726059,
      "num_input_tokens_seen": 69694130,
      "step": 3236,
      "time_per_iteration": 2.708176374435425
    },
    {
      "auxiliary_loss_clip": 0.01207691,
      "auxiliary_loss_mlp": 0.01032754,
      "balance_loss_clip": 1.05289888,
      "balance_loss_mlp": 1.02294302,
      "epoch": 0.38922623699873743,
      "flos": 22893447893760.0,
      "grad_norm": 2.037364064807022,
      "language_loss": 0.82288539,
      "learning_rate": 2.7924509260836543e-06,
      "loss": 0.84528989,
      "num_input_tokens_seen": 69713255,
      "step": 3237,
      "time_per_iteration": 2.68410587310791
    },
    {
      "auxiliary_loss_clip": 0.01199756,
      "auxiliary_loss_mlp": 0.01025409,
      "balance_loss_clip": 1.05171037,
      "balance_loss_mlp": 1.01648045,
      "epoch": 0.38934647988937654,
      "flos": 19792418002560.0,
      "grad_norm": 2.064376962978095,
      "language_loss": 0.68365371,
      "learning_rate": 2.791735650648431e-06,
      "loss": 0.70590532,
      "num_input_tokens_seen": 69732375,
      "step": 3238,
      "time_per_iteration": 2.6284470558166504
    },
    {
      "auxiliary_loss_clip": 0.01195707,
      "auxiliary_loss_mlp": 0.01028726,
      "balance_loss_clip": 1.05405962,
      "balance_loss_mlp": 1.01999438,
      "epoch": 0.38946672278001565,
      "flos": 19202081978880.0,
      "grad_norm": 3.3437638922217277,
      "language_loss": 0.74557883,
      "learning_rate": 2.791020255107538e-06,
      "loss": 0.76782316,
      "num_input_tokens_seen": 69749745,
      "step": 3239,
      "time_per_iteration": 2.5869300365448
    },
    {
      "auxiliary_loss_clip": 0.01196418,
      "auxiliary_loss_mlp": 0.0102797,
      "balance_loss_clip": 1.05096292,
      "balance_loss_mlp": 1.0192318,
      "epoch": 0.3895869656706547,
      "flos": 24936477661440.0,
      "grad_norm": 1.682956897230226,
      "language_loss": 0.80902243,
      "learning_rate": 2.7903047395695023e-06,
      "loss": 0.8312664,
      "num_input_tokens_seen": 69769645,
      "step": 3240,
      "time_per_iteration": 2.78043270111084
    },
    {
      "auxiliary_loss_clip": 0.01197111,
      "auxiliary_loss_mlp": 0.01209371,
      "balance_loss_clip": 1.05789876,
      "balance_loss_mlp": 1.00033832,
      "epoch": 0.3897072085612938,
      "flos": 24133622820480.0,
      "grad_norm": 2.1880941066928856,
      "language_loss": 0.90223086,
      "learning_rate": 2.789589104142865e-06,
      "loss": 0.9262957,
      "num_input_tokens_seen": 69787270,
      "step": 3241,
      "time_per_iteration": 2.6385107040405273
    },
    {
      "auxiliary_loss_clip": 0.0120225,
      "auxiliary_loss_mlp": 0.01027055,
      "balance_loss_clip": 1.05344582,
      "balance_loss_mlp": 1.01839459,
      "epoch": 0.3898274514519329,
      "flos": 17166342672000.0,
      "grad_norm": 2.37577247234666,
      "language_loss": 0.76453507,
      "learning_rate": 2.7888733489361895e-06,
      "loss": 0.7868281,
      "num_input_tokens_seen": 69805685,
      "step": 3242,
      "time_per_iteration": 2.6557016372680664
    },
    {
      "auxiliary_loss_clip": 0.01100516,
      "auxiliary_loss_mlp": 0.01004218,
      "balance_loss_clip": 1.03208148,
      "balance_loss_mlp": 1.00309181,
      "epoch": 0.389947694342572,
      "flos": 66074807952000.0,
      "grad_norm": 0.7238608322723841,
      "language_loss": 0.58697915,
      "learning_rate": 2.788157474058054e-06,
      "loss": 0.6080265,
      "num_input_tokens_seen": 69867960,
      "step": 3243,
      "time_per_iteration": 4.14143967628479
    },
    {
      "auxiliary_loss_clip": 0.01191787,
      "auxiliary_loss_mlp": 0.01027326,
      "balance_loss_clip": 1.05729425,
      "balance_loss_mlp": 1.0187788,
      "epoch": 0.3900679372332111,
      "flos": 25740912700800.0,
      "grad_norm": 1.5626169720597634,
      "language_loss": 0.69749171,
      "learning_rate": 2.7874414796170555e-06,
      "loss": 0.71968287,
      "num_input_tokens_seen": 69889450,
      "step": 3244,
      "time_per_iteration": 3.530400276184082
    },
    {
      "auxiliary_loss_clip": 0.01190262,
      "auxiliary_loss_mlp": 0.01029617,
      "balance_loss_clip": 1.0518105,
      "balance_loss_mlp": 1.0197823,
      "epoch": 0.3901881801238502,
      "flos": 11801611808640.0,
      "grad_norm": 2.3452485050400234,
      "language_loss": 0.84069979,
      "learning_rate": 2.7867253657218113e-06,
      "loss": 0.86289859,
      "num_input_tokens_seen": 69903340,
      "step": 3245,
      "time_per_iteration": 2.5454492568969727
    },
    {
      "auxiliary_loss_clip": 0.01195713,
      "auxiliary_loss_mlp": 0.01208952,
      "balance_loss_clip": 1.05220509,
      "balance_loss_mlp": 1.00028419,
      "epoch": 0.39030842301448926,
      "flos": 27308951994240.0,
      "grad_norm": 1.9231370345993268,
      "language_loss": 0.73178577,
      "learning_rate": 2.7860091324809544e-06,
      "loss": 0.75583237,
      "num_input_tokens_seen": 69924400,
      "step": 3246,
      "time_per_iteration": 2.6367321014404297
    },
    {
      "auxiliary_loss_clip": 0.01194533,
      "auxiliary_loss_mlp": 0.01028475,
      "balance_loss_clip": 1.05871117,
      "balance_loss_mlp": 1.01960003,
      "epoch": 0.39042866590512837,
      "flos": 27163334257920.0,
      "grad_norm": 1.8481936093620832,
      "language_loss": 0.81356919,
      "learning_rate": 2.7852927800031377e-06,
      "loss": 0.83579922,
      "num_input_tokens_seen": 69944565,
      "step": 3247,
      "time_per_iteration": 3.6547963619232178
    },
    {
      "auxiliary_loss_clip": 0.01200767,
      "auxiliary_loss_mlp": 0.010254,
      "balance_loss_clip": 1.05334127,
      "balance_loss_mlp": 1.01730013,
      "epoch": 0.3905489087957674,
      "flos": 29716115886720.0,
      "grad_norm": 1.9117430892141458,
      "language_loss": 0.83170295,
      "learning_rate": 2.7845763083970298e-06,
      "loss": 0.85396457,
      "num_input_tokens_seen": 69964965,
      "step": 3248,
      "time_per_iteration": 2.7013041973114014
    },
    {
      "auxiliary_loss_clip": 0.01188027,
      "auxiliary_loss_mlp": 0.010278,
      "balance_loss_clip": 1.05464053,
      "balance_loss_mlp": 1.01840007,
      "epoch": 0.39066915168640653,
      "flos": 24498618871680.0,
      "grad_norm": 2.2502130070722424,
      "language_loss": 0.82034016,
      "learning_rate": 2.7838597177713205e-06,
      "loss": 0.84249842,
      "num_input_tokens_seen": 69986055,
      "step": 3249,
      "time_per_iteration": 2.6694657802581787
    },
    {
      "auxiliary_loss_clip": 0.01190722,
      "auxiliary_loss_mlp": 0.01029958,
      "balance_loss_clip": 1.04989862,
      "balance_loss_mlp": 1.0200634,
      "epoch": 0.39078939457704565,
      "flos": 20558572122240.0,
      "grad_norm": 2.012561363264574,
      "language_loss": 0.73446351,
      "learning_rate": 2.7831430082347143e-06,
      "loss": 0.75667036,
      "num_input_tokens_seen": 70005260,
      "step": 3250,
      "time_per_iteration": 2.7663161754608154
    },
    {
      "auxiliary_loss_clip": 0.01196161,
      "auxiliary_loss_mlp": 0.01207925,
      "balance_loss_clip": 1.05546451,
      "balance_loss_mlp": 1.0002749,
      "epoch": 0.3909096374676847,
      "flos": 22783417557120.0,
      "grad_norm": 2.476902743675246,
      "language_loss": 0.82356572,
      "learning_rate": 2.7824261798959373e-06,
      "loss": 0.8476066,
      "num_input_tokens_seen": 70023440,
      "step": 3251,
      "time_per_iteration": 3.486757755279541
    },
    {
      "auxiliary_loss_clip": 0.01198502,
      "auxiliary_loss_mlp": 0.01030578,
      "balance_loss_clip": 1.05103862,
      "balance_loss_mlp": 1.02147663,
      "epoch": 0.3910298803583238,
      "flos": 23003119094400.0,
      "grad_norm": 1.9184582862409727,
      "language_loss": 0.80236876,
      "learning_rate": 2.78170923286373e-06,
      "loss": 0.82465947,
      "num_input_tokens_seen": 70043040,
      "step": 3252,
      "time_per_iteration": 2.690263271331787
    },
    {
      "auxiliary_loss_clip": 0.0119866,
      "auxiliary_loss_mlp": 0.01024845,
      "balance_loss_clip": 1.04983807,
      "balance_loss_mlp": 1.0160358,
      "epoch": 0.3911501232489629,
      "flos": 24316264500480.0,
      "grad_norm": 4.426144560180236,
      "language_loss": 0.84315187,
      "learning_rate": 2.780992167246854e-06,
      "loss": 0.86538696,
      "num_input_tokens_seen": 70060565,
      "step": 3253,
      "time_per_iteration": 2.7285804748535156
    },
    {
      "auxiliary_loss_clip": 0.01112173,
      "auxiliary_loss_mlp": 0.01006637,
      "balance_loss_clip": 1.02919078,
      "balance_loss_mlp": 1.00559378,
      "epoch": 0.391270366139602,
      "flos": 60869054684160.0,
      "grad_norm": 0.9775697212481913,
      "language_loss": 0.72149479,
      "learning_rate": 2.7802749831540883e-06,
      "loss": 0.74268293,
      "num_input_tokens_seen": 70119465,
      "step": 3254,
      "time_per_iteration": 3.17732572555542
    },
    {
      "auxiliary_loss_clip": 0.01207036,
      "auxiliary_loss_mlp": 0.01024619,
      "balance_loss_clip": 1.05313218,
      "balance_loss_mlp": 1.0173502,
      "epoch": 0.3913906090302411,
      "flos": 21543494025600.0,
      "grad_norm": 6.711468831810802,
      "language_loss": 0.81848842,
      "learning_rate": 2.7795576806942268e-06,
      "loss": 0.84080499,
      "num_input_tokens_seen": 70138270,
      "step": 3255,
      "time_per_iteration": 2.7037341594696045
    },
    {
      "auxiliary_loss_clip": 0.01109102,
      "auxiliary_loss_mlp": 0.01001885,
      "balance_loss_clip": 1.0373342,
      "balance_loss_mlp": 1.00037694,
      "epoch": 0.3915108519208802,
      "flos": 49839953702400.0,
      "grad_norm": 0.7540359792990026,
      "language_loss": 0.54772103,
      "learning_rate": 2.778840259976085e-06,
      "loss": 0.56883085,
      "num_input_tokens_seen": 70193500,
      "step": 3256,
      "time_per_iteration": 3.188387393951416
    },
    {
      "auxiliary_loss_clip": 0.01198081,
      "auxiliary_loss_mlp": 0.01029495,
      "balance_loss_clip": 1.05675519,
      "balance_loss_mlp": 1.02049446,
      "epoch": 0.39163109481151925,
      "flos": 16506447960960.0,
      "grad_norm": 2.5800724658470706,
      "language_loss": 0.77133322,
      "learning_rate": 2.778122721108495e-06,
      "loss": 0.7936089,
      "num_input_tokens_seen": 70211730,
      "step": 3257,
      "time_per_iteration": 2.5700745582580566
    },
    {
      "auxiliary_loss_clip": 0.01191598,
      "auxiliary_loss_mlp": 0.01026659,
      "balance_loss_clip": 1.05602026,
      "balance_loss_mlp": 1.01811719,
      "epoch": 0.39175133770215836,
      "flos": 26067484177920.0,
      "grad_norm": 14.01426423598521,
      "language_loss": 0.88392055,
      "learning_rate": 2.7774050642003076e-06,
      "loss": 0.90610313,
      "num_input_tokens_seen": 70232540,
      "step": 3258,
      "time_per_iteration": 2.6061246395111084
    },
    {
      "auxiliary_loss_clip": 0.01196756,
      "auxiliary_loss_mlp": 0.01034003,
      "balance_loss_clip": 1.05846548,
      "balance_loss_mlp": 1.02426386,
      "epoch": 0.3918715805927975,
      "flos": 21872076664320.0,
      "grad_norm": 1.9644106744675338,
      "language_loss": 0.93350577,
      "learning_rate": 2.7766872893603896e-06,
      "loss": 0.95581341,
      "num_input_tokens_seen": 70252515,
      "step": 3259,
      "time_per_iteration": 2.6644537448883057
    },
    {
      "auxiliary_loss_clip": 0.01195266,
      "auxiliary_loss_mlp": 0.01035788,
      "balance_loss_clip": 1.05530083,
      "balance_loss_mlp": 1.02742577,
      "epoch": 0.39199182348343653,
      "flos": 20376181837440.0,
      "grad_norm": 1.687875428807008,
      "language_loss": 0.73666596,
      "learning_rate": 2.7759693966976275e-06,
      "loss": 0.75897646,
      "num_input_tokens_seen": 70271020,
      "step": 3260,
      "time_per_iteration": 2.6439664363861084
    },
    {
      "auxiliary_loss_clip": 0.01194222,
      "auxiliary_loss_mlp": 0.01029322,
      "balance_loss_clip": 1.04916596,
      "balance_loss_mlp": 1.02010715,
      "epoch": 0.39211206637407564,
      "flos": 21683545153920.0,
      "grad_norm": 2.315926216534867,
      "language_loss": 0.85454077,
      "learning_rate": 2.7752513863209242e-06,
      "loss": 0.87677622,
      "num_input_tokens_seen": 70289600,
      "step": 3261,
      "time_per_iteration": 2.779966115951538
    },
    {
      "auxiliary_loss_clip": 0.01193036,
      "auxiliary_loss_mlp": 0.01208634,
      "balance_loss_clip": 1.05591989,
      "balance_loss_mlp": 1.00025964,
      "epoch": 0.39223230926471475,
      "flos": 21066276908160.0,
      "grad_norm": 2.2908934344075176,
      "language_loss": 0.84681511,
      "learning_rate": 2.774533258339203e-06,
      "loss": 0.87083185,
      "num_input_tokens_seen": 70307060,
      "step": 3262,
      "time_per_iteration": 2.655149221420288
    },
    {
      "auxiliary_loss_clip": 0.01201326,
      "auxiliary_loss_mlp": 0.01033426,
      "balance_loss_clip": 1.04715514,
      "balance_loss_mlp": 1.02339399,
      "epoch": 0.3923525521553538,
      "flos": 17603016312960.0,
      "grad_norm": 2.368769940611315,
      "language_loss": 0.80007052,
      "learning_rate": 2.7738150128614014e-06,
      "loss": 0.82241809,
      "num_input_tokens_seen": 70324465,
      "step": 3263,
      "time_per_iteration": 2.69942045211792
    },
    {
      "auxiliary_loss_clip": 0.01189357,
      "auxiliary_loss_mlp": 0.01028516,
      "balance_loss_clip": 1.05100393,
      "balance_loss_mlp": 1.01950979,
      "epoch": 0.3924727950459929,
      "flos": 20558284813440.0,
      "grad_norm": 2.191855067581044,
      "language_loss": 0.89492369,
      "learning_rate": 2.7730966499964777e-06,
      "loss": 0.91710234,
      "num_input_tokens_seen": 70341415,
      "step": 3264,
      "time_per_iteration": 2.6412060260772705
    },
    {
      "auxiliary_loss_clip": 0.01195084,
      "auxiliary_loss_mlp": 0.01031473,
      "balance_loss_clip": 1.05717826,
      "balance_loss_mlp": 1.02200758,
      "epoch": 0.39259303793663197,
      "flos": 16216110328320.0,
      "grad_norm": 2.655170587885102,
      "language_loss": 0.80894816,
      "learning_rate": 2.772378169853408e-06,
      "loss": 0.83121371,
      "num_input_tokens_seen": 70358985,
      "step": 3265,
      "time_per_iteration": 2.5732903480529785
    },
    {
      "auxiliary_loss_clip": 0.0119948,
      "auxiliary_loss_mlp": 0.01029006,
      "balance_loss_clip": 1.0537411,
      "balance_loss_mlp": 1.01980269,
      "epoch": 0.3927132808272711,
      "flos": 16797001075200.0,
      "grad_norm": 1.8027983226479423,
      "language_loss": 0.74495387,
      "learning_rate": 2.771659572541183e-06,
      "loss": 0.76723874,
      "num_input_tokens_seen": 70376915,
      "step": 3266,
      "time_per_iteration": 2.6296186447143555
    },
    {
      "auxiliary_loss_clip": 0.01200756,
      "auxiliary_loss_mlp": 0.01027144,
      "balance_loss_clip": 1.05931234,
      "balance_loss_mlp": 1.01846588,
      "epoch": 0.3928335237179102,
      "flos": 20267228908800.0,
      "grad_norm": 2.0646585038582983,
      "language_loss": 0.8709479,
      "learning_rate": 2.7709408581688143e-06,
      "loss": 0.89322698,
      "num_input_tokens_seen": 70396900,
      "step": 3267,
      "time_per_iteration": 2.599824905395508
    },
    {
      "auxiliary_loss_clip": 0.01208468,
      "auxiliary_loss_mlp": 0.01029705,
      "balance_loss_clip": 1.05418372,
      "balance_loss_mlp": 1.02109802,
      "epoch": 0.39295376660854925,
      "flos": 24973250209920.0,
      "grad_norm": 1.7400255660389758,
      "language_loss": 0.8778528,
      "learning_rate": 2.7702220268453307e-06,
      "loss": 0.90023458,
      "num_input_tokens_seen": 70417260,
      "step": 3268,
      "time_per_iteration": 2.670276403427124
    },
    {
      "auxiliary_loss_clip": 0.01199535,
      "auxiliary_loss_mlp": 0.01030229,
      "balance_loss_clip": 1.05337203,
      "balance_loss_mlp": 1.02076352,
      "epoch": 0.39307400949918836,
      "flos": 18697788984960.0,
      "grad_norm": 2.4115727260842648,
      "language_loss": 0.84874213,
      "learning_rate": 2.7695030786797785e-06,
      "loss": 0.87103975,
      "num_input_tokens_seen": 70433155,
      "step": 3269,
      "time_per_iteration": 3.533442974090576
    },
    {
      "auxiliary_loss_clip": 0.01196503,
      "auxiliary_loss_mlp": 0.01027808,
      "balance_loss_clip": 1.05027306,
      "balance_loss_mlp": 1.01896834,
      "epoch": 0.39319425238982747,
      "flos": 22415476590720.0,
      "grad_norm": 2.651958524169556,
      "language_loss": 0.75268137,
      "learning_rate": 2.7687840137812206e-06,
      "loss": 0.77492452,
      "num_input_tokens_seen": 70451240,
      "step": 3270,
      "time_per_iteration": 2.7341864109039307
    },
    {
      "auxiliary_loss_clip": 0.01097985,
      "auxiliary_loss_mlp": 0.01001706,
      "balance_loss_clip": 1.02669358,
      "balance_loss_mlp": 1.00055015,
      "epoch": 0.3933144952804665,
      "flos": 66192954762240.0,
      "grad_norm": 0.7946974830126698,
      "language_loss": 0.62041724,
      "learning_rate": 2.7680648322587395e-06,
      "loss": 0.64141417,
      "num_input_tokens_seen": 70516115,
      "step": 3271,
      "time_per_iteration": 4.089141845703125
    },
    {
      "auxiliary_loss_clip": 0.01191719,
      "auxiliary_loss_mlp": 0.01030695,
      "balance_loss_clip": 1.05649972,
      "balance_loss_mlp": 1.02226079,
      "epoch": 0.39343473817110564,
      "flos": 15487159720320.0,
      "grad_norm": 1.995275827536412,
      "language_loss": 0.80939865,
      "learning_rate": 2.7673455342214334e-06,
      "loss": 0.83162284,
      "num_input_tokens_seen": 70533105,
      "step": 3272,
      "time_per_iteration": 2.7486026287078857
    },
    {
      "auxiliary_loss_clip": 0.01196413,
      "auxiliary_loss_mlp": 0.01025276,
      "balance_loss_clip": 1.05657995,
      "balance_loss_mlp": 1.01706195,
      "epoch": 0.39355498106174475,
      "flos": 21324905809920.0,
      "grad_norm": 1.9524130670697908,
      "language_loss": 0.7587744,
      "learning_rate": 2.7666261197784198e-06,
      "loss": 0.78099126,
      "num_input_tokens_seen": 70551920,
      "step": 3273,
      "time_per_iteration": 2.672297954559326
    },
    {
      "auxiliary_loss_clip": 0.01194939,
      "auxiliary_loss_mlp": 0.01028482,
      "balance_loss_clip": 1.05826139,
      "balance_loss_mlp": 1.01960707,
      "epoch": 0.3936752239523838,
      "flos": 13296357400320.0,
      "grad_norm": 2.084187353841783,
      "language_loss": 0.76291364,
      "learning_rate": 2.7659065890388336e-06,
      "loss": 0.78514785,
      "num_input_tokens_seen": 70567920,
      "step": 3274,
      "time_per_iteration": 3.577207088470459
    },
    {
      "auxiliary_loss_clip": 0.01200179,
      "auxiliary_loss_mlp": 0.01034347,
      "balance_loss_clip": 1.05337787,
      "balance_loss_mlp": 1.02545381,
      "epoch": 0.3937954668430229,
      "flos": 16800161472000.0,
      "grad_norm": 2.346154438219563,
      "language_loss": 0.85272008,
      "learning_rate": 2.7651869421118266e-06,
      "loss": 0.87506533,
      "num_input_tokens_seen": 70584530,
      "step": 3275,
      "time_per_iteration": 2.614105701446533
    },
    {
      "auxiliary_loss_clip": 0.01202195,
      "auxiliary_loss_mlp": 0.01029901,
      "balance_loss_clip": 1.05989432,
      "balance_loss_mlp": 1.020859,
      "epoch": 0.393915709733662,
      "flos": 21064229832960.0,
      "grad_norm": 2.131857905795838,
      "language_loss": 0.83077693,
      "learning_rate": 2.76446717910657e-06,
      "loss": 0.8530978,
      "num_input_tokens_seen": 70605235,
      "step": 3276,
      "time_per_iteration": 2.609863758087158
    },
    {
      "auxiliary_loss_clip": 0.01194761,
      "auxiliary_loss_mlp": 0.01028839,
      "balance_loss_clip": 1.05664158,
      "balance_loss_mlp": 1.02042842,
      "epoch": 0.3940359526243011,
      "flos": 17165265264000.0,
      "grad_norm": 2.473805312612611,
      "language_loss": 0.7697311,
      "learning_rate": 2.763747300132249e-06,
      "loss": 0.79196709,
      "num_input_tokens_seen": 70622675,
      "step": 3277,
      "time_per_iteration": 2.537806272506714
    },
    {
      "auxiliary_loss_clip": 0.01194462,
      "auxiliary_loss_mlp": 0.01028845,
      "balance_loss_clip": 1.05793464,
      "balance_loss_mlp": 1.01981449,
      "epoch": 0.3941561955149402,
      "flos": 20995856294400.0,
      "grad_norm": 1.7423037062634206,
      "language_loss": 0.86569166,
      "learning_rate": 2.7630273052980704e-06,
      "loss": 0.88792473,
      "num_input_tokens_seen": 70643265,
      "step": 3278,
      "time_per_iteration": 3.5251095294952393
    },
    {
      "auxiliary_loss_clip": 0.01186446,
      "auxiliary_loss_mlp": 0.01027165,
      "balance_loss_clip": 1.05272961,
      "balance_loss_mlp": 1.01845026,
      "epoch": 0.39427643840557924,
      "flos": 18843406721280.0,
      "grad_norm": 2.3331736146634316,
      "language_loss": 0.67239487,
      "learning_rate": 2.7623071947132554e-06,
      "loss": 0.69453096,
      "num_input_tokens_seen": 70660295,
      "step": 3279,
      "time_per_iteration": 2.615691661834717
    },
    {
      "auxiliary_loss_clip": 0.01202574,
      "auxiliary_loss_mlp": 0.01032674,
      "balance_loss_clip": 1.05287182,
      "balance_loss_mlp": 1.02353024,
      "epoch": 0.39439668129621835,
      "flos": 23258659426560.0,
      "grad_norm": 2.97946146435981,
      "language_loss": 0.78705263,
      "learning_rate": 2.7615869684870458e-06,
      "loss": 0.80940515,
      "num_input_tokens_seen": 70679605,
      "step": 3280,
      "time_per_iteration": 2.694943904876709
    },
    {
      "auxiliary_loss_clip": 0.01195827,
      "auxiliary_loss_mlp": 0.01031129,
      "balance_loss_clip": 1.05847454,
      "balance_loss_mlp": 1.02227736,
      "epoch": 0.39451692418685746,
      "flos": 26652289507200.0,
      "grad_norm": 1.6462115422854204,
      "language_loss": 0.84571451,
      "learning_rate": 2.7608666267286986e-06,
      "loss": 0.86798406,
      "num_input_tokens_seen": 70699835,
      "step": 3281,
      "time_per_iteration": 2.6517295837402344
    },
    {
      "auxiliary_loss_clip": 0.01200941,
      "auxiliary_loss_mlp": 0.01029669,
      "balance_loss_clip": 1.04501927,
      "balance_loss_mlp": 1.02040029,
      "epoch": 0.3946371670774965,
      "flos": 18258709132800.0,
      "grad_norm": 2.5377511105881316,
      "language_loss": 0.86565065,
      "learning_rate": 2.760146169547489e-06,
      "loss": 0.88795674,
      "num_input_tokens_seen": 70716600,
      "step": 3282,
      "time_per_iteration": 2.705167055130005
    },
    {
      "auxiliary_loss_clip": 0.0120177,
      "auxiliary_loss_mlp": 0.0103121,
      "balance_loss_clip": 1.05844498,
      "balance_loss_mlp": 1.02201843,
      "epoch": 0.39475740996813563,
      "flos": 24206126423040.0,
      "grad_norm": 1.5362435135316963,
      "language_loss": 0.76442033,
      "learning_rate": 2.75942559705271e-06,
      "loss": 0.78675008,
      "num_input_tokens_seen": 70736335,
      "step": 3283,
      "time_per_iteration": 2.713721513748169
    },
    {
      "auxiliary_loss_clip": 0.01194565,
      "auxiliary_loss_mlp": 0.01032935,
      "balance_loss_clip": 1.05742025,
      "balance_loss_mlp": 1.02391648,
      "epoch": 0.39487765285877474,
      "flos": 19317858491520.0,
      "grad_norm": 2.047915660855441,
      "language_loss": 0.88970792,
      "learning_rate": 2.7587049093536713e-06,
      "loss": 0.91198289,
      "num_input_tokens_seen": 70752665,
      "step": 3284,
      "time_per_iteration": 2.691941738128662
    },
    {
      "auxiliary_loss_clip": 0.01199159,
      "auxiliary_loss_mlp": 0.01034991,
      "balance_loss_clip": 1.05550444,
      "balance_loss_mlp": 1.0256505,
      "epoch": 0.3949978957494138,
      "flos": 17311744926720.0,
      "grad_norm": 1.9504676366722409,
      "language_loss": 0.81009126,
      "learning_rate": 2.757984106559701e-06,
      "loss": 0.83243287,
      "num_input_tokens_seen": 70771650,
      "step": 3285,
      "time_per_iteration": 2.6880850791931152
    },
    {
      "auxiliary_loss_clip": 0.01192208,
      "auxiliary_loss_mlp": 0.01027856,
      "balance_loss_clip": 1.05629814,
      "balance_loss_mlp": 1.01851046,
      "epoch": 0.3951181386400529,
      "flos": 36317861280000.0,
      "grad_norm": 3.6257942324661214,
      "language_loss": 0.71084058,
      "learning_rate": 2.7572631887801446e-06,
      "loss": 0.73304117,
      "num_input_tokens_seen": 70793275,
      "step": 3286,
      "time_per_iteration": 2.810076951980591
    },
    {
      "auxiliary_loss_clip": 0.01197896,
      "auxiliary_loss_mlp": 0.01029974,
      "balance_loss_clip": 1.05668688,
      "balance_loss_mlp": 1.02043724,
      "epoch": 0.395238381530692,
      "flos": 23110348170240.0,
      "grad_norm": 2.0152731439445972,
      "language_loss": 0.76723957,
      "learning_rate": 2.7565421561243654e-06,
      "loss": 0.78951824,
      "num_input_tokens_seen": 70811440,
      "step": 3287,
      "time_per_iteration": 2.702540636062622
    },
    {
      "auxiliary_loss_clip": 0.01191237,
      "auxiliary_loss_mlp": 0.0102861,
      "balance_loss_clip": 1.05022144,
      "balance_loss_mlp": 1.01958573,
      "epoch": 0.3953586244213311,
      "flos": 24347614095360.0,
      "grad_norm": 2.285867675158336,
      "language_loss": 0.82512546,
      "learning_rate": 2.7558210087017413e-06,
      "loss": 0.84732395,
      "num_input_tokens_seen": 70831375,
      "step": 3288,
      "time_per_iteration": 2.725078821182251
    },
    {
      "auxiliary_loss_clip": 0.01197529,
      "auxiliary_loss_mlp": 0.01030547,
      "balance_loss_clip": 1.05722916,
      "balance_loss_mlp": 1.02063513,
      "epoch": 0.3954788673119702,
      "flos": 23440080044160.0,
      "grad_norm": 2.2001517427594037,
      "language_loss": 0.73240018,
      "learning_rate": 2.7550997466216724e-06,
      "loss": 0.75468093,
      "num_input_tokens_seen": 70849170,
      "step": 3289,
      "time_per_iteration": 2.7550301551818848
    },
    {
      "auxiliary_loss_clip": 0.01196122,
      "auxiliary_loss_mlp": 0.01032415,
      "balance_loss_clip": 1.05875969,
      "balance_loss_mlp": 1.02338481,
      "epoch": 0.3955991102026093,
      "flos": 17494063384320.0,
      "grad_norm": 1.8697489245258,
      "language_loss": 0.81474167,
      "learning_rate": 2.7543783699935714e-06,
      "loss": 0.83702707,
      "num_input_tokens_seen": 70867200,
      "step": 3290,
      "time_per_iteration": 2.6318917274475098
    },
    {
      "auxiliary_loss_clip": 0.01197678,
      "auxiliary_loss_mlp": 0.01033544,
      "balance_loss_clip": 1.06013691,
      "balance_loss_mlp": 1.02437139,
      "epoch": 0.39571935309324835,
      "flos": 18221326053120.0,
      "grad_norm": 2.97281915432254,
      "language_loss": 0.86209309,
      "learning_rate": 2.753656878926872e-06,
      "loss": 0.88440526,
      "num_input_tokens_seen": 70883080,
      "step": 3291,
      "time_per_iteration": 2.6160221099853516
    },
    {
      "auxiliary_loss_clip": 0.01188018,
      "auxiliary_loss_mlp": 0.0103005,
      "balance_loss_clip": 1.05217195,
      "balance_loss_mlp": 1.02049494,
      "epoch": 0.39583959598388746,
      "flos": 17748813617280.0,
      "grad_norm": 1.9473990316332352,
      "language_loss": 0.74405438,
      "learning_rate": 2.752935273531023e-06,
      "loss": 0.76623511,
      "num_input_tokens_seen": 70901230,
      "step": 3292,
      "time_per_iteration": 2.608811855316162
    },
    {
      "auxiliary_loss_clip": 0.01197462,
      "auxiliary_loss_mlp": 0.01032599,
      "balance_loss_clip": 1.0559845,
      "balance_loss_mlp": 1.02274644,
      "epoch": 0.39595983887452657,
      "flos": 19352368483200.0,
      "grad_norm": 1.7565189912425259,
      "language_loss": 0.78537679,
      "learning_rate": 2.752213553915492e-06,
      "loss": 0.80767739,
      "num_input_tokens_seen": 70919585,
      "step": 3293,
      "time_per_iteration": 2.71755313873291
    },
    {
      "auxiliary_loss_clip": 0.0110396,
      "auxiliary_loss_mlp": 0.01001032,
      "balance_loss_clip": 1.02583003,
      "balance_loss_mlp": 0.99967939,
      "epoch": 0.3960800817651656,
      "flos": 60682282940160.0,
      "grad_norm": 0.8163156036389873,
      "language_loss": 0.66107464,
      "learning_rate": 2.751491720189762e-06,
      "loss": 0.6821245,
      "num_input_tokens_seen": 70977695,
      "step": 3294,
      "time_per_iteration": 3.142176628112793
    },
    {
      "auxiliary_loss_clip": 0.01198285,
      "auxiliary_loss_mlp": 0.01208838,
      "balance_loss_clip": 1.0566771,
      "balance_loss_mlp": 1.00039434,
      "epoch": 0.39620032465580474,
      "flos": 16836718538880.0,
      "grad_norm": 2.029166025612379,
      "language_loss": 0.91433406,
      "learning_rate": 2.7507697724633364e-06,
      "loss": 0.93840522,
      "num_input_tokens_seen": 70994455,
      "step": 3295,
      "time_per_iteration": 2.607370376586914
    },
    {
      "auxiliary_loss_clip": 0.01108489,
      "auxiliary_loss_mlp": 0.01000837,
      "balance_loss_clip": 1.03006005,
      "balance_loss_mlp": 0.9991684,
      "epoch": 0.3963205675464438,
      "flos": 69071445941760.0,
      "grad_norm": 0.774367190097767,
      "language_loss": 0.54590797,
      "learning_rate": 2.7500477108457327e-06,
      "loss": 0.56700122,
      "num_input_tokens_seen": 71046465,
      "step": 3296,
      "time_per_iteration": 3.9085638523101807
    },
    {
      "auxiliary_loss_clip": 0.01192893,
      "auxiliary_loss_mlp": 0.01029349,
      "balance_loss_clip": 1.05378389,
      "balance_loss_mlp": 1.01994324,
      "epoch": 0.3964408104370829,
      "flos": 25667439431040.0,
      "grad_norm": 2.5267473477877567,
      "language_loss": 0.80648518,
      "learning_rate": 2.7493255354464877e-06,
      "loss": 0.82870758,
      "num_input_tokens_seen": 71064275,
      "step": 3297,
      "time_per_iteration": 2.5594851970672607
    },
    {
      "auxiliary_loss_clip": 0.01194109,
      "auxiliary_loss_mlp": 0.01028854,
      "balance_loss_clip": 1.0402422,
      "balance_loss_mlp": 1.02008033,
      "epoch": 0.396561053327722,
      "flos": 24277480790400.0,
      "grad_norm": 2.5093291952563805,
      "language_loss": 0.76215696,
      "learning_rate": 2.748603246375156e-06,
      "loss": 0.78438663,
      "num_input_tokens_seen": 71082290,
      "step": 3298,
      "time_per_iteration": 3.46345591545105
    },
    {
      "auxiliary_loss_clip": 0.01193165,
      "auxiliary_loss_mlp": 0.0103208,
      "balance_loss_clip": 1.05821753,
      "balance_loss_mlp": 1.02304995,
      "epoch": 0.39668129621836107,
      "flos": 20522302364160.0,
      "grad_norm": 3.129635668050383,
      "language_loss": 0.69964856,
      "learning_rate": 2.7478808437413055e-06,
      "loss": 0.72190106,
      "num_input_tokens_seen": 71101700,
      "step": 3299,
      "time_per_iteration": 2.610675096511841
    },
    {
      "auxiliary_loss_clip": 0.01202661,
      "auxiliary_loss_mlp": 0.01027021,
      "balance_loss_clip": 1.05519056,
      "balance_loss_mlp": 1.01781774,
      "epoch": 0.3968015391090002,
      "flos": 27052585649280.0,
      "grad_norm": 3.7166398619434653,
      "language_loss": 0.66028726,
      "learning_rate": 2.7471583276545263e-06,
      "loss": 0.68258405,
      "num_input_tokens_seen": 71122360,
      "step": 3300,
      "time_per_iteration": 2.7113845348358154
    },
    {
      "auxiliary_loss_clip": 0.01195755,
      "auxiliary_loss_mlp": 0.01031227,
      "balance_loss_clip": 1.05328035,
      "balance_loss_mlp": 1.02258468,
      "epoch": 0.3969217819996393,
      "flos": 12531819392640.0,
      "grad_norm": 1.9149958825142677,
      "language_loss": 0.71294415,
      "learning_rate": 2.7464356982244224e-06,
      "loss": 0.73521399,
      "num_input_tokens_seen": 71140360,
      "step": 3301,
      "time_per_iteration": 3.7455527782440186
    },
    {
      "auxiliary_loss_clip": 0.01098189,
      "auxiliary_loss_mlp": 0.01001147,
      "balance_loss_clip": 1.03091657,
      "balance_loss_mlp": 0.99951375,
      "epoch": 0.39704202489027834,
      "flos": 66241399230720.0,
      "grad_norm": 0.7797658146490974,
      "language_loss": 0.61735642,
      "learning_rate": 2.745712955560617e-06,
      "loss": 0.63834977,
      "num_input_tokens_seen": 71196565,
      "step": 3302,
      "time_per_iteration": 3.1401305198669434
    },
    {
      "auxiliary_loss_clip": 0.01203321,
      "auxiliary_loss_mlp": 0.01027165,
      "balance_loss_clip": 1.05013037,
      "balance_loss_mlp": 1.01738381,
      "epoch": 0.39716226778091746,
      "flos": 16982982720000.0,
      "grad_norm": 2.396874211017772,
      "language_loss": 0.77172488,
      "learning_rate": 2.7449900997727496e-06,
      "loss": 0.79402977,
      "num_input_tokens_seen": 71214675,
      "step": 3303,
      "time_per_iteration": 2.6935367584228516
    },
    {
      "auxiliary_loss_clip": 0.01194322,
      "auxiliary_loss_mlp": 0.01027296,
      "balance_loss_clip": 1.05560708,
      "balance_loss_mlp": 1.01930904,
      "epoch": 0.39728251067155657,
      "flos": 23477139901440.0,
      "grad_norm": 1.674011468060072,
      "language_loss": 0.84202266,
      "learning_rate": 2.744267130970476e-06,
      "loss": 0.86423886,
      "num_input_tokens_seen": 71234400,
      "step": 3304,
      "time_per_iteration": 3.6816112995147705
    },
    {
      "auxiliary_loss_clip": 0.01193925,
      "auxiliary_loss_mlp": 0.01028985,
      "balance_loss_clip": 1.05545044,
      "balance_loss_mlp": 1.01975834,
      "epoch": 0.3974027535621956,
      "flos": 20704441253760.0,
      "grad_norm": 1.746758893555768,
      "language_loss": 0.76869327,
      "learning_rate": 2.7435440492634697e-06,
      "loss": 0.79092228,
      "num_input_tokens_seen": 71253725,
      "step": 3305,
      "time_per_iteration": 2.668192148208618
    },
    {
      "auxiliary_loss_clip": 0.01197567,
      "auxiliary_loss_mlp": 0.01031631,
      "balance_loss_clip": 1.05396831,
      "balance_loss_mlp": 1.02176058,
      "epoch": 0.39752299645283473,
      "flos": 21543278544000.0,
      "grad_norm": 1.8339149185368386,
      "language_loss": 0.6717962,
      "learning_rate": 2.7428208547614228e-06,
      "loss": 0.69408816,
      "num_input_tokens_seen": 71273220,
      "step": 3306,
      "time_per_iteration": 2.6955764293670654
    },
    {
      "auxiliary_loss_clip": 0.01195753,
      "auxiliary_loss_mlp": 0.01033806,
      "balance_loss_clip": 1.05561829,
      "balance_loss_mlp": 1.02485299,
      "epoch": 0.39764323934347384,
      "flos": 19208295031680.0,
      "grad_norm": 1.992943907985529,
      "language_loss": 0.77098465,
      "learning_rate": 2.742097547574043e-06,
      "loss": 0.79328024,
      "num_input_tokens_seen": 71291445,
      "step": 3307,
      "time_per_iteration": 2.625645637512207
    },
    {
      "auxiliary_loss_clip": 0.01202683,
      "auxiliary_loss_mlp": 0.01208889,
      "balance_loss_clip": 1.05479038,
      "balance_loss_mlp": 1.00036764,
      "epoch": 0.3977634822341129,
      "flos": 20850202644480.0,
      "grad_norm": 1.792573903532521,
      "language_loss": 0.77745974,
      "learning_rate": 2.7413741278110544e-06,
      "loss": 0.80157542,
      "num_input_tokens_seen": 71310135,
      "step": 3308,
      "time_per_iteration": 2.6517608165740967
    },
    {
      "auxiliary_loss_clip": 0.0120246,
      "auxiliary_loss_mlp": 0.01032043,
      "balance_loss_clip": 1.05605209,
      "balance_loss_mlp": 1.02239299,
      "epoch": 0.397883725124752,
      "flos": 39786042038400.0,
      "grad_norm": 1.9913385245352444,
      "language_loss": 0.69356877,
      "learning_rate": 2.7406505955822016e-06,
      "loss": 0.71591377,
      "num_input_tokens_seen": 71331160,
      "step": 3309,
      "time_per_iteration": 2.7644405364990234
    },
    {
      "auxiliary_loss_clip": 0.0119783,
      "auxiliary_loss_mlp": 0.01030231,
      "balance_loss_clip": 1.05371857,
      "balance_loss_mlp": 1.02143312,
      "epoch": 0.39800396801539106,
      "flos": 17379507934080.0,
      "grad_norm": 2.5395951391065643,
      "language_loss": 0.6627894,
      "learning_rate": 2.7399269509972415e-06,
      "loss": 0.68507004,
      "num_input_tokens_seen": 71345315,
      "step": 3310,
      "time_per_iteration": 2.6026952266693115
    },
    {
      "auxiliary_loss_clip": 0.01188532,
      "auxiliary_loss_mlp": 0.01033474,
      "balance_loss_clip": 1.04849005,
      "balance_loss_mlp": 1.02346611,
      "epoch": 0.3981242109060302,
      "flos": 19202764337280.0,
      "grad_norm": 2.6291285204492914,
      "language_loss": 0.85129589,
      "learning_rate": 2.7392031941659514e-06,
      "loss": 0.87351596,
      "num_input_tokens_seen": 71363160,
      "step": 3311,
      "time_per_iteration": 2.598923683166504
    },
    {
      "auxiliary_loss_clip": 0.01198144,
      "auxiliary_loss_mlp": 0.01035724,
      "balance_loss_clip": 1.05748212,
      "balance_loss_mlp": 1.02621746,
      "epoch": 0.3982444537966693,
      "flos": 24565124903040.0,
      "grad_norm": 2.0920850862565934,
      "language_loss": 0.86027253,
      "learning_rate": 2.7384793251981244e-06,
      "loss": 0.88261127,
      "num_input_tokens_seen": 71382145,
      "step": 3312,
      "time_per_iteration": 2.7759506702423096
    },
    {
      "auxiliary_loss_clip": 0.01200965,
      "auxiliary_loss_mlp": 0.01027756,
      "balance_loss_clip": 1.05608666,
      "balance_loss_mlp": 1.01920283,
      "epoch": 0.39836469668730834,
      "flos": 26213856099840.0,
      "grad_norm": 1.6739546737652282,
      "language_loss": 0.80764627,
      "learning_rate": 2.737755344203571e-06,
      "loss": 0.82993352,
      "num_input_tokens_seen": 71402095,
      "step": 3313,
      "time_per_iteration": 2.7005584239959717
    },
    {
      "auxiliary_loss_clip": 0.01202438,
      "auxiliary_loss_mlp": 0.01030217,
      "balance_loss_clip": 1.06073856,
      "balance_loss_mlp": 1.02144337,
      "epoch": 0.39848493957794745,
      "flos": 27636134002560.0,
      "grad_norm": 1.598801229961996,
      "language_loss": 0.79973304,
      "learning_rate": 2.7370312512921186e-06,
      "loss": 0.82205963,
      "num_input_tokens_seen": 71423875,
      "step": 3314,
      "time_per_iteration": 2.6558117866516113
    },
    {
      "auxiliary_loss_clip": 0.01199628,
      "auxiliary_loss_mlp": 0.01030036,
      "balance_loss_clip": 1.05300462,
      "balance_loss_mlp": 1.01982009,
      "epoch": 0.39860518246858656,
      "flos": 12239326944000.0,
      "grad_norm": 2.3963245468505154,
      "language_loss": 0.76754177,
      "learning_rate": 2.736307046573611e-06,
      "loss": 0.78983837,
      "num_input_tokens_seen": 71439745,
      "step": 3315,
      "time_per_iteration": 2.653184652328491
    },
    {
      "auxiliary_loss_clip": 0.01191488,
      "auxiliary_loss_mlp": 0.0102911,
      "balance_loss_clip": 1.05680513,
      "balance_loss_mlp": 1.02059233,
      "epoch": 0.3987254253592256,
      "flos": 22379135005440.0,
      "grad_norm": 1.5413989989199792,
      "language_loss": 0.81789434,
      "learning_rate": 2.73558273015791e-06,
      "loss": 0.84010023,
      "num_input_tokens_seen": 71459575,
      "step": 3316,
      "time_per_iteration": 2.634066343307495
    },
    {
      "auxiliary_loss_clip": 0.01197033,
      "auxiliary_loss_mlp": 0.01031129,
      "balance_loss_clip": 1.05799699,
      "balance_loss_mlp": 1.02118731,
      "epoch": 0.3988456682498647,
      "flos": 23514020190720.0,
      "grad_norm": 2.1150487944374743,
      "language_loss": 0.70374876,
      "learning_rate": 2.734858302154894e-06,
      "loss": 0.72603035,
      "num_input_tokens_seen": 71481075,
      "step": 3317,
      "time_per_iteration": 2.6786255836486816
    },
    {
      "auxiliary_loss_clip": 0.01194038,
      "auxiliary_loss_mlp": 0.01031542,
      "balance_loss_clip": 1.05537963,
      "balance_loss_mlp": 1.02210629,
      "epoch": 0.39896591114050384,
      "flos": 19208761908480.0,
      "grad_norm": 2.025990626610595,
      "language_loss": 0.76417387,
      "learning_rate": 2.734133762674457e-06,
      "loss": 0.78642964,
      "num_input_tokens_seen": 71500665,
      "step": 3318,
      "time_per_iteration": 2.6137120723724365
    },
    {
      "auxiliary_loss_clip": 0.01198998,
      "auxiliary_loss_mlp": 0.01030216,
      "balance_loss_clip": 1.05471802,
      "balance_loss_mlp": 1.02085257,
      "epoch": 0.3990861540311429,
      "flos": 28401031146240.0,
      "grad_norm": 2.3290509420710324,
      "language_loss": 0.70751917,
      "learning_rate": 2.7334091118265124e-06,
      "loss": 0.72981131,
      "num_input_tokens_seen": 71522560,
      "step": 3319,
      "time_per_iteration": 2.7226626873016357
    },
    {
      "auxiliary_loss_clip": 0.01098067,
      "auxiliary_loss_mlp": 0.01004102,
      "balance_loss_clip": 1.02256656,
      "balance_loss_mlp": 1.00281417,
      "epoch": 0.399206396921782,
      "flos": 61758563086080.0,
      "grad_norm": 2.2798358225535322,
      "language_loss": 0.5779233,
      "learning_rate": 2.732684349720989e-06,
      "loss": 0.5989449,
      "num_input_tokens_seen": 71590520,
      "step": 3320,
      "time_per_iteration": 3.1658105850219727
    },
    {
      "auxiliary_loss_clip": 0.01203504,
      "auxiliary_loss_mlp": 0.01027948,
      "balance_loss_clip": 1.052073,
      "balance_loss_mlp": 1.01904917,
      "epoch": 0.3993266398124211,
      "flos": 28074567409920.0,
      "grad_norm": 1.715090455650931,
      "language_loss": 0.75799465,
      "learning_rate": 2.7319594764678318e-06,
      "loss": 0.78030908,
      "num_input_tokens_seen": 71612620,
      "step": 3321,
      "time_per_iteration": 2.730740547180176
    },
    {
      "auxiliary_loss_clip": 0.01209909,
      "auxiliary_loss_mlp": 0.01031835,
      "balance_loss_clip": 1.0533762,
      "balance_loss_mlp": 1.02217293,
      "epoch": 0.39944688270306017,
      "flos": 23225083188480.0,
      "grad_norm": 2.532537071797327,
      "language_loss": 0.83235204,
      "learning_rate": 2.7312344921770044e-06,
      "loss": 0.85476947,
      "num_input_tokens_seen": 71634320,
      "step": 3322,
      "time_per_iteration": 3.7651968002319336
    },
    {
      "auxiliary_loss_clip": 0.0119777,
      "auxiliary_loss_mlp": 0.0103272,
      "balance_loss_clip": 1.05248237,
      "balance_loss_mlp": 1.02385056,
      "epoch": 0.3995671255936993,
      "flos": 19390433921280.0,
      "grad_norm": 1.891038532182222,
      "language_loss": 0.77995384,
      "learning_rate": 2.7305093969584857e-06,
      "loss": 0.80225867,
      "num_input_tokens_seen": 71653145,
      "step": 3323,
      "time_per_iteration": 2.6460373401641846
    },
    {
      "auxiliary_loss_clip": 0.01189704,
      "auxiliary_loss_mlp": 0.01031192,
      "balance_loss_clip": 1.05444634,
      "balance_loss_mlp": 1.02229333,
      "epoch": 0.3996873684843384,
      "flos": 23842638743040.0,
      "grad_norm": 2.754249636237274,
      "language_loss": 0.79974592,
      "learning_rate": 2.729784190922272e-06,
      "loss": 0.82195497,
      "num_input_tokens_seen": 71674580,
      "step": 3324,
      "time_per_iteration": 3.540285587310791
    },
    {
      "auxiliary_loss_clip": 0.01098772,
      "auxiliary_loss_mlp": 0.01002234,
      "balance_loss_clip": 1.02118468,
      "balance_loss_mlp": 1.00091684,
      "epoch": 0.39980761137497745,
      "flos": 66576877280640.0,
      "grad_norm": 0.9507149758303252,
      "language_loss": 0.57181323,
      "learning_rate": 2.729058874178378e-06,
      "loss": 0.59282327,
      "num_input_tokens_seen": 71745260,
      "step": 3325,
      "time_per_iteration": 3.262040376663208
    },
    {
      "auxiliary_loss_clip": 0.0120421,
      "auxiliary_loss_mlp": 0.01034887,
      "balance_loss_clip": 1.05718803,
      "balance_loss_mlp": 1.02527833,
      "epoch": 0.39992785426561656,
      "flos": 28549162834560.0,
      "grad_norm": 2.0571974931464907,
      "language_loss": 0.69187629,
      "learning_rate": 2.7283334468368315e-06,
      "loss": 0.71426731,
      "num_input_tokens_seen": 71766540,
      "step": 3326,
      "time_per_iteration": 2.6882083415985107
    },
    {
      "auxiliary_loss_clip": 0.0120781,
      "auxiliary_loss_mlp": 0.01032349,
      "balance_loss_clip": 1.04415727,
      "balance_loss_mlp": 1.02286625,
      "epoch": 0.4000480971562556,
      "flos": 15049408671360.0,
      "grad_norm": 1.8203313226570619,
      "language_loss": 0.72934318,
      "learning_rate": 2.72760790900768e-06,
      "loss": 0.75174487,
      "num_input_tokens_seen": 71783125,
      "step": 3327,
      "time_per_iteration": 3.6922690868377686
    },
    {
      "auxiliary_loss_clip": 0.01197388,
      "auxiliary_loss_mlp": 0.01031447,
      "balance_loss_clip": 1.0599966,
      "balance_loss_mlp": 1.02256036,
      "epoch": 0.4001683400468947,
      "flos": 23915609222400.0,
      "grad_norm": 2.7876779815539208,
      "language_loss": 0.78870404,
      "learning_rate": 2.7268822608009875e-06,
      "loss": 0.81099236,
      "num_input_tokens_seen": 71802500,
      "step": 3328,
      "time_per_iteration": 2.646475315093994
    },
    {
      "auxiliary_loss_clip": 0.01207528,
      "auxiliary_loss_mlp": 0.0103028,
      "balance_loss_clip": 1.05460119,
      "balance_loss_mlp": 1.02135122,
      "epoch": 0.40028858293753383,
      "flos": 24352677912960.0,
      "grad_norm": 3.3877963445537107,
      "language_loss": 0.78195548,
      "learning_rate": 2.726156502326834e-06,
      "loss": 0.80433345,
      "num_input_tokens_seen": 71823800,
      "step": 3329,
      "time_per_iteration": 2.8005595207214355
    },
    {
      "auxiliary_loss_clip": 0.01126755,
      "auxiliary_loss_mlp": 0.01001571,
      "balance_loss_clip": 1.02791226,
      "balance_loss_mlp": 0.99996197,
      "epoch": 0.4004088258281729,
      "flos": 66787025800320.0,
      "grad_norm": 0.6971687944147075,
      "language_loss": 0.60273981,
      "learning_rate": 2.725430633695316e-06,
      "loss": 0.62402308,
      "num_input_tokens_seen": 71886880,
      "step": 3330,
      "time_per_iteration": 3.385920286178589
    },
    {
      "auxiliary_loss_clip": 0.01087516,
      "auxiliary_loss_mlp": 0.01003097,
      "balance_loss_clip": 1.01944327,
      "balance_loss_mlp": 1.00196493,
      "epoch": 0.400529068718812,
      "flos": 58598386473600.0,
      "grad_norm": 0.9001502471132683,
      "language_loss": 0.57909274,
      "learning_rate": 2.7247046550165485e-06,
      "loss": 0.59999895,
      "num_input_tokens_seen": 71939005,
      "step": 3331,
      "time_per_iteration": 4.162299871444702
    },
    {
      "auxiliary_loss_clip": 0.01199001,
      "auxiliary_loss_mlp": 0.01034391,
      "balance_loss_clip": 1.06113768,
      "balance_loss_mlp": 1.02521169,
      "epoch": 0.4006493116094511,
      "flos": 25377460934400.0,
      "grad_norm": 1.6219880114625604,
      "language_loss": 0.75971711,
      "learning_rate": 2.7239785664006606e-06,
      "loss": 0.78205109,
      "num_input_tokens_seen": 71962545,
      "step": 3332,
      "time_per_iteration": 2.7009520530700684
    },
    {
      "auxiliary_loss_clip": 0.01093716,
      "auxiliary_loss_mlp": 0.01002913,
      "balance_loss_clip": 1.01842368,
      "balance_loss_mlp": 1.00179243,
      "epoch": 0.40076955450009016,
      "flos": 60280729822080.0,
      "grad_norm": 0.8133168400375854,
      "language_loss": 0.61757863,
      "learning_rate": 2.7232523679578002e-06,
      "loss": 0.63854492,
      "num_input_tokens_seen": 72025625,
      "step": 3333,
      "time_per_iteration": 3.2431349754333496
    },
    {
      "auxiliary_loss_clip": 0.01196665,
      "auxiliary_loss_mlp": 0.01024054,
      "balance_loss_clip": 1.06015015,
      "balance_loss_mlp": 1.01545882,
      "epoch": 0.4008897973907293,
      "flos": 16617268396800.0,
      "grad_norm": 2.857625677669229,
      "language_loss": 0.79791862,
      "learning_rate": 2.7225260597981295e-06,
      "loss": 0.82012582,
      "num_input_tokens_seen": 72043330,
      "step": 3334,
      "time_per_iteration": 2.7097673416137695
    },
    {
      "auxiliary_loss_clip": 0.01200027,
      "auxiliary_loss_mlp": 0.01209417,
      "balance_loss_clip": 1.05431759,
      "balance_loss_mlp": 1.00031328,
      "epoch": 0.4010100402813684,
      "flos": 15377344865280.0,
      "grad_norm": 2.922061985476497,
      "language_loss": 0.7866711,
      "learning_rate": 2.721799642031831e-06,
      "loss": 0.8107655,
      "num_input_tokens_seen": 72059500,
      "step": 3335,
      "time_per_iteration": 2.6188595294952393
    },
    {
      "auxiliary_loss_clip": 0.01205024,
      "auxiliary_loss_mlp": 0.01033206,
      "balance_loss_clip": 1.05270493,
      "balance_loss_mlp": 1.02405143,
      "epoch": 0.40113028317200744,
      "flos": 13298835438720.0,
      "grad_norm": 2.1198669626505646,
      "language_loss": 0.77738369,
      "learning_rate": 2.721073114769101e-06,
      "loss": 0.79976606,
      "num_input_tokens_seen": 72077175,
      "step": 3336,
      "time_per_iteration": 2.807177782058716
    },
    {
      "auxiliary_loss_clip": 0.0119977,
      "auxiliary_loss_mlp": 0.01032095,
      "balance_loss_clip": 1.05448389,
      "balance_loss_mlp": 1.0231781,
      "epoch": 0.40125052606264655,
      "flos": 20668027841280.0,
      "grad_norm": 1.7768276624688408,
      "language_loss": 0.74846405,
      "learning_rate": 2.7203464781201523e-06,
      "loss": 0.77078271,
      "num_input_tokens_seen": 72096490,
      "step": 3337,
      "time_per_iteration": 2.672950267791748
    },
    {
      "auxiliary_loss_clip": 0.01197863,
      "auxiliary_loss_mlp": 0.01035476,
      "balance_loss_clip": 1.06043386,
      "balance_loss_mlp": 1.02654135,
      "epoch": 0.40137076895328566,
      "flos": 24607679541120.0,
      "grad_norm": 2.2124883319239803,
      "language_loss": 0.78147691,
      "learning_rate": 2.719619732195215e-06,
      "loss": 0.8038103,
      "num_input_tokens_seen": 72118130,
      "step": 3338,
      "time_per_iteration": 2.6188247203826904
    },
    {
      "auxiliary_loss_clip": 0.01204264,
      "auxiliary_loss_mlp": 0.01029212,
      "balance_loss_clip": 1.05381966,
      "balance_loss_mlp": 1.02040863,
      "epoch": 0.4014910118439247,
      "flos": 24206593299840.0,
      "grad_norm": 1.5587035555813598,
      "language_loss": 0.72697389,
      "learning_rate": 2.7188928771045377e-06,
      "loss": 0.74930859,
      "num_input_tokens_seen": 72139450,
      "step": 3339,
      "time_per_iteration": 2.7433669567108154
    },
    {
      "auxiliary_loss_clip": 0.01198096,
      "auxiliary_loss_mlp": 0.0102722,
      "balance_loss_clip": 1.05293834,
      "balance_loss_mlp": 1.01839864,
      "epoch": 0.4016112547345638,
      "flos": 26725080418560.0,
      "grad_norm": 2.4845943136433726,
      "language_loss": 0.79787797,
      "learning_rate": 2.7181659129583815e-06,
      "loss": 0.82013118,
      "num_input_tokens_seen": 72159040,
      "step": 3340,
      "time_per_iteration": 2.758986473083496
    },
    {
      "auxiliary_loss_clip": 0.01188994,
      "auxiliary_loss_mlp": 0.01027866,
      "balance_loss_clip": 1.04836333,
      "balance_loss_mlp": 1.01887715,
      "epoch": 0.4017314976252029,
      "flos": 21288025520640.0,
      "grad_norm": 1.7532640519669132,
      "language_loss": 0.76298487,
      "learning_rate": 2.7174388398670276e-06,
      "loss": 0.78515345,
      "num_input_tokens_seen": 72178220,
      "step": 3341,
      "time_per_iteration": 2.652343511581421
    },
    {
      "auxiliary_loss_clip": 0.01197496,
      "auxiliary_loss_mlp": 0.01036154,
      "balance_loss_clip": 1.0559932,
      "balance_loss_mlp": 1.02688551,
      "epoch": 0.401851740515842,
      "flos": 25484690010240.0,
      "grad_norm": 2.350550857255171,
      "language_loss": 0.92361623,
      "learning_rate": 2.716711657940773e-06,
      "loss": 0.94595277,
      "num_input_tokens_seen": 72199230,
      "step": 3342,
      "time_per_iteration": 2.671809434890747
    },
    {
      "auxiliary_loss_clip": 0.01106795,
      "auxiliary_loss_mlp": 0.0100474,
      "balance_loss_clip": 1.0188241,
      "balance_loss_mlp": 1.00358331,
      "epoch": 0.4019719834064811,
      "flos": 55395334978560.0,
      "grad_norm": 1.2739081841954674,
      "language_loss": 0.56456208,
      "learning_rate": 2.7159843672899284e-06,
      "loss": 0.58567739,
      "num_input_tokens_seen": 72263430,
      "step": 3343,
      "time_per_iteration": 3.3798930644989014
    },
    {
      "auxiliary_loss_clip": 0.01201286,
      "auxiliary_loss_mlp": 0.0103089,
      "balance_loss_clip": 1.06000328,
      "balance_loss_mlp": 1.02150846,
      "epoch": 0.40209222629712016,
      "flos": 18180100218240.0,
      "grad_norm": 1.9913860764635738,
      "language_loss": 0.81531179,
      "learning_rate": 2.715256968024825e-06,
      "loss": 0.83763349,
      "num_input_tokens_seen": 72280505,
      "step": 3344,
      "time_per_iteration": 2.8320488929748535
    },
    {
      "auxiliary_loss_clip": 0.01206066,
      "auxiliary_loss_mlp": 0.01028737,
      "balance_loss_clip": 1.05602539,
      "balance_loss_mlp": 1.01954651,
      "epoch": 0.40221246918775927,
      "flos": 25961009287680.0,
      "grad_norm": 1.654541775874132,
      "language_loss": 0.82363641,
      "learning_rate": 2.7145294602558083e-06,
      "loss": 0.84598446,
      "num_input_tokens_seen": 72301215,
      "step": 3345,
      "time_per_iteration": 2.815033435821533
    },
    {
      "auxiliary_loss_clip": 0.011985,
      "auxiliary_loss_mlp": 0.01028305,
      "balance_loss_clip": 1.0568763,
      "balance_loss_mlp": 1.0181427,
      "epoch": 0.4023327120783984,
      "flos": 33838912056960.0,
      "grad_norm": 6.83984558539499,
      "language_loss": 0.70877433,
      "learning_rate": 2.713801844093241e-06,
      "loss": 0.73104239,
      "num_input_tokens_seen": 72322365,
      "step": 3346,
      "time_per_iteration": 2.837744951248169
    },
    {
      "auxiliary_loss_clip": 0.01198755,
      "auxiliary_loss_mlp": 0.01032394,
      "balance_loss_clip": 1.057513,
      "balance_loss_mlp": 1.02348351,
      "epoch": 0.40245295496903744,
      "flos": 26900252069760.0,
      "grad_norm": 2.1659826503740436,
      "language_loss": 0.8838805,
      "learning_rate": 2.7130741196475014e-06,
      "loss": 0.90619206,
      "num_input_tokens_seen": 72340495,
      "step": 3347,
      "time_per_iteration": 2.7062532901763916
    },
    {
      "auxiliary_loss_clip": 0.01204028,
      "auxiliary_loss_mlp": 0.01036767,
      "balance_loss_clip": 1.05847812,
      "balance_loss_mlp": 1.02693236,
      "epoch": 0.40257319785967655,
      "flos": 36902738436480.0,
      "grad_norm": 3.0484094570506177,
      "language_loss": 0.79146838,
      "learning_rate": 2.7123462870289848e-06,
      "loss": 0.81387627,
      "num_input_tokens_seen": 72360545,
      "step": 3348,
      "time_per_iteration": 3.717029571533203
    },
    {
      "auxiliary_loss_clip": 0.01201881,
      "auxiliary_loss_mlp": 0.0103056,
      "balance_loss_clip": 1.05289245,
      "balance_loss_mlp": 1.02096391,
      "epoch": 0.40269344075031566,
      "flos": 24353180703360.0,
      "grad_norm": 3.3054923676444967,
      "language_loss": 0.80969268,
      "learning_rate": 2.711618346348102e-06,
      "loss": 0.83201712,
      "num_input_tokens_seen": 72381070,
      "step": 3349,
      "time_per_iteration": 2.642728328704834
    },
    {
      "auxiliary_loss_clip": 0.01191615,
      "auxiliary_loss_mlp": 0.01034215,
      "balance_loss_clip": 1.05381751,
      "balance_loss_mlp": 1.02506018,
      "epoch": 0.4028136836409547,
      "flos": 14389657614720.0,
      "grad_norm": 1.6282541800475,
      "language_loss": 0.63529724,
      "learning_rate": 2.7108902977152825e-06,
      "loss": 0.65755552,
      "num_input_tokens_seen": 72398970,
      "step": 3350,
      "time_per_iteration": 3.64851713180542
    },
    {
      "auxiliary_loss_clip": 0.01192014,
      "auxiliary_loss_mlp": 0.01030169,
      "balance_loss_clip": 1.05375934,
      "balance_loss_mlp": 1.02101982,
      "epoch": 0.4029339265315938,
      "flos": 26136037284480.0,
      "grad_norm": 2.576464186843232,
      "language_loss": 0.75184751,
      "learning_rate": 2.7101621412409704e-06,
      "loss": 0.77406931,
      "num_input_tokens_seen": 72418455,
      "step": 3351,
      "time_per_iteration": 2.702744722366333
    },
    {
      "auxiliary_loss_clip": 0.01194896,
      "auxiliary_loss_mlp": 0.01036727,
      "balance_loss_clip": 1.05696023,
      "balance_loss_mlp": 1.02734494,
      "epoch": 0.40305416942223293,
      "flos": 23256325042560.0,
      "grad_norm": 2.852844681697008,
      "language_loss": 0.85692775,
      "learning_rate": 2.7094338770356256e-06,
      "loss": 0.87924391,
      "num_input_tokens_seen": 72437540,
      "step": 3352,
      "time_per_iteration": 2.597398519515991
    },
    {
      "auxiliary_loss_clip": 0.01196376,
      "auxiliary_loss_mlp": 0.01035086,
      "balance_loss_clip": 1.05467582,
      "balance_loss_mlp": 1.02618146,
      "epoch": 0.403174412312872,
      "flos": 27089645506560.0,
      "grad_norm": 2.769812254773546,
      "language_loss": 0.64157122,
      "learning_rate": 2.708705505209726e-06,
      "loss": 0.66388583,
      "num_input_tokens_seen": 72458315,
      "step": 3353,
      "time_per_iteration": 3.689796209335327
    },
    {
      "auxiliary_loss_clip": 0.01196139,
      "auxiliary_loss_mlp": 0.01028303,
      "balance_loss_clip": 1.04827929,
      "balance_loss_mlp": 1.01968455,
      "epoch": 0.4032946552035111,
      "flos": 21756336065280.0,
      "grad_norm": 2.0475359537723534,
      "language_loss": 0.91947353,
      "learning_rate": 2.7079770258737646e-06,
      "loss": 0.94171792,
      "num_input_tokens_seen": 72476225,
      "step": 3354,
      "time_per_iteration": 2.8516979217529297
    },
    {
      "auxiliary_loss_clip": 0.01199371,
      "auxiliary_loss_mlp": 0.0103249,
      "balance_loss_clip": 1.05245471,
      "balance_loss_mlp": 1.022506,
      "epoch": 0.4034148980941502,
      "flos": 17343956448000.0,
      "grad_norm": 2.1966337003190546,
      "language_loss": 0.75598776,
      "learning_rate": 2.707248439138251e-06,
      "loss": 0.77830637,
      "num_input_tokens_seen": 72492460,
      "step": 3355,
      "time_per_iteration": 2.6197054386138916
    },
    {
      "auxiliary_loss_clip": 0.0119586,
      "auxiliary_loss_mlp": 0.01035621,
      "balance_loss_clip": 1.05648375,
      "balance_loss_mlp": 1.02691269,
      "epoch": 0.40353514098478926,
      "flos": 22017838055040.0,
      "grad_norm": 1.9813968074206558,
      "language_loss": 0.65400743,
      "learning_rate": 2.7065197451137114e-06,
      "loss": 0.67632222,
      "num_input_tokens_seen": 72513840,
      "step": 3356,
      "time_per_iteration": 2.6935982704162598
    },
    {
      "auxiliary_loss_clip": 0.01200051,
      "auxiliary_loss_mlp": 0.01028387,
      "balance_loss_clip": 1.05586982,
      "balance_loss_mlp": 1.01915479,
      "epoch": 0.4036553838754284,
      "flos": 14246446089600.0,
      "grad_norm": 2.442027196234911,
      "language_loss": 0.67607903,
      "learning_rate": 2.7057909439106894e-06,
      "loss": 0.69836348,
      "num_input_tokens_seen": 72531695,
      "step": 3357,
      "time_per_iteration": 2.6975913047790527
    },
    {
      "auxiliary_loss_clip": 0.01189733,
      "auxiliary_loss_mlp": 0.01209826,
      "balance_loss_clip": 1.05571246,
      "balance_loss_mlp": 1.00027943,
      "epoch": 0.40377562676606743,
      "flos": 24790644443520.0,
      "grad_norm": 2.1244475599913204,
      "language_loss": 0.78479338,
      "learning_rate": 2.7050620356397417e-06,
      "loss": 0.80878901,
      "num_input_tokens_seen": 72550645,
      "step": 3358,
      "time_per_iteration": 3.602893829345703
    },
    {
      "auxiliary_loss_clip": 0.01192908,
      "auxiliary_loss_mlp": 0.01026187,
      "balance_loss_clip": 1.05835009,
      "balance_loss_mlp": 1.01721096,
      "epoch": 0.40389586965670654,
      "flos": 24061226958720.0,
      "grad_norm": 2.072033079297855,
      "language_loss": 0.72441077,
      "learning_rate": 2.7043330204114437e-06,
      "loss": 0.74660176,
      "num_input_tokens_seen": 72569355,
      "step": 3359,
      "time_per_iteration": 2.592270851135254
    },
    {
      "auxiliary_loss_clip": 0.01190017,
      "auxiliary_loss_mlp": 0.01028165,
      "balance_loss_clip": 1.05527735,
      "balance_loss_mlp": 1.01919496,
      "epoch": 0.40401611254734565,
      "flos": 16399613934720.0,
      "grad_norm": 2.8249761318983384,
      "language_loss": 0.85727382,
      "learning_rate": 2.7036038983363862e-06,
      "loss": 0.87945569,
      "num_input_tokens_seen": 72585960,
      "step": 3360,
      "time_per_iteration": 2.6381046772003174
    },
    {
      "auxiliary_loss_clip": 0.01191665,
      "auxiliary_loss_mlp": 0.01027587,
      "balance_loss_clip": 1.056566,
      "balance_loss_mlp": 1.01911116,
      "epoch": 0.4041363554379847,
      "flos": 23988220565760.0,
      "grad_norm": 2.3319103701294073,
      "language_loss": 0.84193212,
      "learning_rate": 2.702874669525177e-06,
      "loss": 0.86412466,
      "num_input_tokens_seen": 72604440,
      "step": 3361,
      "time_per_iteration": 2.6988866329193115
    },
    {
      "auxiliary_loss_clip": 0.01204043,
      "auxiliary_loss_mlp": 0.01029354,
      "balance_loss_clip": 1.05663145,
      "balance_loss_mlp": 1.02009106,
      "epoch": 0.4042565983286238,
      "flos": 28401964899840.0,
      "grad_norm": 1.9423849757577858,
      "language_loss": 0.69798028,
      "learning_rate": 2.7021453340884394e-06,
      "loss": 0.7203142,
      "num_input_tokens_seen": 72622165,
      "step": 3362,
      "time_per_iteration": 2.8168253898620605
    },
    {
      "auxiliary_loss_clip": 0.01190703,
      "auxiliary_loss_mlp": 0.01208957,
      "balance_loss_clip": 1.055686,
      "balance_loss_mlp": 1.00026035,
      "epoch": 0.40437684121926293,
      "flos": 17710963660800.0,
      "grad_norm": 2.0534306736072216,
      "language_loss": 0.7312479,
      "learning_rate": 2.7014158921368125e-06,
      "loss": 0.75524449,
      "num_input_tokens_seen": 72640490,
      "step": 3363,
      "time_per_iteration": 2.6132636070251465
    },
    {
      "auxiliary_loss_clip": 0.01195294,
      "auxiliary_loss_mlp": 0.01033789,
      "balance_loss_clip": 1.05889678,
      "balance_loss_mlp": 1.02435327,
      "epoch": 0.404497084109902,
      "flos": 24018959629440.0,
      "grad_norm": 2.681362226771861,
      "language_loss": 0.85277611,
      "learning_rate": 2.700686343780953e-06,
      "loss": 0.875067,
      "num_input_tokens_seen": 72660360,
      "step": 3364,
      "time_per_iteration": 2.6351799964904785
    },
    {
      "auxiliary_loss_clip": 0.01198139,
      "auxiliary_loss_mlp": 0.01029389,
      "balance_loss_clip": 1.05184007,
      "balance_loss_mlp": 1.02022195,
      "epoch": 0.4046173270005411,
      "flos": 22929861306240.0,
      "grad_norm": 1.7046199639144273,
      "language_loss": 0.88780951,
      "learning_rate": 2.699956689131532e-06,
      "loss": 0.91008478,
      "num_input_tokens_seen": 72680345,
      "step": 3365,
      "time_per_iteration": 2.7329752445220947
    },
    {
      "auxiliary_loss_clip": 0.01202193,
      "auxiliary_loss_mlp": 0.01031695,
      "balance_loss_clip": 1.05591476,
      "balance_loss_mlp": 1.02239656,
      "epoch": 0.4047375698911802,
      "flos": 20668135582080.0,
      "grad_norm": 4.7149900399843485,
      "language_loss": 0.8532027,
      "learning_rate": 2.699226928299238e-06,
      "loss": 0.87554157,
      "num_input_tokens_seen": 72698365,
      "step": 3366,
      "time_per_iteration": 2.643519163131714
    },
    {
      "auxiliary_loss_clip": 0.01200844,
      "auxiliary_loss_mlp": 0.0103185,
      "balance_loss_clip": 1.05792356,
      "balance_loss_mlp": 1.02324867,
      "epoch": 0.40485781278181926,
      "flos": 28912865996160.0,
      "grad_norm": 2.2648423160824276,
      "language_loss": 0.78936815,
      "learning_rate": 2.698497061394774e-06,
      "loss": 0.81169498,
      "num_input_tokens_seen": 72716850,
      "step": 3367,
      "time_per_iteration": 2.772561550140381
    },
    {
      "auxiliary_loss_clip": 0.01208931,
      "auxiliary_loss_mlp": 0.01208968,
      "balance_loss_clip": 1.05593848,
      "balance_loss_mlp": 1.00027156,
      "epoch": 0.40497805567245837,
      "flos": 23148377694720.0,
      "grad_norm": 1.6048684176993966,
      "language_loss": 0.80944163,
      "learning_rate": 2.6977670885288627e-06,
      "loss": 0.83362061,
      "num_input_tokens_seen": 72738250,
      "step": 3368,
      "time_per_iteration": 2.669588327407837
    },
    {
      "auxiliary_loss_clip": 0.01187986,
      "auxiliary_loss_mlp": 0.01031707,
      "balance_loss_clip": 1.05243516,
      "balance_loss_mlp": 1.02268887,
      "epoch": 0.4050982985630975,
      "flos": 16289404030080.0,
      "grad_norm": 1.8523039826727072,
      "language_loss": 0.75614405,
      "learning_rate": 2.6970370098122378e-06,
      "loss": 0.778341,
      "num_input_tokens_seen": 72755235,
      "step": 3369,
      "time_per_iteration": 2.6263012886047363
    },
    {
      "auxiliary_loss_clip": 0.01192679,
      "auxiliary_loss_mlp": 0.01027716,
      "balance_loss_clip": 1.05534363,
      "balance_loss_mlp": 1.01906097,
      "epoch": 0.40521854145373654,
      "flos": 34459484353920.0,
      "grad_norm": 1.5695487199437386,
      "language_loss": 0.86937201,
      "learning_rate": 2.6963068253556535e-06,
      "loss": 0.89157593,
      "num_input_tokens_seen": 72776620,
      "step": 3370,
      "time_per_iteration": 2.677650213241577
    },
    {
      "auxiliary_loss_clip": 0.01205714,
      "auxiliary_loss_mlp": 0.01030502,
      "balance_loss_clip": 1.05718136,
      "balance_loss_mlp": 1.02082253,
      "epoch": 0.40533878434437565,
      "flos": 25331099454720.0,
      "grad_norm": 2.053675958813697,
      "language_loss": 0.8578192,
      "learning_rate": 2.6955765352698763e-06,
      "loss": 0.88018137,
      "num_input_tokens_seen": 72796765,
      "step": 3371,
      "time_per_iteration": 2.6432478427886963
    },
    {
      "auxiliary_loss_clip": 0.01197418,
      "auxiliary_loss_mlp": 0.01029104,
      "balance_loss_clip": 1.05756736,
      "balance_loss_mlp": 1.01954317,
      "epoch": 0.40545902723501476,
      "flos": 15012061505280.0,
      "grad_norm": 2.1911764777801768,
      "language_loss": 0.73157644,
      "learning_rate": 2.6948461396656923e-06,
      "loss": 0.75384164,
      "num_input_tokens_seen": 72814175,
      "step": 3372,
      "time_per_iteration": 2.534284830093384
    },
    {
      "auxiliary_loss_clip": 0.01203729,
      "auxiliary_loss_mlp": 0.01029944,
      "balance_loss_clip": 1.05790615,
      "balance_loss_mlp": 1.02044261,
      "epoch": 0.4055792701256538,
      "flos": 25521103422720.0,
      "grad_norm": 2.298160120316316,
      "language_loss": 0.74606419,
      "learning_rate": 2.6941156386539013e-06,
      "loss": 0.76840091,
      "num_input_tokens_seen": 72834125,
      "step": 3373,
      "time_per_iteration": 2.664430856704712
    },
    {
      "auxiliary_loss_clip": 0.01194956,
      "auxiliary_loss_mlp": 0.010334,
      "balance_loss_clip": 1.05598855,
      "balance_loss_mlp": 1.0241766,
      "epoch": 0.4056995130162929,
      "flos": 19574583972480.0,
      "grad_norm": 2.6078730754086767,
      "language_loss": 0.81378877,
      "learning_rate": 2.6933850323453203e-06,
      "loss": 0.83607227,
      "num_input_tokens_seen": 72852570,
      "step": 3374,
      "time_per_iteration": 2.6686503887176514
    },
    {
      "auxiliary_loss_clip": 0.01195264,
      "auxiliary_loss_mlp": 0.01030977,
      "balance_loss_clip": 1.060377,
      "balance_loss_mlp": 1.02174413,
      "epoch": 0.405819755906932,
      "flos": 15413794191360.0,
      "grad_norm": 3.067985875390008,
      "language_loss": 0.7443521,
      "learning_rate": 2.6926543208507806e-06,
      "loss": 0.76661444,
      "num_input_tokens_seen": 72871250,
      "step": 3375,
      "time_per_iteration": 3.6374433040618896
    },
    {
      "auxiliary_loss_clip": 0.01196156,
      "auxiliary_loss_mlp": 0.01029644,
      "balance_loss_clip": 1.0562458,
      "balance_loss_mlp": 1.02027977,
      "epoch": 0.4059399987975711,
      "flos": 21433930565760.0,
      "grad_norm": 3.944955304829219,
      "language_loss": 0.80224001,
      "learning_rate": 2.6919235042811316e-06,
      "loss": 0.82449806,
      "num_input_tokens_seen": 72890035,
      "step": 3376,
      "time_per_iteration": 2.6430611610412598
    },
    {
      "auxiliary_loss_clip": 0.01196697,
      "auxiliary_loss_mlp": 0.01033639,
      "balance_loss_clip": 1.05248737,
      "balance_loss_mlp": 1.0237987,
      "epoch": 0.4060602416882102,
      "flos": 25556942217600.0,
      "grad_norm": 3.5267838619788576,
      "language_loss": 0.76209289,
      "learning_rate": 2.691192582747237e-06,
      "loss": 0.78439623,
      "num_input_tokens_seen": 72909665,
      "step": 3377,
      "time_per_iteration": 3.595212936401367
    },
    {
      "auxiliary_loss_clip": 0.01198987,
      "auxiliary_loss_mlp": 0.01026022,
      "balance_loss_clip": 1.06063461,
      "balance_loss_mlp": 1.01752257,
      "epoch": 0.40618048457884925,
      "flos": 23766759262080.0,
      "grad_norm": 1.8109379737443745,
      "language_loss": 0.73657668,
      "learning_rate": 2.6904615563599765e-06,
      "loss": 0.75882679,
      "num_input_tokens_seen": 72929465,
      "step": 3378,
      "time_per_iteration": 2.5984063148498535
    },
    {
      "auxiliary_loss_clip": 0.01191859,
      "auxiliary_loss_mlp": 0.01025494,
      "balance_loss_clip": 1.04966068,
      "balance_loss_mlp": 1.01666629,
      "epoch": 0.40630072746948837,
      "flos": 17639681120640.0,
      "grad_norm": 1.9854973831307734,
      "language_loss": 0.83537978,
      "learning_rate": 2.6897304252302477e-06,
      "loss": 0.85755324,
      "num_input_tokens_seen": 72946785,
      "step": 3379,
      "time_per_iteration": 2.6275582313537598
    },
    {
      "auxiliary_loss_clip": 0.01102621,
      "auxiliary_loss_mlp": 0.0100839,
      "balance_loss_clip": 1.01849735,
      "balance_loss_mlp": 1.00709677,
      "epoch": 0.4064209703601275,
      "flos": 60836053063680.0,
      "grad_norm": 0.7954356118259155,
      "language_loss": 0.54785717,
      "learning_rate": 2.688999189468962e-06,
      "loss": 0.56896722,
      "num_input_tokens_seen": 73003215,
      "step": 3380,
      "time_per_iteration": 3.9492247104644775
    },
    {
      "auxiliary_loss_clip": 0.01194209,
      "auxiliary_loss_mlp": 0.01031292,
      "balance_loss_clip": 1.05583441,
      "balance_loss_mlp": 1.02281666,
      "epoch": 0.40654121325076653,
      "flos": 24024346669440.0,
      "grad_norm": 2.6944976001768723,
      "language_loss": 0.7668429,
      "learning_rate": 2.6882678491870464e-06,
      "loss": 0.78909791,
      "num_input_tokens_seen": 73023650,
      "step": 3381,
      "time_per_iteration": 2.648484468460083
    },
    {
      "auxiliary_loss_clip": 0.01200331,
      "auxiliary_loss_mlp": 0.01024337,
      "balance_loss_clip": 1.05773354,
      "balance_loss_mlp": 1.01444268,
      "epoch": 0.40666145614140564,
      "flos": 27344252085120.0,
      "grad_norm": 1.756406908577883,
      "language_loss": 0.71604127,
      "learning_rate": 2.6875364044954453e-06,
      "loss": 0.73828799,
      "num_input_tokens_seen": 73043880,
      "step": 3382,
      "time_per_iteration": 2.6182563304901123
    },
    {
      "auxiliary_loss_clip": 0.01193988,
      "auxiliary_loss_mlp": 0.01028751,
      "balance_loss_clip": 1.04965842,
      "balance_loss_mlp": 1.02004266,
      "epoch": 0.40678169903204475,
      "flos": 26176724415360.0,
      "grad_norm": 1.5774693221661764,
      "language_loss": 0.8257612,
      "learning_rate": 2.6868048555051185e-06,
      "loss": 0.84798861,
      "num_input_tokens_seen": 73065410,
      "step": 3383,
      "time_per_iteration": 2.644859552383423
    },
    {
      "auxiliary_loss_clip": 0.01204143,
      "auxiliary_loss_mlp": 0.0103051,
      "balance_loss_clip": 1.05250716,
      "balance_loss_mlp": 1.02149212,
      "epoch": 0.4069019419226838,
      "flos": 28622420622720.0,
      "grad_norm": 2.4583042450510324,
      "language_loss": 0.85809469,
      "learning_rate": 2.686073202327041e-06,
      "loss": 0.88044119,
      "num_input_tokens_seen": 73084410,
      "step": 3384,
      "time_per_iteration": 2.67272686958313
    },
    {
      "auxiliary_loss_clip": 0.01187024,
      "auxiliary_loss_mlp": 0.01038988,
      "balance_loss_clip": 1.05020404,
      "balance_loss_mlp": 1.0298388,
      "epoch": 0.4070221848133229,
      "flos": 25229006023680.0,
      "grad_norm": 2.5968391583661568,
      "language_loss": 0.73367471,
      "learning_rate": 2.6853414450722043e-06,
      "loss": 0.75593483,
      "num_input_tokens_seen": 73104075,
      "step": 3385,
      "time_per_iteration": 3.640049457550049
    },
    {
      "auxiliary_loss_clip": 0.01192285,
      "auxiliary_loss_mlp": 0.01027904,
      "balance_loss_clip": 1.05389214,
      "balance_loss_mlp": 1.01870072,
      "epoch": 0.40714242770396203,
      "flos": 18405224709120.0,
      "grad_norm": 2.1799627314526084,
      "language_loss": 0.855995,
      "learning_rate": 2.684609583851616e-06,
      "loss": 0.87819684,
      "num_input_tokens_seen": 73122250,
      "step": 3386,
      "time_per_iteration": 2.6717112064361572
    },
    {
      "auxiliary_loss_clip": 0.01203227,
      "auxiliary_loss_mlp": 0.01030288,
      "balance_loss_clip": 1.0506978,
      "balance_loss_mlp": 1.02136493,
      "epoch": 0.4072626705946011,
      "flos": 30228920403840.0,
      "grad_norm": 1.877912385500874,
      "language_loss": 0.80679649,
      "learning_rate": 2.683877618776297e-06,
      "loss": 0.8291316,
      "num_input_tokens_seen": 73144505,
      "step": 3387,
      "time_per_iteration": 2.78218150138855
    },
    {
      "auxiliary_loss_clip": 0.01191362,
      "auxiliary_loss_mlp": 0.01031936,
      "balance_loss_clip": 1.05044007,
      "balance_loss_mlp": 1.02245855,
      "epoch": 0.4073829134852402,
      "flos": 21834549930240.0,
      "grad_norm": 2.4940619320453923,
      "language_loss": 0.74542195,
      "learning_rate": 2.6831455499572876e-06,
      "loss": 0.7676549,
      "num_input_tokens_seen": 73162440,
      "step": 3388,
      "time_per_iteration": 2.7073872089385986
    },
    {
      "auxiliary_loss_clip": 0.01195971,
      "auxiliary_loss_mlp": 0.0102505,
      "balance_loss_clip": 1.05691135,
      "balance_loss_mlp": 1.01582301,
      "epoch": 0.40750315637587925,
      "flos": 25260211964160.0,
      "grad_norm": 2.9171128580961034,
      "language_loss": 0.77789354,
      "learning_rate": 2.682413377505641e-06,
      "loss": 0.80010378,
      "num_input_tokens_seen": 73181245,
      "step": 3389,
      "time_per_iteration": 2.5966055393218994
    },
    {
      "auxiliary_loss_clip": 0.01195187,
      "auxiliary_loss_mlp": 0.01029707,
      "balance_loss_clip": 1.05387688,
      "balance_loss_mlp": 1.02061164,
      "epoch": 0.40762339926651836,
      "flos": 19712767593600.0,
      "grad_norm": 1.9762115890528742,
      "language_loss": 0.76724434,
      "learning_rate": 2.6816811015324284e-06,
      "loss": 0.7894932,
      "num_input_tokens_seen": 73199295,
      "step": 3390,
      "time_per_iteration": 2.630842447280884
    },
    {
      "auxiliary_loss_clip": 0.01085946,
      "auxiliary_loss_mlp": 0.01005511,
      "balance_loss_clip": 1.01804519,
      "balance_loss_mlp": 1.00437856,
      "epoch": 0.40774364215715747,
      "flos": 71449307314560.0,
      "grad_norm": 0.7244318804993606,
      "language_loss": 0.56692386,
      "learning_rate": 2.6809487221487343e-06,
      "loss": 0.58783841,
      "num_input_tokens_seen": 73258780,
      "step": 3391,
      "time_per_iteration": 3.0598268508911133
    },
    {
      "auxiliary_loss_clip": 0.01186696,
      "auxiliary_loss_mlp": 0.01025674,
      "balance_loss_clip": 1.05253339,
      "balance_loss_mlp": 1.01661408,
      "epoch": 0.4078638850477965,
      "flos": 15084134144640.0,
      "grad_norm": 2.6416632155708353,
      "language_loss": 0.8193512,
      "learning_rate": 2.6802162394656605e-06,
      "loss": 0.84147489,
      "num_input_tokens_seen": 73275490,
      "step": 3392,
      "time_per_iteration": 2.591517686843872
    },
    {
      "auxiliary_loss_clip": 0.01194411,
      "auxiliary_loss_mlp": 0.01034123,
      "balance_loss_clip": 1.05034423,
      "balance_loss_mlp": 1.02552795,
      "epoch": 0.40798412793843564,
      "flos": 23842890138240.0,
      "grad_norm": 1.7532084841581217,
      "language_loss": 0.71809709,
      "learning_rate": 2.679483653594324e-06,
      "loss": 0.74038243,
      "num_input_tokens_seen": 73297260,
      "step": 3393,
      "time_per_iteration": 2.6923108100891113
    },
    {
      "auxiliary_loss_clip": 0.01198658,
      "auxiliary_loss_mlp": 0.01032129,
      "balance_loss_clip": 1.05626833,
      "balance_loss_mlp": 1.02311718,
      "epoch": 0.40810437082907475,
      "flos": 21065774117760.0,
      "grad_norm": 2.8631472076866924,
      "language_loss": 0.7642417,
      "learning_rate": 2.678750964645857e-06,
      "loss": 0.78654957,
      "num_input_tokens_seen": 73316340,
      "step": 3394,
      "time_per_iteration": 2.685669183731079
    },
    {
      "auxiliary_loss_clip": 0.01198588,
      "auxiliary_loss_mlp": 0.01030223,
      "balance_loss_clip": 1.05862534,
      "balance_loss_mlp": 1.02121639,
      "epoch": 0.4082246137197138,
      "flos": 11321377948800.0,
      "grad_norm": 2.4850327182753196,
      "language_loss": 0.83865595,
      "learning_rate": 2.6780181727314094e-06,
      "loss": 0.86094409,
      "num_input_tokens_seen": 73331245,
      "step": 3395,
      "time_per_iteration": 2.767573118209839
    },
    {
      "auxiliary_loss_clip": 0.01204188,
      "auxiliary_loss_mlp": 0.012086,
      "balance_loss_clip": 1.05255628,
      "balance_loss_mlp": 1.00032282,
      "epoch": 0.4083448566103529,
      "flos": 19062569554560.0,
      "grad_norm": 1.7161099332394851,
      "language_loss": 0.77805316,
      "learning_rate": 2.6772852779621435e-06,
      "loss": 0.80218101,
      "num_input_tokens_seen": 73349105,
      "step": 3396,
      "time_per_iteration": 2.6720526218414307
    },
    {
      "auxiliary_loss_clip": 0.01190114,
      "auxiliary_loss_mlp": 0.01208424,
      "balance_loss_clip": 1.05717397,
      "balance_loss_mlp": 1.00032699,
      "epoch": 0.408465099500992,
      "flos": 23550254035200.0,
      "grad_norm": 1.9278199985472284,
      "language_loss": 0.86794007,
      "learning_rate": 2.676552280449239e-06,
      "loss": 0.89192551,
      "num_input_tokens_seen": 73368990,
      "step": 3397,
      "time_per_iteration": 2.6216986179351807
    },
    {
      "auxiliary_loss_clip": 0.01186616,
      "auxiliary_loss_mlp": 0.01035642,
      "balance_loss_clip": 1.05297184,
      "balance_loss_mlp": 1.02605736,
      "epoch": 0.4085853423916311,
      "flos": 12750012558720.0,
      "grad_norm": 2.3677480047207324,
      "language_loss": 0.75776696,
      "learning_rate": 2.6758191803038917e-06,
      "loss": 0.77998954,
      "num_input_tokens_seen": 73387485,
      "step": 3398,
      "time_per_iteration": 2.6048288345336914
    },
    {
      "auxiliary_loss_clip": 0.01193946,
      "auxiliary_loss_mlp": 0.01028116,
      "balance_loss_clip": 1.0484302,
      "balance_loss_mlp": 1.0190742,
      "epoch": 0.4087055852822702,
      "flos": 24353072962560.0,
      "grad_norm": 2.0891592284695224,
      "language_loss": 0.82817149,
      "learning_rate": 2.6750859776373125e-06,
      "loss": 0.8503921,
      "num_input_tokens_seen": 73406940,
      "step": 3399,
      "time_per_iteration": 2.765003204345703
    },
    {
      "auxiliary_loss_clip": 0.01120793,
      "auxiliary_loss_mlp": 0.01001186,
      "balance_loss_clip": 1.01844263,
      "balance_loss_mlp": 0.99996978,
      "epoch": 0.4088258281729093,
      "flos": 66387950720640.0,
      "grad_norm": 0.7696323922308828,
      "language_loss": 0.60341775,
      "learning_rate": 2.674352672560727e-06,
      "loss": 0.62463754,
      "num_input_tokens_seen": 73468385,
      "step": 3400,
      "time_per_iteration": 3.3171393871307373
    },
    {
      "auxiliary_loss_clip": 0.01197589,
      "auxiliary_loss_mlp": 0.01031589,
      "balance_loss_clip": 1.04940104,
      "balance_loss_mlp": 1.02214193,
      "epoch": 0.40894607106354836,
      "flos": 20449260057600.0,
      "grad_norm": 1.8148205968495579,
      "language_loss": 0.76969206,
      "learning_rate": 2.673619265185377e-06,
      "loss": 0.79198384,
      "num_input_tokens_seen": 73488225,
      "step": 3401,
      "time_per_iteration": 2.896512985229492
    },
    {
      "auxiliary_loss_clip": 0.0119774,
      "auxiliary_loss_mlp": 0.01031891,
      "balance_loss_clip": 1.05507612,
      "balance_loss_mlp": 1.02278388,
      "epoch": 0.40906631395418747,
      "flos": 27053627143680.0,
      "grad_norm": 1.641855462739808,
      "language_loss": 0.78164309,
      "learning_rate": 2.672885755622521e-06,
      "loss": 0.8039394,
      "num_input_tokens_seen": 73510640,
      "step": 3402,
      "time_per_iteration": 3.5770065784454346
    },
    {
      "auxiliary_loss_clip": 0.0120012,
      "auxiliary_loss_mlp": 0.01032314,
      "balance_loss_clip": 1.04736137,
      "balance_loss_mlp": 1.02335548,
      "epoch": 0.4091865568448266,
      "flos": 25484151306240.0,
      "grad_norm": 2.694872232698398,
      "language_loss": 0.70554483,
      "learning_rate": 2.67215214398343e-06,
      "loss": 0.72786921,
      "num_input_tokens_seen": 73530655,
      "step": 3403,
      "time_per_iteration": 2.674936532974243
    },
    {
      "auxiliary_loss_clip": 0.01204397,
      "auxiliary_loss_mlp": 0.0103313,
      "balance_loss_clip": 1.04753804,
      "balance_loss_mlp": 1.02382541,
      "epoch": 0.40930679973546563,
      "flos": 28657864368000.0,
      "grad_norm": 2.0681155850079653,
      "language_loss": 0.78351742,
      "learning_rate": 2.671418430379393e-06,
      "loss": 0.80589271,
      "num_input_tokens_seen": 73549340,
      "step": 3404,
      "time_per_iteration": 3.6221978664398193
    },
    {
      "auxiliary_loss_clip": 0.01192956,
      "auxiliary_loss_mlp": 0.01025805,
      "balance_loss_clip": 1.05620337,
      "balance_loss_mlp": 1.01688766,
      "epoch": 0.40942704262610474,
      "flos": 20886292834560.0,
      "grad_norm": 1.8575884435521115,
      "language_loss": 0.83418953,
      "learning_rate": 2.670684614921715e-06,
      "loss": 0.85637718,
      "num_input_tokens_seen": 73568315,
      "step": 3405,
      "time_per_iteration": 2.541703939437866
    },
    {
      "auxiliary_loss_clip": 0.01199288,
      "auxiliary_loss_mlp": 0.0103171,
      "balance_loss_clip": 1.05249929,
      "balance_loss_mlp": 1.02253723,
      "epoch": 0.4095472855167438,
      "flos": 21618080616960.0,
      "grad_norm": 2.880179040701166,
      "language_loss": 0.69433272,
      "learning_rate": 2.6699506977217128e-06,
      "loss": 0.71664274,
      "num_input_tokens_seen": 73588490,
      "step": 3406,
      "time_per_iteration": 2.664883852005005
    },
    {
      "auxiliary_loss_clip": 0.01192453,
      "auxiliary_loss_mlp": 0.01030062,
      "balance_loss_clip": 1.0576601,
      "balance_loss_mlp": 1.02156222,
      "epoch": 0.4096675284073829,
      "flos": 27926112499200.0,
      "grad_norm": 2.0648290431189555,
      "language_loss": 0.69772661,
      "learning_rate": 2.6692166788907233e-06,
      "loss": 0.71995175,
      "num_input_tokens_seen": 73608685,
      "step": 3407,
      "time_per_iteration": 3.7711944580078125
    },
    {
      "auxiliary_loss_clip": 0.01197438,
      "auxiliary_loss_mlp": 0.01035879,
      "balance_loss_clip": 1.05109775,
      "balance_loss_mlp": 1.02654469,
      "epoch": 0.409787771298022,
      "flos": 19206607092480.0,
      "grad_norm": 1.9746399032908264,
      "language_loss": 0.76836628,
      "learning_rate": 2.6684825585400957e-06,
      "loss": 0.79069948,
      "num_input_tokens_seen": 73627630,
      "step": 3408,
      "time_per_iteration": 2.639327049255371
    },
    {
      "auxiliary_loss_clip": 0.01092767,
      "auxiliary_loss_mlp": 0.01003209,
      "balance_loss_clip": 1.0177871,
      "balance_loss_mlp": 1.00199926,
      "epoch": 0.4099080141886611,
      "flos": 59269234832640.0,
      "grad_norm": 0.8163810897204758,
      "language_loss": 0.65083432,
      "learning_rate": 2.6677483367811947e-06,
      "loss": 0.67179412,
      "num_input_tokens_seen": 73687670,
      "step": 3409,
      "time_per_iteration": 3.2842841148376465
    },
    {
      "auxiliary_loss_clip": 0.01197878,
      "auxiliary_loss_mlp": 0.01024727,
      "balance_loss_clip": 1.05395937,
      "balance_loss_mlp": 1.01657271,
      "epoch": 0.4100282570793002,
      "flos": 21906443001600.0,
      "grad_norm": 2.4881972107169656,
      "language_loss": 0.75404143,
      "learning_rate": 2.6670140137254028e-06,
      "loss": 0.77626753,
      "num_input_tokens_seen": 73707145,
      "step": 3410,
      "time_per_iteration": 2.5809874534606934
    },
    {
      "auxiliary_loss_clip": 0.01195852,
      "auxiliary_loss_mlp": 0.01030609,
      "balance_loss_clip": 1.04968452,
      "balance_loss_mlp": 1.02175808,
      "epoch": 0.4101484999699393,
      "flos": 18551596631040.0,
      "grad_norm": 2.405487655799179,
      "language_loss": 0.89592981,
      "learning_rate": 2.666279589484115e-06,
      "loss": 0.91819453,
      "num_input_tokens_seen": 73725045,
      "step": 3411,
      "time_per_iteration": 3.5976433753967285
    },
    {
      "auxiliary_loss_clip": 0.01200594,
      "auxiliary_loss_mlp": 0.01024991,
      "balance_loss_clip": 1.0487771,
      "balance_loss_mlp": 1.01665258,
      "epoch": 0.41026874286057835,
      "flos": 19094529680640.0,
      "grad_norm": 1.9486382718283926,
      "language_loss": 0.81004226,
      "learning_rate": 2.6655450641687435e-06,
      "loss": 0.83229816,
      "num_input_tokens_seen": 73742610,
      "step": 3412,
      "time_per_iteration": 2.6602911949157715
    },
    {
      "auxiliary_loss_clip": 0.01195108,
      "auxiliary_loss_mlp": 0.01031229,
      "balance_loss_clip": 1.06068993,
      "balance_loss_mlp": 1.02188921,
      "epoch": 0.41038898575121746,
      "flos": 31209568588800.0,
      "grad_norm": 1.717984366864555,
      "language_loss": 0.69190931,
      "learning_rate": 2.664810437890715e-06,
      "loss": 0.71417266,
      "num_input_tokens_seen": 73764280,
      "step": 3413,
      "time_per_iteration": 2.6404097080230713
    },
    {
      "auxiliary_loss_clip": 0.01195816,
      "auxiliary_loss_mlp": 0.01026376,
      "balance_loss_clip": 1.05242562,
      "balance_loss_mlp": 1.01826954,
      "epoch": 0.41050922864185657,
      "flos": 14355865895040.0,
      "grad_norm": 1.892789790193408,
      "language_loss": 0.79624641,
      "learning_rate": 2.6640757107614714e-06,
      "loss": 0.81846833,
      "num_input_tokens_seen": 73782375,
      "step": 3414,
      "time_per_iteration": 2.6452367305755615
    },
    {
      "auxiliary_loss_clip": 0.0119093,
      "auxiliary_loss_mlp": 0.01025509,
      "balance_loss_clip": 1.04947138,
      "balance_loss_mlp": 1.01638985,
      "epoch": 0.4106294715324956,
      "flos": 30956290813440.0,
      "grad_norm": 2.4288238750215414,
      "language_loss": 0.69267404,
      "learning_rate": 2.6633408828924697e-06,
      "loss": 0.71483845,
      "num_input_tokens_seen": 73801240,
      "step": 3415,
      "time_per_iteration": 2.686812162399292
    },
    {
      "auxiliary_loss_clip": 0.01206202,
      "auxiliary_loss_mlp": 0.010337,
      "balance_loss_clip": 1.05330539,
      "balance_loss_mlp": 1.02518892,
      "epoch": 0.41074971442313474,
      "flos": 24457321209600.0,
      "grad_norm": 1.6401304990675882,
      "language_loss": 0.70083672,
      "learning_rate": 2.662605954395185e-06,
      "loss": 0.72323573,
      "num_input_tokens_seen": 73821200,
      "step": 3416,
      "time_per_iteration": 2.6825919151306152
    },
    {
      "auxiliary_loss_clip": 0.01196453,
      "auxiliary_loss_mlp": 0.01027014,
      "balance_loss_clip": 1.05443609,
      "balance_loss_mlp": 1.01848447,
      "epoch": 0.41086995731377385,
      "flos": 21542991235200.0,
      "grad_norm": 1.8106082703985427,
      "language_loss": 0.83833236,
      "learning_rate": 2.6618709253811027e-06,
      "loss": 0.86056703,
      "num_input_tokens_seen": 73840655,
      "step": 3417,
      "time_per_iteration": 2.6147797107696533
    },
    {
      "auxiliary_loss_clip": 0.01188514,
      "auxiliary_loss_mlp": 0.01024848,
      "balance_loss_clip": 1.05697727,
      "balance_loss_mlp": 1.01720655,
      "epoch": 0.4109902002044129,
      "flos": 20702753314560.0,
      "grad_norm": 1.6297074386213528,
      "language_loss": 0.87729585,
      "learning_rate": 2.6611357959617277e-06,
      "loss": 0.89942944,
      "num_input_tokens_seen": 73860275,
      "step": 3418,
      "time_per_iteration": 2.6254770755767822
    },
    {
      "auxiliary_loss_clip": 0.01191434,
      "auxiliary_loss_mlp": 0.01035807,
      "balance_loss_clip": 1.04945755,
      "balance_loss_mlp": 1.02658033,
      "epoch": 0.411110443095052,
      "flos": 18179992477440.0,
      "grad_norm": 2.0718454945359013,
      "language_loss": 0.91056001,
      "learning_rate": 2.660400566248578e-06,
      "loss": 0.93283236,
      "num_input_tokens_seen": 73878400,
      "step": 3419,
      "time_per_iteration": 2.5983312129974365
    },
    {
      "auxiliary_loss_clip": 0.0119818,
      "auxiliary_loss_mlp": 0.01033132,
      "balance_loss_clip": 1.05039072,
      "balance_loss_mlp": 1.02363062,
      "epoch": 0.41123068598569107,
      "flos": 14575244209920.0,
      "grad_norm": 2.573417883747503,
      "language_loss": 0.66640323,
      "learning_rate": 2.6596652363531876e-06,
      "loss": 0.68871635,
      "num_input_tokens_seen": 73894275,
      "step": 3420,
      "time_per_iteration": 2.6349916458129883
    },
    {
      "auxiliary_loss_clip": 0.01190923,
      "auxiliary_loss_mlp": 0.0102443,
      "balance_loss_clip": 1.0563916,
      "balance_loss_mlp": 1.0157218,
      "epoch": 0.4113509288763302,
      "flos": 21177995184000.0,
      "grad_norm": 1.811254247415306,
      "language_loss": 0.78143203,
      "learning_rate": 2.6589298063871055e-06,
      "loss": 0.80358553,
      "num_input_tokens_seen": 73914450,
      "step": 3421,
      "time_per_iteration": 2.617804527282715
    },
    {
      "auxiliary_loss_clip": 0.01188151,
      "auxiliary_loss_mlp": 0.01028351,
      "balance_loss_clip": 1.05443549,
      "balance_loss_mlp": 1.01920795,
      "epoch": 0.4114711717669693,
      "flos": 18442212739200.0,
      "grad_norm": 4.238983855849647,
      "language_loss": 0.69949466,
      "learning_rate": 2.658194276461895e-06,
      "loss": 0.72165966,
      "num_input_tokens_seen": 73932375,
      "step": 3422,
      "time_per_iteration": 2.604583740234375
    },
    {
      "auxiliary_loss_clip": 0.01197365,
      "auxiliary_loss_mlp": 0.0102772,
      "balance_loss_clip": 1.05037272,
      "balance_loss_mlp": 1.01845121,
      "epoch": 0.41159141465760835,
      "flos": 27233395735680.0,
      "grad_norm": 2.3597936756204185,
      "language_loss": 0.66998792,
      "learning_rate": 2.6574586466891368e-06,
      "loss": 0.69223869,
      "num_input_tokens_seen": 73952850,
      "step": 3423,
      "time_per_iteration": 2.7129364013671875
    },
    {
      "auxiliary_loss_clip": 0.01194663,
      "auxiliary_loss_mlp": 0.01208329,
      "balance_loss_clip": 1.05229759,
      "balance_loss_mlp": 1.00038886,
      "epoch": 0.41171165754824746,
      "flos": 20006876154240.0,
      "grad_norm": 1.9926026627116922,
      "language_loss": 0.64641595,
      "learning_rate": 2.6567229171804247e-06,
      "loss": 0.67044586,
      "num_input_tokens_seen": 73970735,
      "step": 3424,
      "time_per_iteration": 2.6739437580108643
    },
    {
      "auxiliary_loss_clip": 0.0120436,
      "auxiliary_loss_mlp": 0.01038138,
      "balance_loss_clip": 1.04816699,
      "balance_loss_mlp": 1.02857149,
      "epoch": 0.41183190043888657,
      "flos": 18004318035840.0,
      "grad_norm": 3.496857948300498,
      "language_loss": 0.87738359,
      "learning_rate": 2.655987088047368e-06,
      "loss": 0.89980853,
      "num_input_tokens_seen": 73989080,
      "step": 3425,
      "time_per_iteration": 2.6571998596191406
    },
    {
      "auxiliary_loss_clip": 0.01189936,
      "auxiliary_loss_mlp": 0.01030769,
      "balance_loss_clip": 1.05108321,
      "balance_loss_mlp": 1.0214591,
      "epoch": 0.4119521433295256,
      "flos": 27163370171520.0,
      "grad_norm": 2.1496669263326464,
      "language_loss": 0.78843921,
      "learning_rate": 2.6552511594015912e-06,
      "loss": 0.81064624,
      "num_input_tokens_seen": 74009470,
      "step": 3426,
      "time_per_iteration": 2.7572953701019287
    },
    {
      "auxiliary_loss_clip": 0.01193773,
      "auxiliary_loss_mlp": 0.01026812,
      "balance_loss_clip": 1.04968357,
      "balance_loss_mlp": 1.01779366,
      "epoch": 0.41207238622016473,
      "flos": 15122020014720.0,
      "grad_norm": 2.743448458873937,
      "language_loss": 0.85521877,
      "learning_rate": 2.654515131354735e-06,
      "loss": 0.87742466,
      "num_input_tokens_seen": 74027735,
      "step": 3427,
      "time_per_iteration": 2.6685492992401123
    },
    {
      "auxiliary_loss_clip": 0.01195759,
      "auxiliary_loss_mlp": 0.0102546,
      "balance_loss_clip": 1.0507021,
      "balance_loss_mlp": 1.01741958,
      "epoch": 0.41219262911080384,
      "flos": 27052872958080.0,
      "grad_norm": 2.055019046309232,
      "language_loss": 0.8556568,
      "learning_rate": 2.653779004018453e-06,
      "loss": 0.87786895,
      "num_input_tokens_seen": 74048300,
      "step": 3428,
      "time_per_iteration": 2.7053065299987793
    },
    {
      "auxiliary_loss_clip": 0.01188801,
      "auxiliary_loss_mlp": 0.01022473,
      "balance_loss_clip": 1.05296791,
      "balance_loss_mlp": 1.01375234,
      "epoch": 0.4123128720014429,
      "flos": 24686360282880.0,
      "grad_norm": 1.840953515819826,
      "language_loss": 0.8239224,
      "learning_rate": 2.653042777504417e-06,
      "loss": 0.84603512,
      "num_input_tokens_seen": 74070890,
      "step": 3429,
      "time_per_iteration": 3.6515583992004395
    },
    {
      "auxiliary_loss_clip": 0.01202268,
      "auxiliary_loss_mlp": 0.01026178,
      "balance_loss_clip": 1.05278623,
      "balance_loss_mlp": 1.01718354,
      "epoch": 0.412433114892082,
      "flos": 26244774731520.0,
      "grad_norm": 1.8460611154739155,
      "language_loss": 0.79897529,
      "learning_rate": 2.6523064519243105e-06,
      "loss": 0.82125974,
      "num_input_tokens_seen": 74090460,
      "step": 3430,
      "time_per_iteration": 2.660224676132202
    },
    {
      "auxiliary_loss_clip": 0.01194558,
      "auxiliary_loss_mlp": 0.01034447,
      "balance_loss_clip": 1.05544829,
      "balance_loss_mlp": 1.02494574,
      "epoch": 0.4125533577827211,
      "flos": 21361031913600.0,
      "grad_norm": 2.4826251267663366,
      "language_loss": 0.78785276,
      "learning_rate": 2.6515700273898333e-06,
      "loss": 0.81014287,
      "num_input_tokens_seen": 74108335,
      "step": 3431,
      "time_per_iteration": 3.5250039100646973
    },
    {
      "auxiliary_loss_clip": 0.01183162,
      "auxiliary_loss_mlp": 0.01031073,
      "balance_loss_clip": 1.05189419,
      "balance_loss_mlp": 1.02157211,
      "epoch": 0.4126736006733602,
      "flos": 26067556005120.0,
      "grad_norm": 3.5902110367313287,
      "language_loss": 0.68666959,
      "learning_rate": 2.6508335040127018e-06,
      "loss": 0.70881188,
      "num_input_tokens_seen": 74128030,
      "step": 3432,
      "time_per_iteration": 2.712006092071533
    },
    {
      "auxiliary_loss_clip": 0.01198034,
      "auxiliary_loss_mlp": 0.01030179,
      "balance_loss_clip": 1.05721498,
      "balance_loss_mlp": 1.0216738,
      "epoch": 0.4127938435639993,
      "flos": 25666146541440.0,
      "grad_norm": 2.160347907201746,
      "language_loss": 0.77007693,
      "learning_rate": 2.6500968819046446e-06,
      "loss": 0.79235911,
      "num_input_tokens_seen": 74148330,
      "step": 3433,
      "time_per_iteration": 2.627690315246582
    },
    {
      "auxiliary_loss_clip": 0.01184881,
      "auxiliary_loss_mlp": 0.01032682,
      "balance_loss_clip": 1.04684854,
      "balance_loss_mlp": 1.02369976,
      "epoch": 0.4129140864546384,
      "flos": 17995914253440.0,
      "grad_norm": 7.206985917356712,
      "language_loss": 0.58782089,
      "learning_rate": 2.649360161177408e-06,
      "loss": 0.60999644,
      "num_input_tokens_seen": 74163390,
      "step": 3434,
      "time_per_iteration": 3.5754096508026123
    },
    {
      "auxiliary_loss_clip": 0.01198842,
      "auxiliary_loss_mlp": 0.01032562,
      "balance_loss_clip": 1.05394495,
      "balance_loss_mlp": 1.02372861,
      "epoch": 0.41303432934527745,
      "flos": 23732895715200.0,
      "grad_norm": 1.8262179511798613,
      "language_loss": 0.73115206,
      "learning_rate": 2.6486233419427504e-06,
      "loss": 0.75346613,
      "num_input_tokens_seen": 74183205,
      "step": 3435,
      "time_per_iteration": 2.6401636600494385
    },
    {
      "auxiliary_loss_clip": 0.01187793,
      "auxiliary_loss_mlp": 0.01027412,
      "balance_loss_clip": 1.04995179,
      "balance_loss_mlp": 1.0179112,
      "epoch": 0.41315457223591656,
      "flos": 19755286318080.0,
      "grad_norm": 2.3036011498456537,
      "language_loss": 0.74802595,
      "learning_rate": 2.6478864243124484e-06,
      "loss": 0.77017796,
      "num_input_tokens_seen": 74202870,
      "step": 3436,
      "time_per_iteration": 2.728515625
    },
    {
      "auxiliary_loss_clip": 0.01193425,
      "auxiliary_loss_mlp": 0.01026077,
      "balance_loss_clip": 1.05206132,
      "balance_loss_mlp": 1.01789367,
      "epoch": 0.4132748151265556,
      "flos": 20923316778240.0,
      "grad_norm": 15.30681328278622,
      "language_loss": 0.85338253,
      "learning_rate": 2.6471494083982903e-06,
      "loss": 0.87557757,
      "num_input_tokens_seen": 74222255,
      "step": 3437,
      "time_per_iteration": 2.629293203353882
    },
    {
      "auxiliary_loss_clip": 0.01200279,
      "auxiliary_loss_mlp": 0.01025919,
      "balance_loss_clip": 1.04952049,
      "balance_loss_mlp": 1.01775336,
      "epoch": 0.4133950580171947,
      "flos": 32232520016640.0,
      "grad_norm": 1.7581980090101963,
      "language_loss": 0.74752128,
      "learning_rate": 2.6464122943120818e-06,
      "loss": 0.76978326,
      "num_input_tokens_seen": 74242480,
      "step": 3438,
      "time_per_iteration": 3.644092559814453
    },
    {
      "auxiliary_loss_clip": 0.01195928,
      "auxiliary_loss_mlp": 0.01023622,
      "balance_loss_clip": 1.05288291,
      "balance_loss_mlp": 1.01466322,
      "epoch": 0.41351530090783384,
      "flos": 23292487059840.0,
      "grad_norm": 4.015412260713357,
      "language_loss": 0.81947094,
      "learning_rate": 2.645675082165642e-06,
      "loss": 0.84166646,
      "num_input_tokens_seen": 74258690,
      "step": 3439,
      "time_per_iteration": 2.687889337539673
    },
    {
      "auxiliary_loss_clip": 0.01196354,
      "auxiliary_loss_mlp": 0.01028928,
      "balance_loss_clip": 1.05397534,
      "balance_loss_mlp": 1.01959991,
      "epoch": 0.4136355437984729,
      "flos": 25593571111680.0,
      "grad_norm": 2.414513617879284,
      "language_loss": 0.75915444,
      "learning_rate": 2.644937772070806e-06,
      "loss": 0.78140724,
      "num_input_tokens_seen": 74277135,
      "step": 3440,
      "time_per_iteration": 2.733140230178833
    },
    {
      "auxiliary_loss_clip": 0.01193002,
      "auxiliary_loss_mlp": 0.01027947,
      "balance_loss_clip": 1.05643809,
      "balance_loss_mlp": 1.01892865,
      "epoch": 0.413755786689112,
      "flos": 19828615933440.0,
      "grad_norm": 2.60388709021552,
      "language_loss": 0.83840019,
      "learning_rate": 2.6442003641394225e-06,
      "loss": 0.86060965,
      "num_input_tokens_seen": 74294730,
      "step": 3441,
      "time_per_iteration": 2.561756134033203
    },
    {
      "auxiliary_loss_clip": 0.0119231,
      "auxiliary_loss_mlp": 0.01026349,
      "balance_loss_clip": 1.04982638,
      "balance_loss_mlp": 1.01803398,
      "epoch": 0.4138760295797511,
      "flos": 26870446759680.0,
      "grad_norm": 1.9728827653098746,
      "language_loss": 0.83970112,
      "learning_rate": 2.643462858483356e-06,
      "loss": 0.86188769,
      "num_input_tokens_seen": 74315015,
      "step": 3442,
      "time_per_iteration": 2.653765916824341
    },
    {
      "auxiliary_loss_clip": 0.01193469,
      "auxiliary_loss_mlp": 0.01029304,
      "balance_loss_clip": 1.04725409,
      "balance_loss_mlp": 1.01989257,
      "epoch": 0.41399627247039017,
      "flos": 16399254798720.0,
      "grad_norm": 2.071982743659835,
      "language_loss": 0.73174727,
      "learning_rate": 2.6427252552144856e-06,
      "loss": 0.75397497,
      "num_input_tokens_seen": 74333665,
      "step": 3443,
      "time_per_iteration": 2.703866481781006
    },
    {
      "auxiliary_loss_clip": 0.01191112,
      "auxiliary_loss_mlp": 0.0103283,
      "balance_loss_clip": 1.05478549,
      "balance_loss_mlp": 1.02384782,
      "epoch": 0.4141165153610293,
      "flos": 22930220442240.0,
      "grad_norm": 2.1124865204870593,
      "language_loss": 0.75065327,
      "learning_rate": 2.6419875544447044e-06,
      "loss": 0.77289271,
      "num_input_tokens_seen": 74355065,
      "step": 3444,
      "time_per_iteration": 2.6274569034576416
    },
    {
      "auxiliary_loss_clip": 0.01192517,
      "auxiliary_loss_mlp": 0.01031456,
      "balance_loss_clip": 1.05545402,
      "balance_loss_mlp": 1.02236056,
      "epoch": 0.4142367582516684,
      "flos": 25192556697600.0,
      "grad_norm": 3.586144129569201,
      "language_loss": 0.7165615,
      "learning_rate": 2.6412497562859218e-06,
      "loss": 0.73880124,
      "num_input_tokens_seen": 74376345,
      "step": 3445,
      "time_per_iteration": 2.6340994834899902
    },
    {
      "auxiliary_loss_clip": 0.01197605,
      "auxiliary_loss_mlp": 0.01027003,
      "balance_loss_clip": 1.05358458,
      "balance_loss_mlp": 1.0179553,
      "epoch": 0.41435700114230745,
      "flos": 21690476478720.0,
      "grad_norm": 3.158167341265037,
      "language_loss": 0.75734872,
      "learning_rate": 2.6405118608500617e-06,
      "loss": 0.77959478,
      "num_input_tokens_seen": 74395170,
      "step": 3446,
      "time_per_iteration": 2.6003525257110596
    },
    {
      "auxiliary_loss_clip": 0.01190108,
      "auxiliary_loss_mlp": 0.01028409,
      "balance_loss_clip": 1.05340564,
      "balance_loss_mlp": 1.02001667,
      "epoch": 0.41447724403294656,
      "flos": 25995160143360.0,
      "grad_norm": 1.874235109214498,
      "language_loss": 0.81491733,
      "learning_rate": 2.6397738682490613e-06,
      "loss": 0.83710253,
      "num_input_tokens_seen": 74416070,
      "step": 3447,
      "time_per_iteration": 2.693113327026367
    },
    {
      "auxiliary_loss_clip": 0.01191147,
      "auxiliary_loss_mlp": 0.0102765,
      "balance_loss_clip": 1.05562997,
      "balance_loss_mlp": 1.01894212,
      "epoch": 0.41459748692358567,
      "flos": 18259678800000.0,
      "grad_norm": 2.144810707439164,
      "language_loss": 0.75191152,
      "learning_rate": 2.6390357785948734e-06,
      "loss": 0.77409947,
      "num_input_tokens_seen": 74433185,
      "step": 3448,
      "time_per_iteration": 2.599738121032715
    },
    {
      "auxiliary_loss_clip": 0.01196162,
      "auxiliary_loss_mlp": 0.01030649,
      "balance_loss_clip": 1.05673707,
      "balance_loss_mlp": 1.02140999,
      "epoch": 0.4147177298142247,
      "flos": 24168456034560.0,
      "grad_norm": 1.8149589340858172,
      "language_loss": 0.80335218,
      "learning_rate": 2.6382975919994667e-06,
      "loss": 0.82562029,
      "num_input_tokens_seen": 74453760,
      "step": 3449,
      "time_per_iteration": 2.633561372756958
    },
    {
      "auxiliary_loss_clip": 0.01198747,
      "auxiliary_loss_mlp": 0.0102462,
      "balance_loss_clip": 1.05404043,
      "balance_loss_mlp": 1.01667762,
      "epoch": 0.41483797270486383,
      "flos": 20084659056000.0,
      "grad_norm": 1.8256709327216372,
      "language_loss": 0.7309221,
      "learning_rate": 2.637559308574822e-06,
      "loss": 0.75315583,
      "num_input_tokens_seen": 74473505,
      "step": 3450,
      "time_per_iteration": 2.729243040084839
    },
    {
      "auxiliary_loss_clip": 0.01189306,
      "auxiliary_loss_mlp": 0.01027409,
      "balance_loss_clip": 1.05388713,
      "balance_loss_mlp": 1.01889145,
      "epoch": 0.4149582155955029,
      "flos": 30081040110720.0,
      "grad_norm": 2.0878538748693467,
      "language_loss": 0.71769655,
      "learning_rate": 2.6368209284329376e-06,
      "loss": 0.73986375,
      "num_input_tokens_seen": 74494135,
      "step": 3451,
      "time_per_iteration": 2.7020483016967773
    },
    {
      "auxiliary_loss_clip": 0.01191085,
      "auxiliary_loss_mlp": 0.01028873,
      "balance_loss_clip": 1.05143547,
      "balance_loss_mlp": 1.020028,
      "epoch": 0.415078458486142,
      "flos": 16764394504320.0,
      "grad_norm": 2.2868907522015296,
      "language_loss": 0.7539016,
      "learning_rate": 2.636082451685825e-06,
      "loss": 0.77610123,
      "num_input_tokens_seen": 74512335,
      "step": 3452,
      "time_per_iteration": 2.5856387615203857
    },
    {
      "auxiliary_loss_clip": 0.01199265,
      "auxiliary_loss_mlp": 0.01028348,
      "balance_loss_clip": 1.05558944,
      "balance_loss_mlp": 1.01941323,
      "epoch": 0.4151987013767811,
      "flos": 26033692458240.0,
      "grad_norm": 1.5160126381487429,
      "language_loss": 0.8635906,
      "learning_rate": 2.6353438784455094e-06,
      "loss": 0.88586664,
      "num_input_tokens_seen": 74535620,
      "step": 3453,
      "time_per_iteration": 2.6212618350982666
    },
    {
      "auxiliary_loss_clip": 0.01192881,
      "auxiliary_loss_mlp": 0.0103039,
      "balance_loss_clip": 1.05379808,
      "balance_loss_mlp": 1.02073979,
      "epoch": 0.41531894426742016,
      "flos": 24608002763520.0,
      "grad_norm": 2.8652623750252393,
      "language_loss": 0.71786863,
      "learning_rate": 2.6346052088240326e-06,
      "loss": 0.74010134,
      "num_input_tokens_seen": 74555140,
      "step": 3454,
      "time_per_iteration": 2.695056200027466
    },
    {
      "auxiliary_loss_clip": 0.011977,
      "auxiliary_loss_mlp": 0.0102982,
      "balance_loss_clip": 1.05083501,
      "balance_loss_mlp": 1.02024198,
      "epoch": 0.4154391871580593,
      "flos": 14975791747200.0,
      "grad_norm": 2.1738483110437583,
      "language_loss": 0.77501291,
      "learning_rate": 2.63386644293345e-06,
      "loss": 0.79728818,
      "num_input_tokens_seen": 74571485,
      "step": 3455,
      "time_per_iteration": 2.7670915126800537
    },
    {
      "auxiliary_loss_clip": 0.01192131,
      "auxiliary_loss_mlp": 0.01026936,
      "balance_loss_clip": 1.04743338,
      "balance_loss_mlp": 1.0185374,
      "epoch": 0.4155594300486984,
      "flos": 14647173194880.0,
      "grad_norm": 2.8972392693014584,
      "language_loss": 0.82838047,
      "learning_rate": 2.633127580885833e-06,
      "loss": 0.8505711,
      "num_input_tokens_seen": 74585985,
      "step": 3456,
      "time_per_iteration": 3.57721209526062
    },
    {
      "auxiliary_loss_clip": 0.01191764,
      "auxiliary_loss_mlp": 0.01032585,
      "balance_loss_clip": 1.05886185,
      "balance_loss_mlp": 1.02374005,
      "epoch": 0.41567967293933744,
      "flos": 29497276275840.0,
      "grad_norm": 2.424927214072053,
      "language_loss": 0.65431821,
      "learning_rate": 2.632388622793265e-06,
      "loss": 0.67656171,
      "num_input_tokens_seen": 74605140,
      "step": 3457,
      "time_per_iteration": 3.62406849861145
    },
    {
      "auxiliary_loss_clip": 0.01193689,
      "auxiliary_loss_mlp": 0.01029781,
      "balance_loss_clip": 1.05637884,
      "balance_loss_mlp": 1.02116776,
      "epoch": 0.41579991582997655,
      "flos": 19238387650560.0,
      "grad_norm": 2.189607396088256,
      "language_loss": 0.67527002,
      "learning_rate": 2.6316495687678457e-06,
      "loss": 0.69750476,
      "num_input_tokens_seen": 74623790,
      "step": 3458,
      "time_per_iteration": 2.580960512161255
    },
    {
      "auxiliary_loss_clip": 0.01191786,
      "auxiliary_loss_mlp": 0.01028208,
      "balance_loss_clip": 1.04648638,
      "balance_loss_mlp": 1.01907051,
      "epoch": 0.41592015872061566,
      "flos": 24462061804800.0,
      "grad_norm": 2.614509832356751,
      "language_loss": 0.76596296,
      "learning_rate": 2.6309104189216887e-06,
      "loss": 0.78816289,
      "num_input_tokens_seen": 74641355,
      "step": 3459,
      "time_per_iteration": 2.708326578140259
    },
    {
      "auxiliary_loss_clip": 0.0118367,
      "auxiliary_loss_mlp": 0.01209163,
      "balance_loss_clip": 1.04817057,
      "balance_loss_mlp": 1.00048327,
      "epoch": 0.4160404016112547,
      "flos": 20775651966720.0,
      "grad_norm": 2.218812305347357,
      "language_loss": 0.74842703,
      "learning_rate": 2.630171173366923e-06,
      "loss": 0.77235538,
      "num_input_tokens_seen": 74657155,
      "step": 3460,
      "time_per_iteration": 2.6349306106567383
    },
    {
      "auxiliary_loss_clip": 0.01197616,
      "auxiliary_loss_mlp": 0.01027779,
      "balance_loss_clip": 1.04789805,
      "balance_loss_mlp": 1.01865387,
      "epoch": 0.41616064450189383,
      "flos": 13916462820480.0,
      "grad_norm": 2.797043419191097,
      "language_loss": 0.75168008,
      "learning_rate": 2.629431832215691e-06,
      "loss": 0.77393407,
      "num_input_tokens_seen": 74671960,
      "step": 3461,
      "time_per_iteration": 3.563354253768921
    },
    {
      "auxiliary_loss_clip": 0.01191961,
      "auxiliary_loss_mlp": 0.01026043,
      "balance_loss_clip": 1.05237675,
      "balance_loss_mlp": 1.01729918,
      "epoch": 0.41628088739253294,
      "flos": 20010826650240.0,
      "grad_norm": 1.6099514517012032,
      "language_loss": 0.87033045,
      "learning_rate": 2.628692395580151e-06,
      "loss": 0.89251047,
      "num_input_tokens_seen": 74692050,
      "step": 3462,
      "time_per_iteration": 2.745865821838379
    },
    {
      "auxiliary_loss_clip": 0.01182781,
      "auxiliary_loss_mlp": 0.01030344,
      "balance_loss_clip": 1.0447737,
      "balance_loss_mlp": 1.02136803,
      "epoch": 0.416401130283172,
      "flos": 29168801377920.0,
      "grad_norm": 1.7658758609200125,
      "language_loss": 0.79232967,
      "learning_rate": 2.6279528635724747e-06,
      "loss": 0.81446087,
      "num_input_tokens_seen": 74712205,
      "step": 3463,
      "time_per_iteration": 2.765066623687744
    },
    {
      "auxiliary_loss_clip": 0.011935,
      "auxiliary_loss_mlp": 0.01029796,
      "balance_loss_clip": 1.05337477,
      "balance_loss_mlp": 1.02018797,
      "epoch": 0.4165213731738111,
      "flos": 16246813478400.0,
      "grad_norm": 3.142279376517473,
      "language_loss": 0.78847098,
      "learning_rate": 2.627213236304848e-06,
      "loss": 0.81070393,
      "num_input_tokens_seen": 74729005,
      "step": 3464,
      "time_per_iteration": 2.667038917541504
    },
    {
      "auxiliary_loss_clip": 0.01196067,
      "auxiliary_loss_mlp": 0.01023871,
      "balance_loss_clip": 1.05509734,
      "balance_loss_mlp": 1.01501346,
      "epoch": 0.4166416160644502,
      "flos": 33765438787200.0,
      "grad_norm": 2.3127532567922993,
      "language_loss": 0.70854414,
      "learning_rate": 2.626473513889472e-06,
      "loss": 0.73074353,
      "num_input_tokens_seen": 74751385,
      "step": 3465,
      "time_per_iteration": 3.7413907051086426
    },
    {
      "auxiliary_loss_clip": 0.01186032,
      "auxiliary_loss_mlp": 0.0103173,
      "balance_loss_clip": 1.05415702,
      "balance_loss_mlp": 1.02284849,
      "epoch": 0.41676185895508927,
      "flos": 20917498775040.0,
      "grad_norm": 1.871357831475488,
      "language_loss": 0.8302834,
      "learning_rate": 2.625733696438562e-06,
      "loss": 0.85246098,
      "num_input_tokens_seen": 74768890,
      "step": 3466,
      "time_per_iteration": 2.676854133605957
    },
    {
      "auxiliary_loss_clip": 0.01194145,
      "auxiliary_loss_mlp": 0.0103132,
      "balance_loss_clip": 1.05371666,
      "balance_loss_mlp": 1.022439,
      "epoch": 0.4168821018457284,
      "flos": 18406122549120.0,
      "grad_norm": 1.7673490192261563,
      "language_loss": 0.75698555,
      "learning_rate": 2.6249937840643476e-06,
      "loss": 0.77924019,
      "num_input_tokens_seen": 74787195,
      "step": 3467,
      "time_per_iteration": 2.6177725791931152
    },
    {
      "auxiliary_loss_clip": 0.01192589,
      "auxiliary_loss_mlp": 0.01209053,
      "balance_loss_clip": 1.05811977,
      "balance_loss_mlp": 1.00048184,
      "epoch": 0.41700234473636744,
      "flos": 18698399516160.0,
      "grad_norm": 1.6788828502314068,
      "language_loss": 0.66805398,
      "learning_rate": 2.6242537768790733e-06,
      "loss": 0.69207042,
      "num_input_tokens_seen": 74806350,
      "step": 3468,
      "time_per_iteration": 2.5782477855682373
    },
    {
      "auxiliary_loss_clip": 0.01195347,
      "auxiliary_loss_mlp": 0.01031279,
      "balance_loss_clip": 1.05703247,
      "balance_loss_mlp": 1.02207601,
      "epoch": 0.41712258762700655,
      "flos": 31033283616000.0,
      "grad_norm": 1.8791875945391667,
      "language_loss": 0.6851753,
      "learning_rate": 2.6235136749949975e-06,
      "loss": 0.70744157,
      "num_input_tokens_seen": 74829800,
      "step": 3469,
      "time_per_iteration": 2.691143035888672
    },
    {
      "auxiliary_loss_clip": 0.01189587,
      "auxiliary_loss_mlp": 0.01027203,
      "balance_loss_clip": 1.05551422,
      "balance_loss_mlp": 1.01763654,
      "epoch": 0.41724283051764566,
      "flos": 35914763877120.0,
      "grad_norm": 2.0940516001507983,
      "language_loss": 0.6132443,
      "learning_rate": 2.6227734785243924e-06,
      "loss": 0.63541216,
      "num_input_tokens_seen": 74849760,
      "step": 3470,
      "time_per_iteration": 2.675229072570801
    },
    {
      "auxiliary_loss_clip": 0.01194932,
      "auxiliary_loss_mlp": 0.01026858,
      "balance_loss_clip": 1.04561007,
      "balance_loss_mlp": 1.01822186,
      "epoch": 0.4173630734082847,
      "flos": 25333649320320.0,
      "grad_norm": 2.3657399793317158,
      "language_loss": 0.79066944,
      "learning_rate": 2.6220331875795466e-06,
      "loss": 0.81288731,
      "num_input_tokens_seen": 74869110,
      "step": 3471,
      "time_per_iteration": 2.7234702110290527
    },
    {
      "auxiliary_loss_clip": 0.01190385,
      "auxiliary_loss_mlp": 0.01029372,
      "balance_loss_clip": 1.05470002,
      "balance_loss_mlp": 1.02019262,
      "epoch": 0.4174833162989238,
      "flos": 26685398868480.0,
      "grad_norm": 1.6071189465506797,
      "language_loss": 0.75222552,
      "learning_rate": 2.62129280227276e-06,
      "loss": 0.77442306,
      "num_input_tokens_seen": 74889110,
      "step": 3472,
      "time_per_iteration": 2.635765790939331
    },
    {
      "auxiliary_loss_clip": 0.01198537,
      "auxiliary_loss_mlp": 0.01032451,
      "balance_loss_clip": 1.05534387,
      "balance_loss_mlp": 1.02306294,
      "epoch": 0.41760355918956293,
      "flos": 74739584010240.0,
      "grad_norm": 2.0718901496323854,
      "language_loss": 0.68653691,
      "learning_rate": 2.62055232271635e-06,
      "loss": 0.70884681,
      "num_input_tokens_seen": 74916260,
      "step": 3473,
      "time_per_iteration": 2.9713118076324463
    },
    {
      "auxiliary_loss_clip": 0.01186468,
      "auxiliary_loss_mlp": 0.01030575,
      "balance_loss_clip": 1.04884696,
      "balance_loss_mlp": 1.02184272,
      "epoch": 0.417723802080202,
      "flos": 14317513148160.0,
      "grad_norm": 3.6863411720962604,
      "language_loss": 0.87533176,
      "learning_rate": 2.619811749022646e-06,
      "loss": 0.89750206,
      "num_input_tokens_seen": 74931570,
      "step": 3474,
      "time_per_iteration": 2.531658172607422
    },
    {
      "auxiliary_loss_clip": 0.01193798,
      "auxiliary_loss_mlp": 0.01027382,
      "balance_loss_clip": 1.05492449,
      "balance_loss_mlp": 1.01797664,
      "epoch": 0.4178440449708411,
      "flos": 14643797316480.0,
      "grad_norm": 2.407862287675624,
      "language_loss": 0.71494114,
      "learning_rate": 2.6190710813039917e-06,
      "loss": 0.73715299,
      "num_input_tokens_seen": 74944695,
      "step": 3475,
      "time_per_iteration": 2.5206122398376465
    },
    {
      "auxiliary_loss_clip": 0.01191837,
      "auxiliary_loss_mlp": 0.0120971,
      "balance_loss_clip": 1.04477334,
      "balance_loss_mlp": 1.00047255,
      "epoch": 0.4179642878614802,
      "flos": 21507296094720.0,
      "grad_norm": 11.360911790344703,
      "language_loss": 0.84370977,
      "learning_rate": 2.618330319672747e-06,
      "loss": 0.86772525,
      "num_input_tokens_seen": 74964115,
      "step": 3476,
      "time_per_iteration": 2.6451969146728516
    },
    {
      "auxiliary_loss_clip": 0.01194073,
      "auxiliary_loss_mlp": 0.0103028,
      "balance_loss_clip": 1.05765736,
      "balance_loss_mlp": 1.02139914,
      "epoch": 0.41808453075211927,
      "flos": 18441997257600.0,
      "grad_norm": 2.615304814738653,
      "language_loss": 0.91859102,
      "learning_rate": 2.617589464241284e-06,
      "loss": 0.94083458,
      "num_input_tokens_seen": 74978515,
      "step": 3477,
      "time_per_iteration": 2.5960869789123535
    },
    {
      "auxiliary_loss_clip": 0.01203458,
      "auxiliary_loss_mlp": 0.01024383,
      "balance_loss_clip": 1.05323827,
      "balance_loss_mlp": 1.01590753,
      "epoch": 0.4182047736427584,
      "flos": 20301020628480.0,
      "grad_norm": 2.1341471923929327,
      "language_loss": 0.74622887,
      "learning_rate": 2.6168485151219914e-06,
      "loss": 0.7685073,
      "num_input_tokens_seen": 74998135,
      "step": 3478,
      "time_per_iteration": 2.650007724761963
    },
    {
      "auxiliary_loss_clip": 0.01195549,
      "auxiliary_loss_mlp": 0.01028627,
      "balance_loss_clip": 1.05599213,
      "balance_loss_mlp": 1.01956129,
      "epoch": 0.4183250165333975,
      "flos": 18876623823360.0,
      "grad_norm": 4.463837815361589,
      "language_loss": 0.71303952,
      "learning_rate": 2.616107472427269e-06,
      "loss": 0.73528135,
      "num_input_tokens_seen": 75012830,
      "step": 3479,
      "time_per_iteration": 2.6025986671447754
    },
    {
      "auxiliary_loss_clip": 0.01199446,
      "auxiliary_loss_mlp": 0.01023442,
      "balance_loss_clip": 1.05461061,
      "balance_loss_mlp": 1.01438832,
      "epoch": 0.41844525942403654,
      "flos": 17740050698880.0,
      "grad_norm": 2.6660931089199384,
      "language_loss": 0.76223505,
      "learning_rate": 2.615366336269533e-06,
      "loss": 0.78446388,
      "num_input_tokens_seen": 75026495,
      "step": 3480,
      "time_per_iteration": 2.596750020980835
    },
    {
      "auxiliary_loss_clip": 0.01194667,
      "auxiliary_loss_mlp": 0.01035619,
      "balance_loss_clip": 1.05636096,
      "balance_loss_mlp": 1.02570713,
      "epoch": 0.41856550231467565,
      "flos": 18361377181440.0,
      "grad_norm": 2.397073385931514,
      "language_loss": 0.80389202,
      "learning_rate": 2.6146251067612126e-06,
      "loss": 0.82619488,
      "num_input_tokens_seen": 75041970,
      "step": 3481,
      "time_per_iteration": 3.6402907371520996
    },
    {
      "auxiliary_loss_clip": 0.01194246,
      "auxiliary_loss_mlp": 0.01027535,
      "balance_loss_clip": 1.05800438,
      "balance_loss_mlp": 1.01867223,
      "epoch": 0.41868574520531476,
      "flos": 22781801445120.0,
      "grad_norm": 1.6322100381007982,
      "language_loss": 0.82519394,
      "learning_rate": 2.6138837840147525e-06,
      "loss": 0.84741175,
      "num_input_tokens_seen": 75061005,
      "step": 3482,
      "time_per_iteration": 2.65586519241333
    },
    {
      "auxiliary_loss_clip": 0.01196008,
      "auxiliary_loss_mlp": 0.01027521,
      "balance_loss_clip": 1.05175793,
      "balance_loss_mlp": 1.01884842,
      "epoch": 0.4188059880959538,
      "flos": 13699167494400.0,
      "grad_norm": 2.391607053396399,
      "language_loss": 0.76270103,
      "learning_rate": 2.6131423681426103e-06,
      "loss": 0.78493631,
      "num_input_tokens_seen": 75076920,
      "step": 3483,
      "time_per_iteration": 2.7028117179870605
    },
    {
      "auxiliary_loss_clip": 0.01192432,
      "auxiliary_loss_mlp": 0.01026604,
      "balance_loss_clip": 1.05784106,
      "balance_loss_mlp": 1.01851285,
      "epoch": 0.41892623098659293,
      "flos": 37818281220480.0,
      "grad_norm": 1.7753585474922424,
      "language_loss": 0.73283964,
      "learning_rate": 2.6124008592572587e-06,
      "loss": 0.75502992,
      "num_input_tokens_seen": 75100905,
      "step": 3484,
      "time_per_iteration": 3.635254144668579
    },
    {
      "auxiliary_loss_clip": 0.01196592,
      "auxiliary_loss_mlp": 0.01025719,
      "balance_loss_clip": 1.05722237,
      "balance_loss_mlp": 1.01597357,
      "epoch": 0.419046473877232,
      "flos": 23258874908160.0,
      "grad_norm": 2.3228378973884056,
      "language_loss": 0.81625283,
      "learning_rate": 2.6116592574711835e-06,
      "loss": 0.83847594,
      "num_input_tokens_seen": 75119205,
      "step": 3485,
      "time_per_iteration": 2.6011228561401367
    },
    {
      "auxiliary_loss_clip": 0.01197683,
      "auxiliary_loss_mlp": 0.01036436,
      "balance_loss_clip": 1.05916476,
      "balance_loss_mlp": 1.02704263,
      "epoch": 0.4191667167678711,
      "flos": 20741034234240.0,
      "grad_norm": 2.34266582476483,
      "language_loss": 0.84327686,
      "learning_rate": 2.6109175628968853e-06,
      "loss": 0.86561811,
      "num_input_tokens_seen": 75138970,
      "step": 3486,
      "time_per_iteration": 2.594137191772461
    },
    {
      "auxiliary_loss_clip": 0.01185788,
      "auxiliary_loss_mlp": 0.01033034,
      "balance_loss_clip": 1.05400848,
      "balance_loss_mlp": 1.02440882,
      "epoch": 0.4192869596585102,
      "flos": 23586416052480.0,
      "grad_norm": 1.8638078976576993,
      "language_loss": 0.83075321,
      "learning_rate": 2.610175775646878e-06,
      "loss": 0.85294139,
      "num_input_tokens_seen": 75157550,
      "step": 3487,
      "time_per_iteration": 2.5551435947418213
    },
    {
      "auxiliary_loss_clip": 0.01193491,
      "auxiliary_loss_mlp": 0.01031153,
      "balance_loss_clip": 1.05086541,
      "balance_loss_mlp": 1.02205753,
      "epoch": 0.41940720254914926,
      "flos": 25081269384960.0,
      "grad_norm": 8.175404201954523,
      "language_loss": 0.73060775,
      "learning_rate": 2.6094338958336907e-06,
      "loss": 0.75285423,
      "num_input_tokens_seen": 75176220,
      "step": 3488,
      "time_per_iteration": 3.610905170440674
    },
    {
      "auxiliary_loss_clip": 0.01196577,
      "auxiliary_loss_mlp": 0.01027478,
      "balance_loss_clip": 1.05433905,
      "balance_loss_mlp": 1.01876342,
      "epoch": 0.41952744543978837,
      "flos": 15554132628480.0,
      "grad_norm": 2.0141160793447734,
      "language_loss": 0.82120699,
      "learning_rate": 2.608691923569867e-06,
      "loss": 0.84344757,
      "num_input_tokens_seen": 75193095,
      "step": 3489,
      "time_per_iteration": 2.6116373538970947
    },
    {
      "auxiliary_loss_clip": 0.01197246,
      "auxiliary_loss_mlp": 0.01032231,
      "balance_loss_clip": 1.0569222,
      "balance_loss_mlp": 1.0231179,
      "epoch": 0.4196476883304275,
      "flos": 24644775312000.0,
      "grad_norm": 1.646021896049024,
      "language_loss": 0.761195,
      "learning_rate": 2.6079498589679616e-06,
      "loss": 0.78348976,
      "num_input_tokens_seen": 75214185,
      "step": 3490,
      "time_per_iteration": 2.66839861869812
    },
    {
      "auxiliary_loss_clip": 0.01196685,
      "auxiliary_loss_mlp": 0.01038071,
      "balance_loss_clip": 1.04484653,
      "balance_loss_mlp": 1.02767015,
      "epoch": 0.41976793122106654,
      "flos": 24531333183360.0,
      "grad_norm": 2.3839270248002746,
      "language_loss": 0.76065075,
      "learning_rate": 2.6072077021405465e-06,
      "loss": 0.78299826,
      "num_input_tokens_seen": 75233020,
      "step": 3491,
      "time_per_iteration": 2.75496244430542
    },
    {
      "auxiliary_loss_clip": 0.01209696,
      "auxiliary_loss_mlp": 0.0103246,
      "balance_loss_clip": 1.05346346,
      "balance_loss_mlp": 1.02377558,
      "epoch": 0.41988817411170565,
      "flos": 21175301664000.0,
      "grad_norm": 2.1206318334191785,
      "language_loss": 0.69548059,
      "learning_rate": 2.6064654532002054e-06,
      "loss": 0.71790218,
      "num_input_tokens_seen": 75252030,
      "step": 3492,
      "time_per_iteration": 3.5712687969207764
    },
    {
      "auxiliary_loss_clip": 0.01193396,
      "auxiliary_loss_mlp": 0.01033691,
      "balance_loss_clip": 1.05796623,
      "balance_loss_mlp": 1.02459538,
      "epoch": 0.42000841700234476,
      "flos": 31649402626560.0,
      "grad_norm": 1.6195357229472824,
      "language_loss": 0.76037425,
      "learning_rate": 2.6057231122595375e-06,
      "loss": 0.78264511,
      "num_input_tokens_seen": 75273340,
      "step": 3493,
      "time_per_iteration": 2.6975419521331787
    },
    {
      "auxiliary_loss_clip": 0.01197541,
      "auxiliary_loss_mlp": 0.01033057,
      "balance_loss_clip": 1.0516293,
      "balance_loss_mlp": 1.02380633,
      "epoch": 0.4201286598929838,
      "flos": 21281525159040.0,
      "grad_norm": 1.8041284864590363,
      "language_loss": 0.73018229,
      "learning_rate": 2.604980679431154e-06,
      "loss": 0.75248826,
      "num_input_tokens_seen": 75291580,
      "step": 3494,
      "time_per_iteration": 2.6289234161376953
    },
    {
      "auxiliary_loss_clip": 0.01198831,
      "auxiliary_loss_mlp": 0.01028516,
      "balance_loss_clip": 1.05469048,
      "balance_loss_mlp": 1.01946783,
      "epoch": 0.4202489027836229,
      "flos": 18546532813440.0,
      "grad_norm": 2.062249425532304,
      "language_loss": 0.74368656,
      "learning_rate": 2.604238154827684e-06,
      "loss": 0.7659601,
      "num_input_tokens_seen": 75308205,
      "step": 3495,
      "time_per_iteration": 2.5774340629577637
    },
    {
      "auxiliary_loss_clip": 0.01196995,
      "auxiliary_loss_mlp": 0.01025777,
      "balance_loss_clip": 1.05574465,
      "balance_loss_mlp": 1.01712871,
      "epoch": 0.42036914567426203,
      "flos": 19317643009920.0,
      "grad_norm": 2.254812473824473,
      "language_loss": 0.72941148,
      "learning_rate": 2.6034955385617656e-06,
      "loss": 0.75163925,
      "num_input_tokens_seen": 75326535,
      "step": 3496,
      "time_per_iteration": 2.6422150135040283
    },
    {
      "auxiliary_loss_clip": 0.01117403,
      "auxiliary_loss_mlp": 0.01002739,
      "balance_loss_clip": 1.03047836,
      "balance_loss_mlp": 1.00169039,
      "epoch": 0.4204893885649011,
      "flos": 67842942935040.0,
      "grad_norm": 0.7223748936011465,
      "language_loss": 0.61629391,
      "learning_rate": 2.6027528307460544e-06,
      "loss": 0.6374954,
      "num_input_tokens_seen": 75390540,
      "step": 3497,
      "time_per_iteration": 3.3484580516815186
    },
    {
      "auxiliary_loss_clip": 0.01193124,
      "auxiliary_loss_mlp": 0.01027077,
      "balance_loss_clip": 1.05690587,
      "balance_loss_mlp": 1.01908445,
      "epoch": 0.4206096314555402,
      "flos": 21908777385600.0,
      "grad_norm": 3.05734282628713,
      "language_loss": 0.86730182,
      "learning_rate": 2.602010031493217e-06,
      "loss": 0.88950378,
      "num_input_tokens_seen": 75408770,
      "step": 3498,
      "time_per_iteration": 2.6971943378448486
    },
    {
      "auxiliary_loss_clip": 0.01194299,
      "auxiliary_loss_mlp": 0.01028465,
      "balance_loss_clip": 1.05022955,
      "balance_loss_mlp": 1.01944101,
      "epoch": 0.42072987434617926,
      "flos": 29278185269760.0,
      "grad_norm": 2.908227003399854,
      "language_loss": 0.87221819,
      "learning_rate": 2.6012671409159367e-06,
      "loss": 0.89444584,
      "num_input_tokens_seen": 75430105,
      "step": 3499,
      "time_per_iteration": 2.821342945098877
    },
    {
      "auxiliary_loss_clip": 0.01191,
      "auxiliary_loss_mlp": 0.01033473,
      "balance_loss_clip": 1.05439401,
      "balance_loss_mlp": 1.0237751,
      "epoch": 0.42085011723681837,
      "flos": 27600726170880.0,
      "grad_norm": 1.868549882904537,
      "language_loss": 0.82044029,
      "learning_rate": 2.6005241591269097e-06,
      "loss": 0.84268498,
      "num_input_tokens_seen": 75449475,
      "step": 3500,
      "time_per_iteration": 2.844896078109741
    },
    {
      "auxiliary_loss_clip": 0.01194515,
      "auxiliary_loss_mlp": 0.01030588,
      "balance_loss_clip": 1.05382061,
      "balance_loss_mlp": 1.02164173,
      "epoch": 0.4209703601274575,
      "flos": 27818632028160.0,
      "grad_norm": 1.656312946189618,
      "language_loss": 0.79231864,
      "learning_rate": 2.5997810862388454e-06,
      "loss": 0.81456965,
      "num_input_tokens_seen": 75469315,
      "step": 3501,
      "time_per_iteration": 2.789642333984375
    },
    {
      "auxiliary_loss_clip": 0.01198391,
      "auxiliary_loss_mlp": 0.01031083,
      "balance_loss_clip": 1.05240893,
      "balance_loss_mlp": 1.02215445,
      "epoch": 0.42109060301809653,
      "flos": 27525529048320.0,
      "grad_norm": 2.5722758365314076,
      "language_loss": 0.75910091,
      "learning_rate": 2.599037922364467e-06,
      "loss": 0.78139567,
      "num_input_tokens_seen": 75488215,
      "step": 3502,
      "time_per_iteration": 2.720567226409912
    },
    {
      "auxiliary_loss_clip": 0.01195083,
      "auxiliary_loss_mlp": 0.01026034,
      "balance_loss_clip": 1.05263758,
      "balance_loss_mlp": 1.01746333,
      "epoch": 0.42121084590873564,
      "flos": 29314275459840.0,
      "grad_norm": 2.199523069267517,
      "language_loss": 0.75190979,
      "learning_rate": 2.5982946676165112e-06,
      "loss": 0.77412105,
      "num_input_tokens_seen": 75507985,
      "step": 3503,
      "time_per_iteration": 2.7838542461395264
    },
    {
      "auxiliary_loss_clip": 0.01116102,
      "auxiliary_loss_mlp": 0.01004443,
      "balance_loss_clip": 1.03707087,
      "balance_loss_mlp": 1.0032028,
      "epoch": 0.42133108879937475,
      "flos": 67398835178880.0,
      "grad_norm": 0.7261686294029569,
      "language_loss": 0.57557386,
      "learning_rate": 2.5975513221077313e-06,
      "loss": 0.59677935,
      "num_input_tokens_seen": 75571955,
      "step": 3504,
      "time_per_iteration": 3.384316921234131
    },
    {
      "auxiliary_loss_clip": 0.01188015,
      "auxiliary_loss_mlp": 0.01036546,
      "balance_loss_clip": 1.05313659,
      "balance_loss_mlp": 1.02753329,
      "epoch": 0.4214513316900138,
      "flos": 23106038538240.0,
      "grad_norm": 2.6987635465656825,
      "language_loss": 0.88527048,
      "learning_rate": 2.5968078859508897e-06,
      "loss": 0.90751612,
      "num_input_tokens_seen": 75589155,
      "step": 3505,
      "time_per_iteration": 2.6669201850891113
    },
    {
      "auxiliary_loss_clip": 0.01194983,
      "auxiliary_loss_mlp": 0.0103029,
      "balance_loss_clip": 1.05562687,
      "balance_loss_mlp": 1.02148592,
      "epoch": 0.4215715745806529,
      "flos": 15336190857600.0,
      "grad_norm": 2.3078902766558507,
      "language_loss": 0.79909474,
      "learning_rate": 2.5960643592587673e-06,
      "loss": 0.82134748,
      "num_input_tokens_seen": 75606565,
      "step": 3506,
      "time_per_iteration": 2.6166813373565674
    },
    {
      "auxiliary_loss_clip": 0.01197139,
      "auxiliary_loss_mlp": 0.01030021,
      "balance_loss_clip": 1.05031955,
      "balance_loss_mlp": 1.02171254,
      "epoch": 0.42169181747129203,
      "flos": 22127257860480.0,
      "grad_norm": 2.600969299319444,
      "language_loss": 0.81747198,
      "learning_rate": 2.5953207421441553e-06,
      "loss": 0.83974361,
      "num_input_tokens_seen": 75625165,
      "step": 3507,
      "time_per_iteration": 2.6838624477386475
    },
    {
      "auxiliary_loss_clip": 0.01199767,
      "auxiliary_loss_mlp": 0.01033056,
      "balance_loss_clip": 1.05176258,
      "balance_loss_mlp": 1.02463412,
      "epoch": 0.4218120603619311,
      "flos": 22630724841600.0,
      "grad_norm": 3.0702447075796884,
      "language_loss": 0.75202513,
      "learning_rate": 2.5945770347198603e-06,
      "loss": 0.77435333,
      "num_input_tokens_seen": 75643320,
      "step": 3508,
      "time_per_iteration": 3.7036962509155273
    },
    {
      "auxiliary_loss_clip": 0.01191979,
      "auxiliary_loss_mlp": 0.01021803,
      "balance_loss_clip": 1.05155802,
      "balance_loss_mlp": 1.01363146,
      "epoch": 0.4219323032525702,
      "flos": 19682818629120.0,
      "grad_norm": 1.8021779181939106,
      "language_loss": 0.82029313,
      "learning_rate": 2.593833237098701e-06,
      "loss": 0.84243095,
      "num_input_tokens_seen": 75660920,
      "step": 3509,
      "time_per_iteration": 2.6737301349639893
    },
    {
      "auxiliary_loss_clip": 0.01191452,
      "auxiliary_loss_mlp": 0.01029414,
      "balance_loss_clip": 1.05109596,
      "balance_loss_mlp": 1.02046144,
      "epoch": 0.4220525461432093,
      "flos": 30190747224960.0,
      "grad_norm": 2.0237911612768302,
      "language_loss": 0.62563258,
      "learning_rate": 2.593089349393512e-06,
      "loss": 0.64784122,
      "num_input_tokens_seen": 75681410,
      "step": 3510,
      "time_per_iteration": 3.6675825119018555
    },
    {
      "auxiliary_loss_clip": 0.0119255,
      "auxiliary_loss_mlp": 0.01024674,
      "balance_loss_clip": 1.05639172,
      "balance_loss_mlp": 1.01611447,
      "epoch": 0.42217278903384836,
      "flos": 24315941278080.0,
      "grad_norm": 2.9163320269248447,
      "language_loss": 0.83466953,
      "learning_rate": 2.592345371717141e-06,
      "loss": 0.8568418,
      "num_input_tokens_seen": 75700940,
      "step": 3511,
      "time_per_iteration": 2.6041650772094727
    },
    {
      "auxiliary_loss_clip": 0.01195724,
      "auxiliary_loss_mlp": 0.01030351,
      "balance_loss_clip": 1.06015122,
      "balance_loss_mlp": 1.02167225,
      "epoch": 0.42229303192448747,
      "flos": 17092474352640.0,
      "grad_norm": 2.27681589193895,
      "language_loss": 0.72231472,
      "learning_rate": 2.591601304182448e-06,
      "loss": 0.7445755,
      "num_input_tokens_seen": 75718910,
      "step": 3512,
      "time_per_iteration": 2.633512496948242
    },
    {
      "auxiliary_loss_clip": 0.01196178,
      "auxiliary_loss_mlp": 0.01027396,
      "balance_loss_clip": 1.05603588,
      "balance_loss_mlp": 1.01965356,
      "epoch": 0.4224132748151266,
      "flos": 22784530878720.0,
      "grad_norm": 2.1941140656864935,
      "language_loss": 0.79550272,
      "learning_rate": 2.5908571469023067e-06,
      "loss": 0.81773841,
      "num_input_tokens_seen": 75738395,
      "step": 3513,
      "time_per_iteration": 2.6818490028381348
    },
    {
      "auxiliary_loss_clip": 0.01190866,
      "auxiliary_loss_mlp": 0.01028639,
      "balance_loss_clip": 1.05575693,
      "balance_loss_mlp": 1.02008033,
      "epoch": 0.42253351770576564,
      "flos": 17819090576640.0,
      "grad_norm": 2.3300721075615174,
      "language_loss": 0.76065385,
      "learning_rate": 2.5901128999896067e-06,
      "loss": 0.78284889,
      "num_input_tokens_seen": 75753825,
      "step": 3514,
      "time_per_iteration": 3.5083529949188232
    },
    {
      "auxiliary_loss_clip": 0.01192094,
      "auxiliary_loss_mlp": 0.01027591,
      "balance_loss_clip": 1.05572891,
      "balance_loss_mlp": 1.01943088,
      "epoch": 0.42265376059640475,
      "flos": 28512390286080.0,
      "grad_norm": 1.5771881223979238,
      "language_loss": 0.68068945,
      "learning_rate": 2.5893685635572487e-06,
      "loss": 0.70288622,
      "num_input_tokens_seen": 75774675,
      "step": 3515,
      "time_per_iteration": 2.7828328609466553
    },
    {
      "auxiliary_loss_clip": 0.01195628,
      "auxiliary_loss_mlp": 0.01028955,
      "balance_loss_clip": 1.0543412,
      "balance_loss_mlp": 1.019979,
      "epoch": 0.4227740034870438,
      "flos": 16253349753600.0,
      "grad_norm": 1.9078150009347423,
      "language_loss": 0.69155896,
      "learning_rate": 2.5886241377181483e-06,
      "loss": 0.71380478,
      "num_input_tokens_seen": 75793545,
      "step": 3516,
      "time_per_iteration": 2.6116726398468018
    },
    {
      "auxiliary_loss_clip": 0.01197351,
      "auxiliary_loss_mlp": 0.01029038,
      "balance_loss_clip": 1.05565059,
      "balance_loss_mlp": 1.01954341,
      "epoch": 0.4228942463776829,
      "flos": 25295691623040.0,
      "grad_norm": 1.8160885666786755,
      "language_loss": 0.81567299,
      "learning_rate": 2.587879622585234e-06,
      "loss": 0.83793694,
      "num_input_tokens_seen": 75812145,
      "step": 3517,
      "time_per_iteration": 2.686466932296753
    },
    {
      "auxiliary_loss_clip": 0.01193887,
      "auxiliary_loss_mlp": 0.01036459,
      "balance_loss_clip": 1.05686283,
      "balance_loss_mlp": 1.02769732,
      "epoch": 0.423014489268322,
      "flos": 26395779507840.0,
      "grad_norm": 3.571213188772891,
      "language_loss": 0.76223707,
      "learning_rate": 2.5871350182714486e-06,
      "loss": 0.78454053,
      "num_input_tokens_seen": 75833025,
      "step": 3518,
      "time_per_iteration": 2.6628215312957764
    },
    {
      "auxiliary_loss_clip": 0.01189198,
      "auxiliary_loss_mlp": 0.01032909,
      "balance_loss_clip": 1.05561554,
      "balance_loss_mlp": 1.02450466,
      "epoch": 0.4231347321589611,
      "flos": 17274002711040.0,
      "grad_norm": 2.5160241824805447,
      "language_loss": 0.80646574,
      "learning_rate": 2.586390324889748e-06,
      "loss": 0.82868677,
      "num_input_tokens_seen": 75848925,
      "step": 3519,
      "time_per_iteration": 3.5418996810913086
    },
    {
      "auxiliary_loss_clip": 0.01190419,
      "auxiliary_loss_mlp": 0.01030778,
      "balance_loss_clip": 1.05515373,
      "balance_loss_mlp": 1.02267754,
      "epoch": 0.4232549750496002,
      "flos": 22999635475200.0,
      "grad_norm": 2.238842352079369,
      "language_loss": 0.67748624,
      "learning_rate": 2.5856455425531003e-06,
      "loss": 0.69969821,
      "num_input_tokens_seen": 75870400,
      "step": 3520,
      "time_per_iteration": 2.6891794204711914
    },
    {
      "auxiliary_loss_clip": 0.0119376,
      "auxiliary_loss_mlp": 0.01022987,
      "balance_loss_clip": 1.05558276,
      "balance_loss_mlp": 1.01448107,
      "epoch": 0.4233752179402393,
      "flos": 21248343970560.0,
      "grad_norm": 1.779714619744325,
      "language_loss": 0.80517566,
      "learning_rate": 2.5849006713744902e-06,
      "loss": 0.82734311,
      "num_input_tokens_seen": 75889195,
      "step": 3521,
      "time_per_iteration": 2.6267378330230713
    },
    {
      "auxiliary_loss_clip": 0.01194235,
      "auxiliary_loss_mlp": 0.01027183,
      "balance_loss_clip": 1.05303526,
      "balance_loss_mlp": 1.01845121,
      "epoch": 0.42349546083087836,
      "flos": 20704297599360.0,
      "grad_norm": 3.0172683595305907,
      "language_loss": 0.73481691,
      "learning_rate": 2.5841557114669135e-06,
      "loss": 0.75703108,
      "num_input_tokens_seen": 75906055,
      "step": 3522,
      "time_per_iteration": 2.6368777751922607
    },
    {
      "auxiliary_loss_clip": 0.01195064,
      "auxiliary_loss_mlp": 0.01030011,
      "balance_loss_clip": 1.05549741,
      "balance_loss_mlp": 1.02036726,
      "epoch": 0.42361570372151747,
      "flos": 18585065128320.0,
      "grad_norm": 3.1239641566281717,
      "language_loss": 0.6768887,
      "learning_rate": 2.58341066294338e-06,
      "loss": 0.69913948,
      "num_input_tokens_seen": 75922720,
      "step": 3523,
      "time_per_iteration": 2.541163921356201
    },
    {
      "auxiliary_loss_clip": 0.01206602,
      "auxiliary_loss_mlp": 0.01209303,
      "balance_loss_clip": 1.05166292,
      "balance_loss_mlp": 1.00064421,
      "epoch": 0.4237359466121566,
      "flos": 20959478795520.0,
      "grad_norm": 19.22218011697924,
      "language_loss": 0.85792387,
      "learning_rate": 2.5826655259169124e-06,
      "loss": 0.88208294,
      "num_input_tokens_seen": 75941375,
      "step": 3524,
      "time_per_iteration": 2.661229372024536
    },
    {
      "auxiliary_loss_clip": 0.01193824,
      "auxiliary_loss_mlp": 0.01032576,
      "balance_loss_clip": 1.05847788,
      "balance_loss_mlp": 1.02368271,
      "epoch": 0.42385618950279563,
      "flos": 18038181582720.0,
      "grad_norm": 1.8680353110010794,
      "language_loss": 0.90655851,
      "learning_rate": 2.5819203005005475e-06,
      "loss": 0.92882252,
      "num_input_tokens_seen": 75958710,
      "step": 3525,
      "time_per_iteration": 2.567401170730591
    },
    {
      "auxiliary_loss_clip": 0.01190786,
      "auxiliary_loss_mlp": 0.01028908,
      "balance_loss_clip": 1.05383992,
      "balance_loss_mlp": 1.02058733,
      "epoch": 0.42397643239343474,
      "flos": 23769129559680.0,
      "grad_norm": 1.5974122512100264,
      "language_loss": 0.78458649,
      "learning_rate": 2.581174986807336e-06,
      "loss": 0.80678338,
      "num_input_tokens_seen": 75978945,
      "step": 3526,
      "time_per_iteration": 2.6684892177581787
    },
    {
      "auxiliary_loss_clip": 0.01185891,
      "auxiliary_loss_mlp": 0.01209318,
      "balance_loss_clip": 1.05417037,
      "balance_loss_mlp": 1.00061929,
      "epoch": 0.42409667528407385,
      "flos": 16545088016640.0,
      "grad_norm": 2.315555052555759,
      "language_loss": 0.9161939,
      "learning_rate": 2.580429584950341e-06,
      "loss": 0.94014597,
      "num_input_tokens_seen": 75994695,
      "step": 3527,
      "time_per_iteration": 2.530042886734009
    },
    {
      "auxiliary_loss_clip": 0.01203821,
      "auxiliary_loss_mlp": 0.01028162,
      "balance_loss_clip": 1.05275321,
      "balance_loss_mlp": 1.01822031,
      "epoch": 0.4242169181747129,
      "flos": 16034186920320.0,
      "grad_norm": 2.3762612307573097,
      "language_loss": 0.65939128,
      "learning_rate": 2.5796840950426397e-06,
      "loss": 0.68171114,
      "num_input_tokens_seen": 76011780,
      "step": 3528,
      "time_per_iteration": 2.580244779586792
    },
    {
      "auxiliary_loss_clip": 0.01183014,
      "auxiliary_loss_mlp": 0.01030018,
      "balance_loss_clip": 1.05174136,
      "balance_loss_mlp": 1.02180409,
      "epoch": 0.424337161065352,
      "flos": 20084012611200.0,
      "grad_norm": 1.7867694032177883,
      "language_loss": 0.65883696,
      "learning_rate": 2.578938517197322e-06,
      "loss": 0.68096727,
      "num_input_tokens_seen": 76029875,
      "step": 3529,
      "time_per_iteration": 2.618537187576294
    },
    {
      "auxiliary_loss_clip": 0.01183914,
      "auxiliary_loss_mlp": 0.01029899,
      "balance_loss_clip": 1.05041718,
      "balance_loss_mlp": 1.02107716,
      "epoch": 0.4244574039559911,
      "flos": 23878369797120.0,
      "grad_norm": 5.10337913240849,
      "language_loss": 0.62429613,
      "learning_rate": 2.5781928515274916e-06,
      "loss": 0.64643419,
      "num_input_tokens_seen": 76048595,
      "step": 3530,
      "time_per_iteration": 2.697225332260132
    },
    {
      "auxiliary_loss_clip": 0.01198881,
      "auxiliary_loss_mlp": 0.01029588,
      "balance_loss_clip": 1.05844808,
      "balance_loss_mlp": 1.0210948,
      "epoch": 0.4245776468466302,
      "flos": 17565920542080.0,
      "grad_norm": 2.122936439794338,
      "language_loss": 0.68111736,
      "learning_rate": 2.577447098146265e-06,
      "loss": 0.70340204,
      "num_input_tokens_seen": 76065770,
      "step": 3531,
      "time_per_iteration": 2.5814154148101807
    },
    {
      "auxiliary_loss_clip": 0.01201527,
      "auxiliary_loss_mlp": 0.01036378,
      "balance_loss_clip": 1.05376494,
      "balance_loss_mlp": 1.02774692,
      "epoch": 0.4246978897372693,
      "flos": 27776256958080.0,
      "grad_norm": 2.256637777900892,
      "language_loss": 0.7889545,
      "learning_rate": 2.5767012571667724e-06,
      "loss": 0.81133354,
      "num_input_tokens_seen": 76085250,
      "step": 3532,
      "time_per_iteration": 2.6927154064178467
    },
    {
      "auxiliary_loss_clip": 0.01194325,
      "auxiliary_loss_mlp": 0.01026889,
      "balance_loss_clip": 1.0523665,
      "balance_loss_mlp": 1.01722693,
      "epoch": 0.42481813262790835,
      "flos": 15596615439360.0,
      "grad_norm": 2.018454888134479,
      "language_loss": 0.68083894,
      "learning_rate": 2.5759553287021587e-06,
      "loss": 0.70305109,
      "num_input_tokens_seen": 76103580,
      "step": 3533,
      "time_per_iteration": 2.578737497329712
    },
    {
      "auxiliary_loss_clip": 0.01195252,
      "auxiliary_loss_mlp": 0.01030115,
      "balance_loss_clip": 1.05524683,
      "balance_loss_mlp": 1.02061415,
      "epoch": 0.42493837551854746,
      "flos": 23951088881280.0,
      "grad_norm": 1.8526635912391118,
      "language_loss": 0.77536547,
      "learning_rate": 2.5752093128655786e-06,
      "loss": 0.79761922,
      "num_input_tokens_seen": 76121825,
      "step": 3534,
      "time_per_iteration": 2.6503570079803467
    },
    {
      "auxiliary_loss_clip": 0.01186362,
      "auxiliary_loss_mlp": 0.01026275,
      "balance_loss_clip": 1.04937398,
      "balance_loss_mlp": 1.01730418,
      "epoch": 0.4250586184091866,
      "flos": 20813466009600.0,
      "grad_norm": 1.9291209185971432,
      "language_loss": 0.7411167,
      "learning_rate": 2.574463209770204e-06,
      "loss": 0.76324308,
      "num_input_tokens_seen": 76141140,
      "step": 3535,
      "time_per_iteration": 3.5264041423797607
    },
    {
      "auxiliary_loss_clip": 0.01194635,
      "auxiliary_loss_mlp": 0.01031275,
      "balance_loss_clip": 1.04886508,
      "balance_loss_mlp": 1.02223253,
      "epoch": 0.42517886129982563,
      "flos": 30371018607360.0,
      "grad_norm": 1.6833452058581815,
      "language_loss": 0.79558522,
      "learning_rate": 2.5737170195292165e-06,
      "loss": 0.81784427,
      "num_input_tokens_seen": 76164475,
      "step": 3536,
      "time_per_iteration": 2.7336583137512207
    },
    {
      "auxiliary_loss_clip": 0.01197401,
      "auxiliary_loss_mlp": 0.01030796,
      "balance_loss_clip": 1.05008507,
      "balance_loss_mlp": 1.02146745,
      "epoch": 0.42529910419046474,
      "flos": 20080636732800.0,
      "grad_norm": 3.0476086209693722,
      "language_loss": 0.78264701,
      "learning_rate": 2.572970742255814e-06,
      "loss": 0.80492902,
      "num_input_tokens_seen": 76182965,
      "step": 3537,
      "time_per_iteration": 3.549419403076172
    },
    {
      "auxiliary_loss_clip": 0.01193082,
      "auxiliary_loss_mlp": 0.01028944,
      "balance_loss_clip": 1.05646157,
      "balance_loss_mlp": 1.02059317,
      "epoch": 0.42541934708110385,
      "flos": 22632448694400.0,
      "grad_norm": 3.0115117971809675,
      "language_loss": 0.81576884,
      "learning_rate": 2.5722243780632046e-06,
      "loss": 0.83798909,
      "num_input_tokens_seen": 76201230,
      "step": 3538,
      "time_per_iteration": 2.622464895248413
    },
    {
      "auxiliary_loss_clip": 0.01116684,
      "auxiliary_loss_mlp": 0.0100428,
      "balance_loss_clip": 1.02370572,
      "balance_loss_mlp": 1.00324869,
      "epoch": 0.4255395899717429,
      "flos": 66200676186240.0,
      "grad_norm": 0.7615106738951026,
      "language_loss": 0.60464704,
      "learning_rate": 2.5714779270646125e-06,
      "loss": 0.62585664,
      "num_input_tokens_seen": 76262000,
      "step": 3539,
      "time_per_iteration": 3.153671979904175
    },
    {
      "auxiliary_loss_clip": 0.01200145,
      "auxiliary_loss_mlp": 0.01208984,
      "balance_loss_clip": 1.05466962,
      "balance_loss_mlp": 1.00060821,
      "epoch": 0.425659832862382,
      "flos": 17931814433280.0,
      "grad_norm": 2.825517170048629,
      "language_loss": 0.77723074,
      "learning_rate": 2.5707313893732735e-06,
      "loss": 0.80132198,
      "num_input_tokens_seen": 76280540,
      "step": 3540,
      "time_per_iteration": 3.5219221115112305
    },
    {
      "auxiliary_loss_clip": 0.01196872,
      "auxiliary_loss_mlp": 0.01025623,
      "balance_loss_clip": 1.04261029,
      "balance_loss_mlp": 1.01668835,
      "epoch": 0.4257800757530211,
      "flos": 24022550989440.0,
      "grad_norm": 1.9630739838301774,
      "language_loss": 0.77175081,
      "learning_rate": 2.5699847651024364e-06,
      "loss": 0.79397571,
      "num_input_tokens_seen": 76301180,
      "step": 3541,
      "time_per_iteration": 2.7465150356292725
    },
    {
      "auxiliary_loss_clip": 0.01194377,
      "auxiliary_loss_mlp": 0.01031983,
      "balance_loss_clip": 1.05805373,
      "balance_loss_mlp": 1.02337646,
      "epoch": 0.4259003186436602,
      "flos": 23696015425920.0,
      "grad_norm": 2.0321335394965026,
      "language_loss": 0.76343322,
      "learning_rate": 2.5692380543653627e-06,
      "loss": 0.78569686,
      "num_input_tokens_seen": 76319335,
      "step": 3542,
      "time_per_iteration": 2.6495893001556396
    },
    {
      "auxiliary_loss_clip": 0.01199436,
      "auxiliary_loss_mlp": 0.01209872,
      "balance_loss_clip": 1.05718791,
      "balance_loss_mlp": 1.00076151,
      "epoch": 0.4260205615342993,
      "flos": 15259772672640.0,
      "grad_norm": 2.1586026494973947,
      "language_loss": 0.69884837,
      "learning_rate": 2.5684912572753293e-06,
      "loss": 0.72294152,
      "num_input_tokens_seen": 76335010,
      "step": 3543,
      "time_per_iteration": 2.6247150897979736
    },
    {
      "auxiliary_loss_clip": 0.01189475,
      "auxiliary_loss_mlp": 0.0102453,
      "balance_loss_clip": 1.05713892,
      "balance_loss_mlp": 1.01606631,
      "epoch": 0.4261408044249384,
      "flos": 30665306736000.0,
      "grad_norm": 11.873067252823793,
      "language_loss": 0.8396588,
      "learning_rate": 2.5677443739456245e-06,
      "loss": 0.86179882,
      "num_input_tokens_seen": 76356670,
      "step": 3544,
      "time_per_iteration": 2.703394651412964
    },
    {
      "auxiliary_loss_clip": 0.01196663,
      "auxiliary_loss_mlp": 0.01028519,
      "balance_loss_clip": 1.05553627,
      "balance_loss_mlp": 1.01999521,
      "epoch": 0.42626104731557746,
      "flos": 23257905240960.0,
      "grad_norm": 2.46476247640177,
      "language_loss": 0.79840672,
      "learning_rate": 2.5669974044895495e-06,
      "loss": 0.82065856,
      "num_input_tokens_seen": 76373065,
      "step": 3545,
      "time_per_iteration": 2.60493803024292
    },
    {
      "auxiliary_loss_clip": 0.01206413,
      "auxiliary_loss_mlp": 0.0102525,
      "balance_loss_clip": 1.05194461,
      "balance_loss_mlp": 1.01640511,
      "epoch": 0.42638129020621657,
      "flos": 25884770670720.0,
      "grad_norm": 1.793461639047948,
      "language_loss": 0.79336649,
      "learning_rate": 2.5662503490204187e-06,
      "loss": 0.81568313,
      "num_input_tokens_seen": 76393230,
      "step": 3546,
      "time_per_iteration": 3.6414101123809814
    },
    {
      "auxiliary_loss_clip": 0.01192657,
      "auxiliary_loss_mlp": 0.01025827,
      "balance_loss_clip": 1.05117106,
      "balance_loss_mlp": 1.01739335,
      "epoch": 0.4265015330968556,
      "flos": 26502362138880.0,
      "grad_norm": 2.0931832474207166,
      "language_loss": 0.76087022,
      "learning_rate": 2.5655032076515603e-06,
      "loss": 0.78305507,
      "num_input_tokens_seen": 76412555,
      "step": 3547,
      "time_per_iteration": 2.6803112030029297
    },
    {
      "auxiliary_loss_clip": 0.01196003,
      "auxiliary_loss_mlp": 0.01027363,
      "balance_loss_clip": 1.05167377,
      "balance_loss_mlp": 1.01860702,
      "epoch": 0.42662177598749473,
      "flos": 24389522288640.0,
      "grad_norm": 2.2744636272567282,
      "language_loss": 0.82145095,
      "learning_rate": 2.5647559804963155e-06,
      "loss": 0.84368461,
      "num_input_tokens_seen": 76432485,
      "step": 3548,
      "time_per_iteration": 2.623753786087036
    },
    {
      "auxiliary_loss_clip": 0.01209152,
      "auxiliary_loss_mlp": 0.01032414,
      "balance_loss_clip": 1.05228066,
      "balance_loss_mlp": 1.02412271,
      "epoch": 0.42674201887813384,
      "flos": 23148629089920.0,
      "grad_norm": 1.9963706955157514,
      "language_loss": 0.78451818,
      "learning_rate": 2.5640086676680364e-06,
      "loss": 0.80693388,
      "num_input_tokens_seen": 76453980,
      "step": 3549,
      "time_per_iteration": 2.723564863204956
    },
    {
      "auxiliary_loss_clip": 0.01196896,
      "auxiliary_loss_mlp": 0.01028488,
      "balance_loss_clip": 1.05690396,
      "balance_loss_mlp": 1.01923776,
      "epoch": 0.4268622617687729,
      "flos": 21689614552320.0,
      "grad_norm": 2.756133361693614,
      "language_loss": 0.80897653,
      "learning_rate": 2.5632612692800923e-06,
      "loss": 0.8312304,
      "num_input_tokens_seen": 76473045,
      "step": 3550,
      "time_per_iteration": 2.5851776599884033
    },
    {
      "auxiliary_loss_clip": 0.0119703,
      "auxiliary_loss_mlp": 0.01036125,
      "balance_loss_clip": 1.04920971,
      "balance_loss_mlp": 1.02676141,
      "epoch": 0.426982504659412,
      "flos": 23440151871360.0,
      "grad_norm": 3.408331202991418,
      "language_loss": 0.75386292,
      "learning_rate": 2.5625137854458603e-06,
      "loss": 0.77619445,
      "num_input_tokens_seen": 76492060,
      "step": 3551,
      "time_per_iteration": 2.7711734771728516
    },
    {
      "auxiliary_loss_clip": 0.01198094,
      "auxiliary_loss_mlp": 0.01030588,
      "balance_loss_clip": 1.05435061,
      "balance_loss_mlp": 1.02224946,
      "epoch": 0.4271027475500511,
      "flos": 18916556768640.0,
      "grad_norm": 1.9670401641201516,
      "language_loss": 0.8016777,
      "learning_rate": 2.561766216278735e-06,
      "loss": 0.82396448,
      "num_input_tokens_seen": 76509655,
      "step": 3552,
      "time_per_iteration": 2.624176263809204
    },
    {
      "auxiliary_loss_clip": 0.01197925,
      "auxiliary_loss_mlp": 0.01027257,
      "balance_loss_clip": 1.04998624,
      "balance_loss_mlp": 1.01809549,
      "epoch": 0.4272229904406902,
      "flos": 26870554500480.0,
      "grad_norm": 2.346499600269311,
      "language_loss": 0.81330329,
      "learning_rate": 2.561018561892121e-06,
      "loss": 0.83555514,
      "num_input_tokens_seen": 76528795,
      "step": 3553,
      "time_per_iteration": 2.6972532272338867
    },
    {
      "auxiliary_loss_clip": 0.01192968,
      "auxiliary_loss_mlp": 0.01032259,
      "balance_loss_clip": 1.05015635,
      "balance_loss_mlp": 1.02395606,
      "epoch": 0.4273432333313293,
      "flos": 23951376190080.0,
      "grad_norm": 1.5936455081420426,
      "language_loss": 0.76734138,
      "learning_rate": 2.5602708223994363e-06,
      "loss": 0.7895937,
      "num_input_tokens_seen": 76550660,
      "step": 3554,
      "time_per_iteration": 2.667886257171631
    },
    {
      "auxiliary_loss_clip": 0.01196031,
      "auxiliary_loss_mlp": 0.01026937,
      "balance_loss_clip": 1.04633737,
      "balance_loss_mlp": 1.0184375,
      "epoch": 0.4274634762219684,
      "flos": 29570354496000.0,
      "grad_norm": 2.2795022100270623,
      "language_loss": 0.67868948,
      "learning_rate": 2.559522997914115e-06,
      "loss": 0.70091915,
      "num_input_tokens_seen": 76570240,
      "step": 3555,
      "time_per_iteration": 2.7325401306152344
    },
    {
      "auxiliary_loss_clip": 0.01188052,
      "auxiliary_loss_mlp": 0.01030087,
      "balance_loss_clip": 1.05729997,
      "balance_loss_mlp": 1.0219332,
      "epoch": 0.42758371911260745,
      "flos": 21434146047360.0,
      "grad_norm": 2.4129264096298337,
      "language_loss": 0.84388614,
      "learning_rate": 2.558775088549599e-06,
      "loss": 0.86606753,
      "num_input_tokens_seen": 76589820,
      "step": 3556,
      "time_per_iteration": 2.532132863998413
    },
    {
      "auxiliary_loss_clip": 0.01198315,
      "auxiliary_loss_mlp": 0.01029045,
      "balance_loss_clip": 1.05412054,
      "balance_loss_mlp": 1.02031922,
      "epoch": 0.42770396200324656,
      "flos": 14752822072320.0,
      "grad_norm": 2.387775770066941,
      "language_loss": 0.66763186,
      "learning_rate": 2.5580270944193467e-06,
      "loss": 0.68990541,
      "num_input_tokens_seen": 76606640,
      "step": 3557,
      "time_per_iteration": 2.587329626083374
    },
    {
      "auxiliary_loss_clip": 0.01088842,
      "auxiliary_loss_mlp": 0.01001128,
      "balance_loss_clip": 1.02234793,
      "balance_loss_mlp": 1.00015664,
      "epoch": 0.4278242048938857,
      "flos": 70654712601600.0,
      "grad_norm": 0.7388554318121621,
      "language_loss": 0.55462956,
      "learning_rate": 2.557279015636827e-06,
      "loss": 0.57552922,
      "num_input_tokens_seen": 76667050,
      "step": 3558,
      "time_per_iteration": 3.115216016769409
    },
    {
      "auxiliary_loss_clip": 0.01091669,
      "auxiliary_loss_mlp": 0.01002582,
      "balance_loss_clip": 1.02317214,
      "balance_loss_mlp": 1.00174201,
      "epoch": 0.42794444778452473,
      "flos": 69366165033600.0,
      "grad_norm": 0.7681805525411383,
      "language_loss": 0.61204612,
      "learning_rate": 2.5565308523155245e-06,
      "loss": 0.63298869,
      "num_input_tokens_seen": 76726650,
      "step": 3559,
      "time_per_iteration": 3.1154026985168457
    },
    {
      "auxiliary_loss_clip": 0.01192968,
      "auxiliary_loss_mlp": 0.01028371,
      "balance_loss_clip": 1.04848528,
      "balance_loss_mlp": 1.01981807,
      "epoch": 0.42806469067516384,
      "flos": 18215328481920.0,
      "grad_norm": 3.4548074804058357,
      "language_loss": 0.81825519,
      "learning_rate": 2.5557826045689336e-06,
      "loss": 0.84046865,
      "num_input_tokens_seen": 76742890,
      "step": 3560,
      "time_per_iteration": 2.663248300552368
    },
    {
      "auxiliary_loss_clip": 0.01100379,
      "auxiliary_loss_mlp": 0.01004268,
      "balance_loss_clip": 1.02121377,
      "balance_loss_mlp": 1.00302243,
      "epoch": 0.4281849335658029,
      "flos": 54535814432640.0,
      "grad_norm": 0.8320812876360821,
      "language_loss": 0.58788753,
      "learning_rate": 2.5550342725105643e-06,
      "loss": 0.60893393,
      "num_input_tokens_seen": 76801055,
      "step": 3561,
      "time_per_iteration": 3.1533162593841553
    },
    {
      "auxiliary_loss_clip": 0.01195866,
      "auxiliary_loss_mlp": 0.01033614,
      "balance_loss_clip": 1.05820322,
      "balance_loss_mlp": 1.02509642,
      "epoch": 0.428305176456442,
      "flos": 17274828723840.0,
      "grad_norm": 1.6825216527798388,
      "language_loss": 0.81184328,
      "learning_rate": 2.554285856253937e-06,
      "loss": 0.83413804,
      "num_input_tokens_seen": 76819890,
      "step": 3562,
      "time_per_iteration": 3.6037349700927734
    },
    {
      "auxiliary_loss_clip": 0.01192728,
      "auxiliary_loss_mlp": 0.01032945,
      "balance_loss_clip": 1.0543493,
      "balance_loss_mlp": 1.02394485,
      "epoch": 0.4284254193470811,
      "flos": 26359509749760.0,
      "grad_norm": 1.7857792544158946,
      "language_loss": 0.77494395,
      "learning_rate": 2.5535373559125855e-06,
      "loss": 0.79720068,
      "num_input_tokens_seen": 76840255,
      "step": 3563,
      "time_per_iteration": 2.6781184673309326
    },
    {
      "auxiliary_loss_clip": 0.01202893,
      "auxiliary_loss_mlp": 0.01024522,
      "balance_loss_clip": 1.04502869,
      "balance_loss_mlp": 1.01564717,
      "epoch": 0.42854566223772017,
      "flos": 29714248379520.0,
      "grad_norm": 1.6203941937812927,
      "language_loss": 0.82124549,
      "learning_rate": 2.552788771600057e-06,
      "loss": 0.84351963,
      "num_input_tokens_seen": 76860565,
      "step": 3564,
      "time_per_iteration": 3.6328980922698975
    },
    {
      "auxiliary_loss_clip": 0.01200218,
      "auxiliary_loss_mlp": 0.01037239,
      "balance_loss_clip": 1.05363822,
      "balance_loss_mlp": 1.02782142,
      "epoch": 0.4286659051283593,
      "flos": 22018161277440.0,
      "grad_norm": 1.8532504103862222,
      "language_loss": 0.82304537,
      "learning_rate": 2.5520401034299118e-06,
      "loss": 0.84541988,
      "num_input_tokens_seen": 76878325,
      "step": 3565,
      "time_per_iteration": 2.742218255996704
    },
    {
      "auxiliary_loss_clip": 0.01193442,
      "auxiliary_loss_mlp": 0.01030458,
      "balance_loss_clip": 1.0536859,
      "balance_loss_mlp": 1.02147007,
      "epoch": 0.4287861480189984,
      "flos": 13334422838400.0,
      "grad_norm": 2.3926929773694834,
      "language_loss": 0.88137221,
      "learning_rate": 2.551291351515722e-06,
      "loss": 0.90361124,
      "num_input_tokens_seen": 76895340,
      "step": 3566,
      "time_per_iteration": 2.5964431762695312
    },
    {
      "auxiliary_loss_clip": 0.01191146,
      "auxiliary_loss_mlp": 0.01209862,
      "balance_loss_clip": 1.04679966,
      "balance_loss_mlp": 1.00073886,
      "epoch": 0.42890639090963745,
      "flos": 26651535321600.0,
      "grad_norm": 1.7546750294356133,
      "language_loss": 0.8569628,
      "learning_rate": 2.5505425159710726e-06,
      "loss": 0.88097286,
      "num_input_tokens_seen": 76915150,
      "step": 3567,
      "time_per_iteration": 3.7046165466308594
    },
    {
      "auxiliary_loss_clip": 0.01201386,
      "auxiliary_loss_mlp": 0.0120967,
      "balance_loss_clip": 1.05127358,
      "balance_loss_mlp": 1.00076318,
      "epoch": 0.42902663380027656,
      "flos": 24055768091520.0,
      "grad_norm": 2.412118263093021,
      "language_loss": 0.82857752,
      "learning_rate": 2.549793596909561e-06,
      "loss": 0.85268807,
      "num_input_tokens_seen": 76933770,
      "step": 3568,
      "time_per_iteration": 2.712475299835205
    },
    {
      "auxiliary_loss_clip": 0.01190206,
      "auxiliary_loss_mlp": 0.01026519,
      "balance_loss_clip": 1.05307531,
      "balance_loss_mlp": 1.01747131,
      "epoch": 0.42914687669091567,
      "flos": 15632561975040.0,
      "grad_norm": 2.2240025809121855,
      "language_loss": 0.66187996,
      "learning_rate": 2.5490445944447976e-06,
      "loss": 0.68404722,
      "num_input_tokens_seen": 76952265,
      "step": 3569,
      "time_per_iteration": 2.6088125705718994
    },
    {
      "auxiliary_loss_clip": 0.01191412,
      "auxiliary_loss_mlp": 0.01023827,
      "balance_loss_clip": 1.053262,
      "balance_loss_mlp": 1.01514852,
      "epoch": 0.4292671195815547,
      "flos": 31467802440960.0,
      "grad_norm": 3.2396128620897993,
      "language_loss": 0.65090454,
      "learning_rate": 2.548295508690406e-06,
      "loss": 0.67305696,
      "num_input_tokens_seen": 76973560,
      "step": 3570,
      "time_per_iteration": 2.703958034515381
    },
    {
      "auxiliary_loss_clip": 0.01194492,
      "auxiliary_loss_mlp": 0.01025466,
      "balance_loss_clip": 1.05309594,
      "balance_loss_mlp": 1.01667452,
      "epoch": 0.42938736247219383,
      "flos": 30257756046720.0,
      "grad_norm": 1.7083987086410524,
      "language_loss": 0.76606643,
      "learning_rate": 2.5475463397600217e-06,
      "loss": 0.78826606,
      "num_input_tokens_seen": 76993640,
      "step": 3571,
      "time_per_iteration": 2.6605734825134277
    },
    {
      "auxiliary_loss_clip": 0.01195948,
      "auxiliary_loss_mlp": 0.01026807,
      "balance_loss_clip": 1.05848384,
      "balance_loss_mlp": 1.01779461,
      "epoch": 0.42950760536283294,
      "flos": 29349683291520.0,
      "grad_norm": 2.105245613306248,
      "language_loss": 0.77243191,
      "learning_rate": 2.546797087767293e-06,
      "loss": 0.79465944,
      "num_input_tokens_seen": 77013765,
      "step": 3572,
      "time_per_iteration": 3.5892436504364014
    },
    {
      "auxiliary_loss_clip": 0.01195855,
      "auxiliary_loss_mlp": 0.01032971,
      "balance_loss_clip": 1.04926324,
      "balance_loss_mlp": 1.02398229,
      "epoch": 0.429627848253472,
      "flos": 26869943969280.0,
      "grad_norm": 1.6940716490100265,
      "language_loss": 0.87501085,
      "learning_rate": 2.546047752825881e-06,
      "loss": 0.89729905,
      "num_input_tokens_seen": 77034370,
      "step": 3573,
      "time_per_iteration": 2.671551465988159
    },
    {
      "auxiliary_loss_clip": 0.01204393,
      "auxiliary_loss_mlp": 0.01028981,
      "balance_loss_clip": 1.05070043,
      "balance_loss_mlp": 1.02023125,
      "epoch": 0.4297480911441111,
      "flos": 13881270470400.0,
      "grad_norm": 2.9637628924130004,
      "language_loss": 0.93809617,
      "learning_rate": 2.5452983350494595e-06,
      "loss": 0.96042991,
      "num_input_tokens_seen": 77049925,
      "step": 3574,
      "time_per_iteration": 2.5526034832000732
    },
    {
      "auxiliary_loss_clip": 0.01192985,
      "auxiliary_loss_mlp": 0.01208982,
      "balance_loss_clip": 1.05494893,
      "balance_loss_mlp": 1.0008018,
      "epoch": 0.4298683340347502,
      "flos": 20741141975040.0,
      "grad_norm": 2.3424594918120767,
      "language_loss": 0.65895551,
      "learning_rate": 2.544548834551713e-06,
      "loss": 0.68297529,
      "num_input_tokens_seen": 77068930,
      "step": 3575,
      "time_per_iteration": 2.5677731037139893
    },
    {
      "auxiliary_loss_clip": 0.01194072,
      "auxiliary_loss_mlp": 0.01209096,
      "balance_loss_clip": 1.05074954,
      "balance_loss_mlp": 1.00076723,
      "epoch": 0.4299885769253893,
      "flos": 20882126856960.0,
      "grad_norm": 2.4881143074513514,
      "language_loss": 0.94814289,
      "learning_rate": 2.5437992514463424e-06,
      "loss": 0.97217458,
      "num_input_tokens_seen": 77082255,
      "step": 3576,
      "time_per_iteration": 2.586463212966919
    },
    {
      "auxiliary_loss_clip": 0.01193579,
      "auxiliary_loss_mlp": 0.01031652,
      "balance_loss_clip": 1.05549085,
      "balance_loss_mlp": 1.02237809,
      "epoch": 0.4301088198160284,
      "flos": 25484618183040.0,
      "grad_norm": 1.7738055696834532,
      "language_loss": 0.88170981,
      "learning_rate": 2.5430495858470565e-06,
      "loss": 0.90396214,
      "num_input_tokens_seen": 77101725,
      "step": 3577,
      "time_per_iteration": 2.6463334560394287
    },
    {
      "auxiliary_loss_clip": 0.01191196,
      "auxiliary_loss_mlp": 0.01028581,
      "balance_loss_clip": 1.05552387,
      "balance_loss_mlp": 1.01984859,
      "epoch": 0.43022906270666744,
      "flos": 18259427404800.0,
      "grad_norm": 2.738129795813348,
      "language_loss": 0.78002989,
      "learning_rate": 2.54229983786758e-06,
      "loss": 0.80222762,
      "num_input_tokens_seen": 77119670,
      "step": 3578,
      "time_per_iteration": 2.57539439201355
    },
    {
      "auxiliary_loss_clip": 0.01192592,
      "auxiliary_loss_mlp": 0.01029893,
      "balance_loss_clip": 1.04958296,
      "balance_loss_mlp": 1.02068424,
      "epoch": 0.43034930559730655,
      "flos": 23399536567680.0,
      "grad_norm": 2.0589609912476994,
      "language_loss": 0.84742463,
      "learning_rate": 2.541550007621651e-06,
      "loss": 0.86964953,
      "num_input_tokens_seen": 77138160,
      "step": 3579,
      "time_per_iteration": 2.663200855255127
    },
    {
      "auxiliary_loss_clip": 0.01190821,
      "auxiliary_loss_mlp": 0.01028178,
      "balance_loss_clip": 1.05572474,
      "balance_loss_mlp": 1.01981497,
      "epoch": 0.43046954848794566,
      "flos": 28184382264960.0,
      "grad_norm": 1.9540050703928564,
      "language_loss": 0.8018229,
      "learning_rate": 2.5408000952230156e-06,
      "loss": 0.82401299,
      "num_input_tokens_seen": 77156950,
      "step": 3580,
      "time_per_iteration": 2.6737704277038574
    },
    {
      "auxiliary_loss_clip": 0.01207555,
      "auxiliary_loss_mlp": 0.01027891,
      "balance_loss_clip": 1.05061531,
      "balance_loss_mlp": 1.0186522,
      "epoch": 0.4305897913785847,
      "flos": 28580476515840.0,
      "grad_norm": 2.218489294781455,
      "language_loss": 0.90463293,
      "learning_rate": 2.5400501007854357e-06,
      "loss": 0.92698741,
      "num_input_tokens_seen": 77176395,
      "step": 3581,
      "time_per_iteration": 2.735945463180542
    },
    {
      "auxiliary_loss_clip": 0.01197469,
      "auxiliary_loss_mlp": 0.01034148,
      "balance_loss_clip": 1.04584634,
      "balance_loss_mlp": 1.02596402,
      "epoch": 0.43071003426922383,
      "flos": 20448721353600.0,
      "grad_norm": 1.9303306307421024,
      "language_loss": 0.75550044,
      "learning_rate": 2.539300024422685e-06,
      "loss": 0.77781659,
      "num_input_tokens_seen": 77194340,
      "step": 3582,
      "time_per_iteration": 2.6749515533447266
    },
    {
      "auxiliary_loss_clip": 0.01101316,
      "auxiliary_loss_mlp": 0.01007394,
      "balance_loss_clip": 1.01888323,
      "balance_loss_mlp": 1.00643396,
      "epoch": 0.43083027715986294,
      "flos": 51997969883520.0,
      "grad_norm": 0.7864777820851984,
      "language_loss": 0.60908538,
      "learning_rate": 2.538549866248549e-06,
      "loss": 0.63017243,
      "num_input_tokens_seen": 77249320,
      "step": 3583,
      "time_per_iteration": 3.049377918243408
    },
    {
      "auxiliary_loss_clip": 0.01192673,
      "auxiliary_loss_mlp": 0.0102681,
      "balance_loss_clip": 1.05246449,
      "balance_loss_mlp": 1.0177505,
      "epoch": 0.430950520050502,
      "flos": 16690885320960.0,
      "grad_norm": 5.2710389668372075,
      "language_loss": 0.81074667,
      "learning_rate": 2.5377996263768274e-06,
      "loss": 0.83294153,
      "num_input_tokens_seen": 77267400,
      "step": 3584,
      "time_per_iteration": 2.599330186843872
    },
    {
      "auxiliary_loss_clip": 0.01189229,
      "auxiliary_loss_mlp": 0.01033177,
      "balance_loss_clip": 1.052423,
      "balance_loss_mlp": 1.02434373,
      "epoch": 0.4310707629411411,
      "flos": 24608433726720.0,
      "grad_norm": 2.0203077355036116,
      "language_loss": 0.68595272,
      "learning_rate": 2.5370493049213293e-06,
      "loss": 0.70817685,
      "num_input_tokens_seen": 77287045,
      "step": 3585,
      "time_per_iteration": 2.6870310306549072
    },
    {
      "auxiliary_loss_clip": 0.0119735,
      "auxiliary_loss_mlp": 0.01030154,
      "balance_loss_clip": 1.04257894,
      "balance_loss_mlp": 1.02094519,
      "epoch": 0.4311910058317802,
      "flos": 26432983019520.0,
      "grad_norm": 2.8887738443443696,
      "language_loss": 0.80281973,
      "learning_rate": 2.536298901995878e-06,
      "loss": 0.82509482,
      "num_input_tokens_seen": 77306255,
      "step": 3586,
      "time_per_iteration": 2.881281614303589
    },
    {
      "auxiliary_loss_clip": 0.01194942,
      "auxiliary_loss_mlp": 0.01027886,
      "balance_loss_clip": 1.0525732,
      "balance_loss_mlp": 1.01942813,
      "epoch": 0.43131124872241927,
      "flos": 25155891889920.0,
      "grad_norm": 1.6963007302916824,
      "language_loss": 0.80069613,
      "learning_rate": 2.535548417714311e-06,
      "loss": 0.82292444,
      "num_input_tokens_seen": 77325555,
      "step": 3587,
      "time_per_iteration": 2.9108355045318604
    },
    {
      "auxiliary_loss_clip": 0.01198329,
      "auxiliary_loss_mlp": 0.01029236,
      "balance_loss_clip": 1.05401373,
      "balance_loss_mlp": 1.02018178,
      "epoch": 0.4314314916130584,
      "flos": 21614812479360.0,
      "grad_norm": 1.685807203408006,
      "language_loss": 0.87124062,
      "learning_rate": 2.534797852190474e-06,
      "loss": 0.89351624,
      "num_input_tokens_seen": 77345735,
      "step": 3588,
      "time_per_iteration": 2.7898404598236084
    },
    {
      "auxiliary_loss_clip": 0.01188518,
      "auxiliary_loss_mlp": 0.01035489,
      "balance_loss_clip": 1.0518049,
      "balance_loss_mlp": 1.02654207,
      "epoch": 0.4315517345036975,
      "flos": 19275016544640.0,
      "grad_norm": 1.9091860430971832,
      "language_loss": 0.81641626,
      "learning_rate": 2.5340472055382283e-06,
      "loss": 0.83865631,
      "num_input_tokens_seen": 77361765,
      "step": 3589,
      "time_per_iteration": 3.739046812057495
    },
    {
      "auxiliary_loss_clip": 0.01194065,
      "auxiliary_loss_mlp": 0.01025718,
      "balance_loss_clip": 1.04739177,
      "balance_loss_mlp": 1.01769471,
      "epoch": 0.43167197739433655,
      "flos": 24273853516800.0,
      "grad_norm": 2.443130173279769,
      "language_loss": 0.81400508,
      "learning_rate": 2.5332964778714468e-06,
      "loss": 0.83620286,
      "num_input_tokens_seen": 77378950,
      "step": 3590,
      "time_per_iteration": 3.6048762798309326
    },
    {
      "auxiliary_loss_clip": 0.01194704,
      "auxiliary_loss_mlp": 0.01025619,
      "balance_loss_clip": 1.0529238,
      "balance_loss_mlp": 1.01778722,
      "epoch": 0.43179222028497566,
      "flos": 16867816738560.0,
      "grad_norm": 1.7698390225945468,
      "language_loss": 0.66530836,
      "learning_rate": 2.5325456693040123e-06,
      "loss": 0.68751156,
      "num_input_tokens_seen": 77396145,
      "step": 3591,
      "time_per_iteration": 2.6866562366485596
    },
    {
      "auxiliary_loss_clip": 0.01198527,
      "auxiliary_loss_mlp": 0.01025092,
      "balance_loss_clip": 1.05291784,
      "balance_loss_mlp": 1.01624668,
      "epoch": 0.43191246317561477,
      "flos": 17639214243840.0,
      "grad_norm": 1.9956630490132032,
      "language_loss": 0.75272697,
      "learning_rate": 2.531794779949824e-06,
      "loss": 0.7749632,
      "num_input_tokens_seen": 77414045,
      "step": 3592,
      "time_per_iteration": 2.6035332679748535
    },
    {
      "auxiliary_loss_clip": 0.01187305,
      "auxiliary_loss_mlp": 0.01027583,
      "balance_loss_clip": 1.04853749,
      "balance_loss_mlp": 1.01957762,
      "epoch": 0.4320327060662538,
      "flos": 23878800760320.0,
      "grad_norm": 1.7248074231757533,
      "language_loss": 0.88099349,
      "learning_rate": 2.5310438099227903e-06,
      "loss": 0.90314233,
      "num_input_tokens_seen": 77431310,
      "step": 3593,
      "time_per_iteration": 3.5887906551361084
    },
    {
      "auxiliary_loss_clip": 0.01091293,
      "auxiliary_loss_mlp": 0.0100302,
      "balance_loss_clip": 1.01851046,
      "balance_loss_mlp": 1.00201893,
      "epoch": 0.43215294895689293,
      "flos": 66394917959040.0,
      "grad_norm": 0.7965112332987103,
      "language_loss": 0.53401309,
      "learning_rate": 2.530292759336833e-06,
      "loss": 0.55495626,
      "num_input_tokens_seen": 77492045,
      "step": 3594,
      "time_per_iteration": 3.187016487121582
    },
    {
      "auxiliary_loss_clip": 0.01190288,
      "auxiliary_loss_mlp": 0.01028481,
      "balance_loss_clip": 1.05259228,
      "balance_loss_mlp": 1.01966,
      "epoch": 0.432273191847532,
      "flos": 20594267262720.0,
      "grad_norm": 2.4531233037355538,
      "language_loss": 0.69559634,
      "learning_rate": 2.5295416283058855e-06,
      "loss": 0.71778405,
      "num_input_tokens_seen": 77510910,
      "step": 3595,
      "time_per_iteration": 2.648831844329834
    },
    {
      "auxiliary_loss_clip": 0.01189971,
      "auxiliary_loss_mlp": 0.01208651,
      "balance_loss_clip": 1.0519042,
      "balance_loss_mlp": 1.00070214,
      "epoch": 0.4323934347381711,
      "flos": 19282127437440.0,
      "grad_norm": 1.6963217308269503,
      "language_loss": 0.66094607,
      "learning_rate": 2.5287904169438943e-06,
      "loss": 0.68493235,
      "num_input_tokens_seen": 77530115,
      "step": 3596,
      "time_per_iteration": 2.826179027557373
    },
    {
      "auxiliary_loss_clip": 0.01211548,
      "auxiliary_loss_mlp": 0.01035655,
      "balance_loss_clip": 1.04756975,
      "balance_loss_mlp": 1.02599311,
      "epoch": 0.4325136776288102,
      "flos": 21726315273600.0,
      "grad_norm": 3.050607997673442,
      "language_loss": 0.64485443,
      "learning_rate": 2.528039125364817e-06,
      "loss": 0.66732645,
      "num_input_tokens_seen": 77548920,
      "step": 3597,
      "time_per_iteration": 2.70326566696167
    },
    {
      "auxiliary_loss_clip": 0.01198492,
      "auxiliary_loss_mlp": 0.01027887,
      "balance_loss_clip": 1.0507654,
      "balance_loss_mlp": 1.01894665,
      "epoch": 0.43263392051944927,
      "flos": 22340746344960.0,
      "grad_norm": 2.17002100909709,
      "language_loss": 0.75893164,
      "learning_rate": 2.5272877536826246e-06,
      "loss": 0.7811954,
      "num_input_tokens_seen": 77567715,
      "step": 3598,
      "time_per_iteration": 2.6473004817962646
    },
    {
      "auxiliary_loss_clip": 0.01200362,
      "auxiliary_loss_mlp": 0.01030484,
      "balance_loss_clip": 1.04494381,
      "balance_loss_mlp": 1.02163815,
      "epoch": 0.4327541634100884,
      "flos": 29168406328320.0,
      "grad_norm": 2.623687579746369,
      "language_loss": 0.70313609,
      "learning_rate": 2.5265363020112986e-06,
      "loss": 0.7254445,
      "num_input_tokens_seen": 77588035,
      "step": 3599,
      "time_per_iteration": 3.711056709289551
    },
    {
      "auxiliary_loss_clip": 0.01192433,
      "auxiliary_loss_mlp": 0.01034414,
      "balance_loss_clip": 1.05602515,
      "balance_loss_mlp": 1.02538395,
      "epoch": 0.4328744063007275,
      "flos": 26067448264320.0,
      "grad_norm": 1.7405444844539233,
      "language_loss": 0.83783972,
      "learning_rate": 2.5257847704648344e-06,
      "loss": 0.86010814,
      "num_input_tokens_seen": 77609265,
      "step": 3600,
      "time_per_iteration": 2.6753768920898438
    },
    {
      "auxiliary_loss_clip": 0.01188092,
      "auxiliary_loss_mlp": 0.01027994,
      "balance_loss_clip": 1.05483913,
      "balance_loss_mlp": 1.01969695,
      "epoch": 0.43299464919136654,
      "flos": 16581357774720.0,
      "grad_norm": 1.778444879306512,
      "language_loss": 0.75541979,
      "learning_rate": 2.525033159157239e-06,
      "loss": 0.77758062,
      "num_input_tokens_seen": 77625580,
      "step": 3601,
      "time_per_iteration": 2.586550235748291
    },
    {
      "auxiliary_loss_clip": 0.01189622,
      "auxiliary_loss_mlp": 0.01036383,
      "balance_loss_clip": 1.05387354,
      "balance_loss_mlp": 1.02713275,
      "epoch": 0.43311489208200565,
      "flos": 16107265140480.0,
      "grad_norm": 1.9125603748723161,
      "language_loss": 0.77491844,
      "learning_rate": 2.52428146820253e-06,
      "loss": 0.79717845,
      "num_input_tokens_seen": 77643835,
      "step": 3602,
      "time_per_iteration": 2.572911500930786
    },
    {
      "auxiliary_loss_clip": 0.01197769,
      "auxiliary_loss_mlp": 0.01026884,
      "balance_loss_clip": 1.05180955,
      "balance_loss_mlp": 1.0171926,
      "epoch": 0.43323513497264476,
      "flos": 22930220442240.0,
      "grad_norm": 1.744632952526154,
      "language_loss": 0.82020819,
      "learning_rate": 2.52352969771474e-06,
      "loss": 0.84245479,
      "num_input_tokens_seen": 77663060,
      "step": 3603,
      "time_per_iteration": 2.676853895187378
    },
    {
      "auxiliary_loss_clip": 0.01196109,
      "auxiliary_loss_mlp": 0.0102735,
      "balance_loss_clip": 1.05220318,
      "balance_loss_mlp": 1.01883888,
      "epoch": 0.4333553778632838,
      "flos": 25299031587840.0,
      "grad_norm": 2.141915183796452,
      "language_loss": 0.88418782,
      "learning_rate": 2.5227778478079106e-06,
      "loss": 0.90642244,
      "num_input_tokens_seen": 77682470,
      "step": 3604,
      "time_per_iteration": 2.633652448654175
    },
    {
      "auxiliary_loss_clip": 0.01186516,
      "auxiliary_loss_mlp": 0.01034652,
      "balance_loss_clip": 1.05140495,
      "balance_loss_mlp": 1.02634299,
      "epoch": 0.43347562075392293,
      "flos": 19387165783680.0,
      "grad_norm": 1.6667700340628249,
      "language_loss": 0.77092934,
      "learning_rate": 2.522025918596098e-06,
      "loss": 0.79314101,
      "num_input_tokens_seen": 77700770,
      "step": 3605,
      "time_per_iteration": 2.643449068069458
    },
    {
      "auxiliary_loss_clip": 0.011942,
      "auxiliary_loss_mlp": 0.01025272,
      "balance_loss_clip": 1.05459523,
      "balance_loss_mlp": 1.01713634,
      "epoch": 0.43359586364456204,
      "flos": 26325969425280.0,
      "grad_norm": 1.4624317641318538,
      "language_loss": 0.65469903,
      "learning_rate": 2.521273910193368e-06,
      "loss": 0.67689377,
      "num_input_tokens_seen": 77723950,
      "step": 3606,
      "time_per_iteration": 2.627074956893921
    },
    {
      "auxiliary_loss_clip": 0.01199101,
      "auxiliary_loss_mlp": 0.01028996,
      "balance_loss_clip": 1.05597019,
      "balance_loss_mlp": 1.0203234,
      "epoch": 0.4337161065352011,
      "flos": 15989261984640.0,
      "grad_norm": 3.0793643282851537,
      "language_loss": 0.87301767,
      "learning_rate": 2.5205218227138006e-06,
      "loss": 0.8952986,
      "num_input_tokens_seen": 77736905,
      "step": 3607,
      "time_per_iteration": 2.596123218536377
    },
    {
      "auxiliary_loss_clip": 0.01191071,
      "auxiliary_loss_mlp": 0.01029788,
      "balance_loss_clip": 1.05624914,
      "balance_loss_mlp": 1.02096057,
      "epoch": 0.4338363494258402,
      "flos": 20224710184320.0,
      "grad_norm": 2.1345154228158405,
      "language_loss": 0.79419684,
      "learning_rate": 2.519769656271486e-06,
      "loss": 0.81640542,
      "num_input_tokens_seen": 77754325,
      "step": 3608,
      "time_per_iteration": 2.5654587745666504
    },
    {
      "auxiliary_loss_clip": 0.01187718,
      "auxiliary_loss_mlp": 0.0102879,
      "balance_loss_clip": 1.04704738,
      "balance_loss_mlp": 1.01980114,
      "epoch": 0.43395659231647926,
      "flos": 20083904870400.0,
      "grad_norm": 1.9644767610211213,
      "language_loss": 0.67896909,
      "learning_rate": 2.5190174109805285e-06,
      "loss": 0.7011342,
      "num_input_tokens_seen": 77774150,
      "step": 3609,
      "time_per_iteration": 2.696977138519287
    },
    {
      "auxiliary_loss_clip": 0.0118331,
      "auxiliary_loss_mlp": 0.01028238,
      "balance_loss_clip": 1.04862928,
      "balance_loss_mlp": 1.01892734,
      "epoch": 0.43407683520711837,
      "flos": 19901801894400.0,
      "grad_norm": 2.1728366984840513,
      "language_loss": 0.63625669,
      "learning_rate": 2.518265086955042e-06,
      "loss": 0.65837216,
      "num_input_tokens_seen": 77791870,
      "step": 3610,
      "time_per_iteration": 2.606254816055298
    },
    {
      "auxiliary_loss_clip": 0.01190369,
      "auxiliary_loss_mlp": 0.01036406,
      "balance_loss_clip": 1.05538476,
      "balance_loss_mlp": 1.02801347,
      "epoch": 0.4341970780977575,
      "flos": 23108732058240.0,
      "grad_norm": 4.484149202980009,
      "language_loss": 0.83755368,
      "learning_rate": 2.5175126843091534e-06,
      "loss": 0.85982144,
      "num_input_tokens_seen": 77811240,
      "step": 3611,
      "time_per_iteration": 2.677337884902954
    },
    {
      "auxiliary_loss_clip": 0.0119734,
      "auxiliary_loss_mlp": 0.01026282,
      "balance_loss_clip": 1.05255318,
      "balance_loss_mlp": 1.01761532,
      "epoch": 0.43431732098839654,
      "flos": 37408288406400.0,
      "grad_norm": 2.435186158264734,
      "language_loss": 0.75417197,
      "learning_rate": 2.5167602031570034e-06,
      "loss": 0.7764082,
      "num_input_tokens_seen": 77831425,
      "step": 3612,
      "time_per_iteration": 2.8833377361297607
    },
    {
      "auxiliary_loss_clip": 0.01190581,
      "auxiliary_loss_mlp": 0.01030014,
      "balance_loss_clip": 1.05664134,
      "balance_loss_mlp": 1.02119303,
      "epoch": 0.43443756387903565,
      "flos": 31868206323840.0,
      "grad_norm": 4.372847549873819,
      "language_loss": 0.73637968,
      "learning_rate": 2.51600764361274e-06,
      "loss": 0.75858557,
      "num_input_tokens_seen": 77852950,
      "step": 3613,
      "time_per_iteration": 2.7466306686401367
    },
    {
      "auxiliary_loss_clip": 0.011916,
      "auxiliary_loss_mlp": 0.01031,
      "balance_loss_clip": 1.05618072,
      "balance_loss_mlp": 1.02222073,
      "epoch": 0.43455780676967476,
      "flos": 23477139901440.0,
      "grad_norm": 2.4082353706804414,
      "language_loss": 0.79086339,
      "learning_rate": 2.5152550057905283e-06,
      "loss": 0.81308937,
      "num_input_tokens_seen": 77872840,
      "step": 3614,
      "time_per_iteration": 2.5853869915008545
    },
    {
      "auxiliary_loss_clip": 0.01196135,
      "auxiliary_loss_mlp": 0.01209307,
      "balance_loss_clip": 1.05662608,
      "balance_loss_mlp": 1.00078726,
      "epoch": 0.4346780496603138,
      "flos": 24207060176640.0,
      "grad_norm": 2.3262080312481985,
      "language_loss": 0.7727412,
      "learning_rate": 2.5145022898045415e-06,
      "loss": 0.79679567,
      "num_input_tokens_seen": 77892025,
      "step": 3615,
      "time_per_iteration": 2.6406917572021484
    },
    {
      "auxiliary_loss_clip": 0.01196814,
      "auxiliary_loss_mlp": 0.01032347,
      "balance_loss_clip": 1.05031931,
      "balance_loss_mlp": 1.0230068,
      "epoch": 0.4347982925509529,
      "flos": 17092366611840.0,
      "grad_norm": 2.26472052452827,
      "language_loss": 0.90292335,
      "learning_rate": 2.5137494957689664e-06,
      "loss": 0.92521495,
      "num_input_tokens_seen": 77907635,
      "step": 3616,
      "time_per_iteration": 3.5974600315093994
    },
    {
      "auxiliary_loss_clip": 0.01097456,
      "auxiliary_loss_mlp": 0.01003861,
      "balance_loss_clip": 1.01889563,
      "balance_loss_mlp": 1.00278175,
      "epoch": 0.43491853544159204,
      "flos": 60945544696320.0,
      "grad_norm": 0.7640220526519456,
      "language_loss": 0.57297504,
      "learning_rate": 2.5129966237980016e-06,
      "loss": 0.59398824,
      "num_input_tokens_seen": 77970630,
      "step": 3617,
      "time_per_iteration": 4.101282835006714
    },
    {
      "auxiliary_loss_clip": 0.01199559,
      "auxiliary_loss_mlp": 0.01026016,
      "balance_loss_clip": 1.05134916,
      "balance_loss_mlp": 1.01741457,
      "epoch": 0.4350387783322311,
      "flos": 21944652094080.0,
      "grad_norm": 2.3776384507782224,
      "language_loss": 0.7816478,
      "learning_rate": 2.512243674005857e-06,
      "loss": 0.80390352,
      "num_input_tokens_seen": 77989995,
      "step": 3618,
      "time_per_iteration": 2.739701747894287
    },
    {
      "auxiliary_loss_clip": 0.01198631,
      "auxiliary_loss_mlp": 0.01031057,
      "balance_loss_clip": 1.04761767,
      "balance_loss_mlp": 1.02212858,
      "epoch": 0.4351590212228702,
      "flos": 25082705928960.0,
      "grad_norm": 1.7095950247494982,
      "language_loss": 0.86300242,
      "learning_rate": 2.5114906465067537e-06,
      "loss": 0.88529927,
      "num_input_tokens_seen": 78010980,
      "step": 3619,
      "time_per_iteration": 2.759967565536499
    },
    {
      "auxiliary_loss_clip": 0.01193669,
      "auxiliary_loss_mlp": 0.01027444,
      "balance_loss_clip": 1.05254936,
      "balance_loss_mlp": 1.01836622,
      "epoch": 0.4352792641135093,
      "flos": 21506541909120.0,
      "grad_norm": 2.1362635034533937,
      "language_loss": 0.74872786,
      "learning_rate": 2.5107375414149264e-06,
      "loss": 0.77093899,
      "num_input_tokens_seen": 78030225,
      "step": 3620,
      "time_per_iteration": 3.5309243202209473
    },
    {
      "auxiliary_loss_clip": 0.01191319,
      "auxiliary_loss_mlp": 0.0102986,
      "balance_loss_clip": 1.04546428,
      "balance_loss_mlp": 1.02021003,
      "epoch": 0.43539950700414837,
      "flos": 16253457494400.0,
      "grad_norm": 2.3732485160964356,
      "language_loss": 0.71803021,
      "learning_rate": 2.5099843588446197e-06,
      "loss": 0.740242,
      "num_input_tokens_seen": 78048545,
      "step": 3621,
      "time_per_iteration": 2.713259696960449
    },
    {
      "auxiliary_loss_clip": 0.01209764,
      "auxiliary_loss_mlp": 0.01032327,
      "balance_loss_clip": 1.05324757,
      "balance_loss_mlp": 1.02321339,
      "epoch": 0.4355197498947875,
      "flos": 16691819074560.0,
      "grad_norm": 1.596823615426517,
      "language_loss": 0.61320907,
      "learning_rate": 2.509231098910091e-06,
      "loss": 0.63563001,
      "num_input_tokens_seen": 78068415,
      "step": 3622,
      "time_per_iteration": 2.6426212787628174
    },
    {
      "auxiliary_loss_clip": 0.01192317,
      "auxiliary_loss_mlp": 0.01031127,
      "balance_loss_clip": 1.05557144,
      "balance_loss_mlp": 1.02171516,
      "epoch": 0.4356399927854266,
      "flos": 16362733645440.0,
      "grad_norm": 3.161812909338373,
      "language_loss": 0.74647367,
      "learning_rate": 2.508477761725611e-06,
      "loss": 0.76870811,
      "num_input_tokens_seen": 78086690,
      "step": 3623,
      "time_per_iteration": 2.6684865951538086
    },
    {
      "auxiliary_loss_clip": 0.01196884,
      "auxiliary_loss_mlp": 0.01032345,
      "balance_loss_clip": 1.05571485,
      "balance_loss_mlp": 1.02319002,
      "epoch": 0.43576023567606564,
      "flos": 17202037812480.0,
      "grad_norm": 1.9951394707816836,
      "language_loss": 0.80703986,
      "learning_rate": 2.507724347405458e-06,
      "loss": 0.82933217,
      "num_input_tokens_seen": 78104640,
      "step": 3624,
      "time_per_iteration": 2.5502798557281494
    },
    {
      "auxiliary_loss_clip": 0.01193717,
      "auxiliary_loss_mlp": 0.01030678,
      "balance_loss_clip": 1.04674566,
      "balance_loss_mlp": 1.02210128,
      "epoch": 0.43588047856670475,
      "flos": 15917656222080.0,
      "grad_norm": 2.171196307951499,
      "language_loss": 0.82137799,
      "learning_rate": 2.5069708560639243e-06,
      "loss": 0.84362197,
      "num_input_tokens_seen": 78122550,
      "step": 3625,
      "time_per_iteration": 2.8416812419891357
    },
    {
      "auxiliary_loss_clip": 0.01198088,
      "auxiliary_loss_mlp": 0.01028515,
      "balance_loss_clip": 1.05131221,
      "balance_loss_mlp": 1.01930606,
      "epoch": 0.4360007214573438,
      "flos": 23659566099840.0,
      "grad_norm": 1.9255692766618302,
      "language_loss": 0.61306483,
      "learning_rate": 2.5062172878153158e-06,
      "loss": 0.63533086,
      "num_input_tokens_seen": 78141825,
      "step": 3626,
      "time_per_iteration": 3.6035144329071045
    },
    {
      "auxiliary_loss_clip": 0.01210009,
      "auxiliary_loss_mlp": 0.01031225,
      "balance_loss_clip": 1.04917729,
      "balance_loss_mlp": 1.02111018,
      "epoch": 0.4361209643479829,
      "flos": 21978767036160.0,
      "grad_norm": 2.1844639380531485,
      "language_loss": 0.87785721,
      "learning_rate": 2.505463642773947e-06,
      "loss": 0.90026963,
      "num_input_tokens_seen": 78161790,
      "step": 3627,
      "time_per_iteration": 2.7122957706451416
    },
    {
      "auxiliary_loss_clip": 0.01201159,
      "auxiliary_loss_mlp": 0.01209258,
      "balance_loss_clip": 1.05372095,
      "balance_loss_mlp": 1.0006386,
      "epoch": 0.43624120723862203,
      "flos": 17420159151360.0,
      "grad_norm": 3.924284741784235,
      "language_loss": 0.75486743,
      "learning_rate": 2.504709921054146e-06,
      "loss": 0.77897155,
      "num_input_tokens_seen": 78178605,
      "step": 3628,
      "time_per_iteration": 2.6637966632843018
    },
    {
      "auxiliary_loss_clip": 0.01192709,
      "auxiliary_loss_mlp": 0.01030173,
      "balance_loss_clip": 1.04714823,
      "balance_loss_mlp": 1.02080905,
      "epoch": 0.4363614501292611,
      "flos": 17895293280000.0,
      "grad_norm": 2.331566424870759,
      "language_loss": 0.83868051,
      "learning_rate": 2.50395612277025e-06,
      "loss": 0.86090934,
      "num_input_tokens_seen": 78194460,
      "step": 3629,
      "time_per_iteration": 2.722567558288574
    },
    {
      "auxiliary_loss_clip": 0.01199372,
      "auxiliary_loss_mlp": 0.01028038,
      "balance_loss_clip": 1.05281115,
      "balance_loss_mlp": 1.01906204,
      "epoch": 0.4364816930199002,
      "flos": 20302888135680.0,
      "grad_norm": 2.5916066265251336,
      "language_loss": 0.72928298,
      "learning_rate": 2.503202248036612e-06,
      "loss": 0.75155711,
      "num_input_tokens_seen": 78213315,
      "step": 3630,
      "time_per_iteration": 2.6600899696350098
    },
    {
      "auxiliary_loss_clip": 0.01190113,
      "auxiliary_loss_mlp": 0.01036176,
      "balance_loss_clip": 1.05629468,
      "balance_loss_mlp": 1.0268842,
      "epoch": 0.4366019359105393,
      "flos": 24061334699520.0,
      "grad_norm": 1.7319352405015567,
      "language_loss": 0.73188984,
      "learning_rate": 2.5024482969675927e-06,
      "loss": 0.75415272,
      "num_input_tokens_seen": 78233270,
      "step": 3631,
      "time_per_iteration": 2.6089279651641846
    },
    {
      "auxiliary_loss_clip": 0.01207005,
      "auxiliary_loss_mlp": 0.01024894,
      "balance_loss_clip": 1.05205142,
      "balance_loss_mlp": 1.01650214,
      "epoch": 0.43672217880117836,
      "flos": 21754109422080.0,
      "grad_norm": 1.9611072532970508,
      "language_loss": 0.84985453,
      "learning_rate": 2.501694269677566e-06,
      "loss": 0.87217355,
      "num_input_tokens_seen": 78251040,
      "step": 3632,
      "time_per_iteration": 2.6809613704681396
    },
    {
      "auxiliary_loss_clip": 0.01195602,
      "auxiliary_loss_mlp": 0.01027467,
      "balance_loss_clip": 1.05283856,
      "balance_loss_mlp": 1.01853836,
      "epoch": 0.4368424216918175,
      "flos": 18035200753920.0,
      "grad_norm": 2.454267000322443,
      "language_loss": 0.80541676,
      "learning_rate": 2.500940166280918e-06,
      "loss": 0.82764745,
      "num_input_tokens_seen": 78269470,
      "step": 3633,
      "time_per_iteration": 2.5970053672790527
    },
    {
      "auxiliary_loss_clip": 0.01189419,
      "auxiliary_loss_mlp": 0.01030216,
      "balance_loss_clip": 1.05345023,
      "balance_loss_mlp": 1.02127528,
      "epoch": 0.4369626645824566,
      "flos": 25447127362560.0,
      "grad_norm": 1.8364101081974338,
      "language_loss": 0.79118919,
      "learning_rate": 2.500185986892045e-06,
      "loss": 0.81338561,
      "num_input_tokens_seen": 78288955,
      "step": 3634,
      "time_per_iteration": 2.616014242172241
    },
    {
      "auxiliary_loss_clip": 0.01190512,
      "auxiliary_loss_mlp": 0.01030731,
      "balance_loss_clip": 1.05454123,
      "balance_loss_mlp": 1.02169514,
      "epoch": 0.43708290747309564,
      "flos": 25302694775040.0,
      "grad_norm": 2.310999402902722,
      "language_loss": 0.77639192,
      "learning_rate": 2.499431731625355e-06,
      "loss": 0.79860437,
      "num_input_tokens_seen": 78307980,
      "step": 3635,
      "time_per_iteration": 2.7102155685424805
    },
    {
      "auxiliary_loss_clip": 0.01193298,
      "auxiliary_loss_mlp": 0.0102649,
      "balance_loss_clip": 1.05695033,
      "balance_loss_mlp": 1.01692927,
      "epoch": 0.43720315036373475,
      "flos": 31575103344000.0,
      "grad_norm": 2.7563982940183003,
      "language_loss": 0.79585648,
      "learning_rate": 2.4986774005952686e-06,
      "loss": 0.81805438,
      "num_input_tokens_seen": 78330355,
      "step": 3636,
      "time_per_iteration": 2.6852810382843018
    },
    {
      "auxiliary_loss_clip": 0.0119106,
      "auxiliary_loss_mlp": 0.01026831,
      "balance_loss_clip": 1.05661964,
      "balance_loss_mlp": 1.01856339,
      "epoch": 0.43732339325437386,
      "flos": 23112000195840.0,
      "grad_norm": 2.2580451971977906,
      "language_loss": 0.84916621,
      "learning_rate": 2.4979229939162166e-06,
      "loss": 0.8713451,
      "num_input_tokens_seen": 78349135,
      "step": 3637,
      "time_per_iteration": 2.6283490657806396
    },
    {
      "auxiliary_loss_clip": 0.01190397,
      "auxiliary_loss_mlp": 0.01026747,
      "balance_loss_clip": 1.05567336,
      "balance_loss_mlp": 1.01831865,
      "epoch": 0.4374436361450129,
      "flos": 27746272080000.0,
      "grad_norm": 2.4360453163734843,
      "language_loss": 0.80876231,
      "learning_rate": 2.4971685117026433e-06,
      "loss": 0.83093375,
      "num_input_tokens_seen": 78368900,
      "step": 3638,
      "time_per_iteration": 2.6508679389953613
    },
    {
      "auxiliary_loss_clip": 0.01194995,
      "auxiliary_loss_mlp": 0.0102545,
      "balance_loss_clip": 1.05645323,
      "balance_loss_mlp": 1.01671219,
      "epoch": 0.437563879035652,
      "flos": 24172370616960.0,
      "grad_norm": 2.0845123198182725,
      "language_loss": 0.76838529,
      "learning_rate": 2.4964139540690018e-06,
      "loss": 0.79058969,
      "num_input_tokens_seen": 78392235,
      "step": 3639,
      "time_per_iteration": 2.6762123107910156
    },
    {
      "auxiliary_loss_clip": 0.01199288,
      "auxiliary_loss_mlp": 0.01027572,
      "balance_loss_clip": 1.0521847,
      "balance_loss_mlp": 1.01856518,
      "epoch": 0.4376841219262911,
      "flos": 23477211728640.0,
      "grad_norm": 2.0361321931567544,
      "language_loss": 0.72817659,
      "learning_rate": 2.495659321129758e-06,
      "loss": 0.75044513,
      "num_input_tokens_seen": 78409980,
      "step": 3640,
      "time_per_iteration": 2.815234661102295
    },
    {
      "auxiliary_loss_clip": 0.01188405,
      "auxiliary_loss_mlp": 0.01030622,
      "balance_loss_clip": 1.05210543,
      "balance_loss_mlp": 1.02215266,
      "epoch": 0.4378043648169302,
      "flos": 25447809720960.0,
      "grad_norm": 1.8682966369262182,
      "language_loss": 0.75436616,
      "learning_rate": 2.494904612999389e-06,
      "loss": 0.77655649,
      "num_input_tokens_seen": 78428690,
      "step": 3641,
      "time_per_iteration": 2.689746856689453
    },
    {
      "auxiliary_loss_clip": 0.01087177,
      "auxiliary_loss_mlp": 0.01006275,
      "balance_loss_clip": 1.01914239,
      "balance_loss_mlp": 1.0052737,
      "epoch": 0.4379246077075693,
      "flos": 53914056986880.0,
      "grad_norm": 0.8059513174487745,
      "language_loss": 0.56508237,
      "learning_rate": 2.4941498297923843e-06,
      "loss": 0.58601689,
      "num_input_tokens_seen": 78489260,
      "step": 3642,
      "time_per_iteration": 4.102389097213745
    },
    {
      "auxiliary_loss_clip": 0.01191163,
      "auxiliary_loss_mlp": 0.01026211,
      "balance_loss_clip": 1.05427718,
      "balance_loss_mlp": 1.01765776,
      "epoch": 0.43804485059820836,
      "flos": 20588305605120.0,
      "grad_norm": 1.8685380979930228,
      "language_loss": 0.69539696,
      "learning_rate": 2.4933949716232424e-06,
      "loss": 0.71757066,
      "num_input_tokens_seen": 78506785,
      "step": 3643,
      "time_per_iteration": 3.5111517906188965
    },
    {
      "auxiliary_loss_clip": 0.01196801,
      "auxiliary_loss_mlp": 0.01026387,
      "balance_loss_clip": 1.05262387,
      "balance_loss_mlp": 1.01761293,
      "epoch": 0.43816509348884747,
      "flos": 23876214981120.0,
      "grad_norm": 2.5080370765475357,
      "language_loss": 0.73628467,
      "learning_rate": 2.492640038606476e-06,
      "loss": 0.75851655,
      "num_input_tokens_seen": 78525150,
      "step": 3644,
      "time_per_iteration": 2.6439929008483887
    },
    {
      "auxiliary_loss_clip": 0.01191084,
      "auxiliary_loss_mlp": 0.01029971,
      "balance_loss_clip": 1.05172801,
      "balance_loss_mlp": 1.0209229,
      "epoch": 0.4382853363794866,
      "flos": 14684448533760.0,
      "grad_norm": 2.103244499735839,
      "language_loss": 0.78955621,
      "learning_rate": 2.491885030856608e-06,
      "loss": 0.81176674,
      "num_input_tokens_seen": 78543245,
      "step": 3645,
      "time_per_iteration": 2.588502883911133
    },
    {
      "auxiliary_loss_clip": 0.01198594,
      "auxiliary_loss_mlp": 0.01027478,
      "balance_loss_clip": 1.05588675,
      "balance_loss_mlp": 1.0190444,
      "epoch": 0.43840557927012563,
      "flos": 17165301177600.0,
      "grad_norm": 2.5071003103793186,
      "language_loss": 0.83019245,
      "learning_rate": 2.4911299484881713e-06,
      "loss": 0.85245323,
      "num_input_tokens_seen": 78560775,
      "step": 3646,
      "time_per_iteration": 2.593902111053467
    },
    {
      "auxiliary_loss_clip": 0.01190231,
      "auxiliary_loss_mlp": 0.0102427,
      "balance_loss_clip": 1.05250669,
      "balance_loss_mlp": 1.01569247,
      "epoch": 0.43852582216076474,
      "flos": 19390685316480.0,
      "grad_norm": 1.6777055956367029,
      "language_loss": 0.81047696,
      "learning_rate": 2.490374791615712e-06,
      "loss": 0.83262193,
      "num_input_tokens_seen": 78580800,
      "step": 3647,
      "time_per_iteration": 3.6522107124328613
    },
    {
      "auxiliary_loss_clip": 0.01194908,
      "auxiliary_loss_mlp": 0.01209017,
      "balance_loss_clip": 1.05684578,
      "balance_loss_mlp": 1.00063968,
      "epoch": 0.43864606505140386,
      "flos": 18075133699200.0,
      "grad_norm": 13.087601336415501,
      "language_loss": 0.77870733,
      "learning_rate": 2.4896195603537867e-06,
      "loss": 0.80274659,
      "num_input_tokens_seen": 78595410,
      "step": 3648,
      "time_per_iteration": 2.5473806858062744
    },
    {
      "auxiliary_loss_clip": 0.01194733,
      "auxiliary_loss_mlp": 0.01028183,
      "balance_loss_clip": 1.05261683,
      "balance_loss_mlp": 1.01924253,
      "epoch": 0.4387663079420429,
      "flos": 19644896845440.0,
      "grad_norm": 2.138702908734749,
      "language_loss": 0.74040824,
      "learning_rate": 2.488864254816964e-06,
      "loss": 0.76263738,
      "num_input_tokens_seen": 78614100,
      "step": 3649,
      "time_per_iteration": 2.6368777751922607
    },
    {
      "auxiliary_loss_clip": 0.01196517,
      "auxiliary_loss_mlp": 0.0103487,
      "balance_loss_clip": 1.05734658,
      "balance_loss_mlp": 1.02571452,
      "epoch": 0.438886550832682,
      "flos": 19719339782400.0,
      "grad_norm": 4.418311428773128,
      "language_loss": 0.68370253,
      "learning_rate": 2.4881088751198218e-06,
      "loss": 0.70601642,
      "num_input_tokens_seen": 78632260,
      "step": 3650,
      "time_per_iteration": 2.620161771774292
    },
    {
      "auxiliary_loss_clip": 0.01195833,
      "auxiliary_loss_mlp": 0.01028216,
      "balance_loss_clip": 1.05056095,
      "balance_loss_mlp": 1.01928163,
      "epoch": 0.43900679372332113,
      "flos": 14536675981440.0,
      "grad_norm": 2.406340562612107,
      "language_loss": 0.63985497,
      "learning_rate": 2.4873534213769517e-06,
      "loss": 0.66209543,
      "num_input_tokens_seen": 78647490,
      "step": 3651,
      "time_per_iteration": 2.6442580223083496
    },
    {
      "auxiliary_loss_clip": 0.01192901,
      "auxiliary_loss_mlp": 0.01028861,
      "balance_loss_clip": 1.05500698,
      "balance_loss_mlp": 1.0203073,
      "epoch": 0.4391270366139602,
      "flos": 24056234968320.0,
      "grad_norm": 1.8263152753025382,
      "language_loss": 0.72014034,
      "learning_rate": 2.4865978937029547e-06,
      "loss": 0.74235797,
      "num_input_tokens_seen": 78666470,
      "step": 3652,
      "time_per_iteration": 3.5796267986297607
    },
    {
      "auxiliary_loss_clip": 0.01191714,
      "auxiliary_loss_mlp": 0.01034417,
      "balance_loss_clip": 1.05178475,
      "balance_loss_mlp": 1.0249939,
      "epoch": 0.4392472795045993,
      "flos": 31538510363520.0,
      "grad_norm": 1.7565943001783773,
      "language_loss": 0.66650772,
      "learning_rate": 2.485842292212445e-06,
      "loss": 0.68876904,
      "num_input_tokens_seen": 78687685,
      "step": 3653,
      "time_per_iteration": 2.795646905899048
    },
    {
      "auxiliary_loss_clip": 0.01194119,
      "auxiliary_loss_mlp": 0.01033459,
      "balance_loss_clip": 1.05881262,
      "balance_loss_mlp": 1.02427936,
      "epoch": 0.4393675223952384,
      "flos": 14866300114560.0,
      "grad_norm": 1.904196954702534,
      "language_loss": 0.80453062,
      "learning_rate": 2.485086617020045e-06,
      "loss": 0.82680643,
      "num_input_tokens_seen": 78706180,
      "step": 3654,
      "time_per_iteration": 2.5439677238464355
    },
    {
      "auxiliary_loss_clip": 0.01187395,
      "auxiliary_loss_mlp": 0.01029249,
      "balance_loss_clip": 1.05054307,
      "balance_loss_mlp": 1.02006936,
      "epoch": 0.43948776528587746,
      "flos": 14825900292480.0,
      "grad_norm": 2.6176920351947093,
      "language_loss": 0.82188863,
      "learning_rate": 2.4843308682403903e-06,
      "loss": 0.84405506,
      "num_input_tokens_seen": 78723095,
      "step": 3655,
      "time_per_iteration": 2.6458799839019775
    },
    {
      "auxiliary_loss_clip": 0.01190835,
      "auxiliary_loss_mlp": 0.01027839,
      "balance_loss_clip": 1.05620241,
      "balance_loss_mlp": 1.01911306,
      "epoch": 0.4396080081765166,
      "flos": 13914523486080.0,
      "grad_norm": 1.856704961658629,
      "language_loss": 0.82960176,
      "learning_rate": 2.4835750459881294e-06,
      "loss": 0.85178852,
      "num_input_tokens_seen": 78739720,
      "step": 3656,
      "time_per_iteration": 2.5328030586242676
    },
    {
      "auxiliary_loss_clip": 0.01185728,
      "auxiliary_loss_mlp": 0.01037828,
      "balance_loss_clip": 1.04917836,
      "balance_loss_mlp": 1.02791595,
      "epoch": 0.43972825106715563,
      "flos": 18222978078720.0,
      "grad_norm": 1.8290287194304666,
      "language_loss": 0.81705821,
      "learning_rate": 2.4828191503779177e-06,
      "loss": 0.83929378,
      "num_input_tokens_seen": 78757820,
      "step": 3657,
      "time_per_iteration": 2.6298978328704834
    },
    {
      "auxiliary_loss_clip": 0.01195634,
      "auxiliary_loss_mlp": 0.01026108,
      "balance_loss_clip": 1.0505116,
      "balance_loss_mlp": 1.01704848,
      "epoch": 0.43984849395779474,
      "flos": 16873239692160.0,
      "grad_norm": 2.1287396809003636,
      "language_loss": 0.8992762,
      "learning_rate": 2.482063181524425e-06,
      "loss": 0.92149365,
      "num_input_tokens_seen": 78773720,
      "step": 3658,
      "time_per_iteration": 2.6481945514678955
    },
    {
      "auxiliary_loss_clip": 0.01194259,
      "auxiliary_loss_mlp": 0.01037611,
      "balance_loss_clip": 1.05793107,
      "balance_loss_mlp": 1.02831912,
      "epoch": 0.43996873684843385,
      "flos": 18691504104960.0,
      "grad_norm": 2.1496146552150166,
      "language_loss": 0.81241018,
      "learning_rate": 2.4813071395423307e-06,
      "loss": 0.83472884,
      "num_input_tokens_seen": 78791285,
      "step": 3659,
      "time_per_iteration": 2.60628604888916
    },
    {
      "auxiliary_loss_clip": 0.0119222,
      "auxiliary_loss_mlp": 0.01031616,
      "balance_loss_clip": 1.05451369,
      "balance_loss_mlp": 1.02225173,
      "epoch": 0.4400889797390729,
      "flos": 23653460787840.0,
      "grad_norm": 1.932049830666889,
      "language_loss": 0.64588577,
      "learning_rate": 2.4805510245463263e-06,
      "loss": 0.66812414,
      "num_input_tokens_seen": 78811440,
      "step": 3660,
      "time_per_iteration": 2.6348624229431152
    },
    {
      "auxiliary_loss_clip": 0.01192139,
      "auxiliary_loss_mlp": 0.0103379,
      "balance_loss_clip": 1.05332768,
      "balance_loss_mlp": 1.0241878,
      "epoch": 0.440209222629712,
      "flos": 23149203707520.0,
      "grad_norm": 2.290439753832973,
      "language_loss": 0.60811752,
      "learning_rate": 2.4797948366511137e-06,
      "loss": 0.63037682,
      "num_input_tokens_seen": 78831150,
      "step": 3661,
      "time_per_iteration": 2.6358845233917236
    },
    {
      "auxiliary_loss_clip": 0.01197096,
      "auxiliary_loss_mlp": 0.01033924,
      "balance_loss_clip": 1.04952097,
      "balance_loss_mlp": 1.02507305,
      "epoch": 0.4403294655203511,
      "flos": 24823394668800.0,
      "grad_norm": 5.168198428486822,
      "language_loss": 0.76862848,
      "learning_rate": 2.4790385759714055e-06,
      "loss": 0.79093868,
      "num_input_tokens_seen": 78850215,
      "step": 3662,
      "time_per_iteration": 2.6897614002227783
    },
    {
      "auxiliary_loss_clip": 0.01191881,
      "auxiliary_loss_mlp": 0.01026731,
      "balance_loss_clip": 1.05712652,
      "balance_loss_mlp": 1.0180769,
      "epoch": 0.4404497084109902,
      "flos": 22565080736640.0,
      "grad_norm": 1.6630114990676998,
      "language_loss": 0.71155977,
      "learning_rate": 2.478282242621926e-06,
      "loss": 0.73374587,
      "num_input_tokens_seen": 78870675,
      "step": 3663,
      "time_per_iteration": 2.6378114223480225
    },
    {
      "auxiliary_loss_clip": 0.01099385,
      "auxiliary_loss_mlp": 0.01001838,
      "balance_loss_clip": 1.01674199,
      "balance_loss_mlp": 1.00072944,
      "epoch": 0.4405699513016293,
      "flos": 64967073448320.0,
      "grad_norm": 0.842336752576115,
      "language_loss": 0.59481549,
      "learning_rate": 2.477525836717411e-06,
      "loss": 0.61582774,
      "num_input_tokens_seen": 78938440,
      "step": 3664,
      "time_per_iteration": 3.327587366104126
    },
    {
      "auxiliary_loss_clip": 0.01190484,
      "auxiliary_loss_mlp": 0.01031491,
      "balance_loss_clip": 1.05241776,
      "balance_loss_mlp": 1.02275324,
      "epoch": 0.4406901941922684,
      "flos": 35661952978560.0,
      "grad_norm": 3.9157762339201367,
      "language_loss": 0.80019736,
      "learning_rate": 2.476769358372606e-06,
      "loss": 0.82241714,
      "num_input_tokens_seen": 78960090,
      "step": 3665,
      "time_per_iteration": 2.7287518978118896
    },
    {
      "auxiliary_loss_clip": 0.01192544,
      "auxiliary_loss_mlp": 0.01028234,
      "balance_loss_clip": 1.05166876,
      "balance_loss_mlp": 1.02031851,
      "epoch": 0.44081043708290746,
      "flos": 18040767361920.0,
      "grad_norm": 1.9812733113234462,
      "language_loss": 0.74420154,
      "learning_rate": 2.4760128077022683e-06,
      "loss": 0.76640928,
      "num_input_tokens_seen": 78978225,
      "step": 3666,
      "time_per_iteration": 2.6503281593322754
    },
    {
      "auxiliary_loss_clip": 0.01189219,
      "auxiliary_loss_mlp": 0.01025589,
      "balance_loss_clip": 1.04971445,
      "balance_loss_mlp": 1.01730347,
      "epoch": 0.44093067997354657,
      "flos": 30153507799680.0,
      "grad_norm": 1.565981270259253,
      "language_loss": 0.68659371,
      "learning_rate": 2.4752561848211672e-06,
      "loss": 0.70874178,
      "num_input_tokens_seen": 79000625,
      "step": 3667,
      "time_per_iteration": 2.752272129058838
    },
    {
      "auxiliary_loss_clip": 0.01192721,
      "auxiliary_loss_mlp": 0.0103346,
      "balance_loss_clip": 1.05927777,
      "balance_loss_mlp": 1.02466202,
      "epoch": 0.4410509228641857,
      "flos": 23255068066560.0,
      "grad_norm": 1.8217930156033906,
      "language_loss": 0.71488464,
      "learning_rate": 2.4744994898440797e-06,
      "loss": 0.7371465,
      "num_input_tokens_seen": 79019415,
      "step": 3668,
      "time_per_iteration": 2.7449657917022705
    },
    {
      "auxiliary_loss_clip": 0.01203888,
      "auxiliary_loss_mlp": 0.01035607,
      "balance_loss_clip": 1.05355501,
      "balance_loss_mlp": 1.02598071,
      "epoch": 0.44117116575482473,
      "flos": 19500571998720.0,
      "grad_norm": 2.0580543667707274,
      "language_loss": 0.83699608,
      "learning_rate": 2.473742722885797e-06,
      "loss": 0.85939097,
      "num_input_tokens_seen": 79038435,
      "step": 3669,
      "time_per_iteration": 3.595945119857788
    },
    {
      "auxiliary_loss_clip": 0.01194456,
      "auxiliary_loss_mlp": 0.01209086,
      "balance_loss_clip": 1.05782211,
      "balance_loss_mlp": 1.00071287,
      "epoch": 0.44129140864546385,
      "flos": 27053124353280.0,
      "grad_norm": 2.635229343600037,
      "language_loss": 0.65177655,
      "learning_rate": 2.4729858840611197e-06,
      "loss": 0.67581189,
      "num_input_tokens_seen": 79057345,
      "step": 3670,
      "time_per_iteration": 3.6313462257385254
    },
    {
      "auxiliary_loss_clip": 0.01192159,
      "auxiliary_loss_mlp": 0.01029491,
      "balance_loss_clip": 1.05885541,
      "balance_loss_mlp": 1.02098584,
      "epoch": 0.4414116515361029,
      "flos": 26102101910400.0,
      "grad_norm": 3.1561211668567495,
      "language_loss": 0.72643673,
      "learning_rate": 2.4722289734848605e-06,
      "loss": 0.74865323,
      "num_input_tokens_seen": 79077810,
      "step": 3671,
      "time_per_iteration": 2.6916511058807373
    },
    {
      "auxiliary_loss_clip": 0.01193136,
      "auxiliary_loss_mlp": 0.0102928,
      "balance_loss_clip": 1.05428982,
      "balance_loss_mlp": 1.02053618,
      "epoch": 0.441531894426742,
      "flos": 21906083865600.0,
      "grad_norm": 9.840570481318847,
      "language_loss": 0.78080213,
      "learning_rate": 2.471471991271841e-06,
      "loss": 0.80302632,
      "num_input_tokens_seen": 79094935,
      "step": 3672,
      "time_per_iteration": 2.752309799194336
    },
    {
      "auxiliary_loss_clip": 0.01183287,
      "auxiliary_loss_mlp": 0.01024048,
      "balance_loss_clip": 1.05216968,
      "balance_loss_mlp": 1.01525056,
      "epoch": 0.4416521373173811,
      "flos": 23437099215360.0,
      "grad_norm": 6.11457555501888,
      "language_loss": 0.79554296,
      "learning_rate": 2.470714937536896e-06,
      "loss": 0.81761628,
      "num_input_tokens_seen": 79113660,
      "step": 3673,
      "time_per_iteration": 3.648040533065796
    },
    {
      "auxiliary_loss_clip": 0.01193948,
      "auxiliary_loss_mlp": 0.01032429,
      "balance_loss_clip": 1.0478245,
      "balance_loss_mlp": 1.0236609,
      "epoch": 0.4417723802080202,
      "flos": 20334345471360.0,
      "grad_norm": 2.26117962997795,
      "language_loss": 0.7076391,
      "learning_rate": 2.469957812394868e-06,
      "loss": 0.72990292,
      "num_input_tokens_seen": 79132470,
      "step": 3674,
      "time_per_iteration": 2.7187767028808594
    },
    {
      "auxiliary_loss_clip": 0.01188841,
      "auxiliary_loss_mlp": 0.01029005,
      "balance_loss_clip": 1.05739832,
      "balance_loss_mlp": 1.02027273,
      "epoch": 0.4418926230986593,
      "flos": 18880682060160.0,
      "grad_norm": 2.658972107068713,
      "language_loss": 0.76575392,
      "learning_rate": 2.4692006159606148e-06,
      "loss": 0.7879324,
      "num_input_tokens_seen": 79150000,
      "step": 3675,
      "time_per_iteration": 2.599844217300415
    },
    {
      "auxiliary_loss_clip": 0.01189766,
      "auxiliary_loss_mlp": 0.01028589,
      "balance_loss_clip": 1.05622268,
      "balance_loss_mlp": 1.01976156,
      "epoch": 0.4420128659892984,
      "flos": 19464409981440.0,
      "grad_norm": 1.746636962000818,
      "language_loss": 0.78453565,
      "learning_rate": 2.468443348349e-06,
      "loss": 0.80671918,
      "num_input_tokens_seen": 79167875,
      "step": 3676,
      "time_per_iteration": 2.59767746925354
    },
    {
      "auxiliary_loss_clip": 0.01197765,
      "auxiliary_loss_mlp": 0.01028478,
      "balance_loss_clip": 1.04970729,
      "balance_loss_mlp": 1.01878607,
      "epoch": 0.44213310887993745,
      "flos": 17894359526400.0,
      "grad_norm": 2.5715026377395422,
      "language_loss": 0.82124084,
      "learning_rate": 2.467686009674902e-06,
      "loss": 0.84350324,
      "num_input_tokens_seen": 79182325,
      "step": 3677,
      "time_per_iteration": 2.646047830581665
    },
    {
      "auxiliary_loss_clip": 0.01187238,
      "auxiliary_loss_mlp": 0.0102618,
      "balance_loss_clip": 1.0515461,
      "balance_loss_mlp": 1.01696539,
      "epoch": 0.44225335177057656,
      "flos": 19204667758080.0,
      "grad_norm": 3.280127945522521,
      "language_loss": 0.85554862,
      "learning_rate": 2.466928600053209e-06,
      "loss": 0.87768275,
      "num_input_tokens_seen": 79197630,
      "step": 3678,
      "time_per_iteration": 2.577937126159668
    },
    {
      "auxiliary_loss_clip": 0.01192806,
      "auxiliary_loss_mlp": 0.01026742,
      "balance_loss_clip": 1.05172765,
      "balance_loss_mlp": 1.01796174,
      "epoch": 0.4423735946612157,
      "flos": 23471321898240.0,
      "grad_norm": 2.0756611834292205,
      "language_loss": 0.716371,
      "learning_rate": 2.466171119598818e-06,
      "loss": 0.73856652,
      "num_input_tokens_seen": 79217600,
      "step": 3679,
      "time_per_iteration": 3.609272003173828
    },
    {
      "auxiliary_loss_clip": 0.01197185,
      "auxiliary_loss_mlp": 0.01034989,
      "balance_loss_clip": 1.0523541,
      "balance_loss_mlp": 1.0258038,
      "epoch": 0.44249383755185473,
      "flos": 26685398868480.0,
      "grad_norm": 3.568780216731502,
      "language_loss": 0.77702713,
      "learning_rate": 2.465413568426639e-06,
      "loss": 0.79934883,
      "num_input_tokens_seen": 79238550,
      "step": 3680,
      "time_per_iteration": 2.6245129108428955
    },
    {
      "auxiliary_loss_clip": 0.01184419,
      "auxiliary_loss_mlp": 0.01023568,
      "balance_loss_clip": 1.05320215,
      "balance_loss_mlp": 1.01554513,
      "epoch": 0.44261408044249384,
      "flos": 23147659422720.0,
      "grad_norm": 1.7093881872361922,
      "language_loss": 0.81323969,
      "learning_rate": 2.464655946651591e-06,
      "loss": 0.83531958,
      "num_input_tokens_seen": 79257555,
      "step": 3681,
      "time_per_iteration": 2.697303533554077
    },
    {
      "auxiliary_loss_clip": 0.0119584,
      "auxiliary_loss_mlp": 0.01030071,
      "balance_loss_clip": 1.05662537,
      "balance_loss_mlp": 1.02139854,
      "epoch": 0.44273432333313295,
      "flos": 24462564595200.0,
      "grad_norm": 1.891336450842565,
      "language_loss": 0.8101002,
      "learning_rate": 2.4638982543886065e-06,
      "loss": 0.83235931,
      "num_input_tokens_seen": 79277595,
      "step": 3682,
      "time_per_iteration": 2.600142002105713
    },
    {
      "auxiliary_loss_clip": 0.01195382,
      "auxiliary_loss_mlp": 0.01033655,
      "balance_loss_clip": 1.05632794,
      "balance_loss_mlp": 1.02451754,
      "epoch": 0.442854566223772,
      "flos": 17528932512000.0,
      "grad_norm": 3.626094002752787,
      "language_loss": 0.87065244,
      "learning_rate": 2.4631404917526254e-06,
      "loss": 0.89294279,
      "num_input_tokens_seen": 79294550,
      "step": 3683,
      "time_per_iteration": 2.613055944442749
    },
    {
      "auxiliary_loss_clip": 0.01184529,
      "auxiliary_loss_mlp": 0.01026491,
      "balance_loss_clip": 1.05158377,
      "balance_loss_mlp": 1.01854014,
      "epoch": 0.4429748091144111,
      "flos": 24896293320960.0,
      "grad_norm": 2.1660805039021067,
      "language_loss": 0.79297245,
      "learning_rate": 2.4623826588586e-06,
      "loss": 0.81508267,
      "num_input_tokens_seen": 79314820,
      "step": 3684,
      "time_per_iteration": 2.635045289993286
    },
    {
      "auxiliary_loss_clip": 0.01188852,
      "auxiliary_loss_mlp": 0.01027745,
      "balance_loss_clip": 1.05021214,
      "balance_loss_mlp": 1.01863146,
      "epoch": 0.4430950520050502,
      "flos": 21614704738560.0,
      "grad_norm": 1.6807573712079975,
      "language_loss": 0.82864916,
      "learning_rate": 2.461624755821492e-06,
      "loss": 0.85081506,
      "num_input_tokens_seen": 79334300,
      "step": 3685,
      "time_per_iteration": 2.682161331176758
    },
    {
      "auxiliary_loss_clip": 0.0119505,
      "auxiliary_loss_mlp": 0.01026007,
      "balance_loss_clip": 1.05088556,
      "balance_loss_mlp": 1.01786518,
      "epoch": 0.4432152948956893,
      "flos": 24572271709440.0,
      "grad_norm": 1.832288751384694,
      "language_loss": 0.76380014,
      "learning_rate": 2.4608667827562763e-06,
      "loss": 0.7860108,
      "num_input_tokens_seen": 79353630,
      "step": 3686,
      "time_per_iteration": 2.6670472621917725
    },
    {
      "auxiliary_loss_clip": 0.01198964,
      "auxiliary_loss_mlp": 0.01029738,
      "balance_loss_clip": 1.05795264,
      "balance_loss_mlp": 1.02086866,
      "epoch": 0.4433355377863284,
      "flos": 21762261809280.0,
      "grad_norm": 2.0262455792799563,
      "language_loss": 0.9017061,
      "learning_rate": 2.460108739777936e-06,
      "loss": 0.92399311,
      "num_input_tokens_seen": 79372765,
      "step": 3687,
      "time_per_iteration": 2.622248649597168
    },
    {
      "auxiliary_loss_clip": 0.01191209,
      "auxiliary_loss_mlp": 0.01026447,
      "balance_loss_clip": 1.053496,
      "balance_loss_mlp": 1.0177927,
      "epoch": 0.44345578067696745,
      "flos": 20084479488000.0,
      "grad_norm": 2.2382378288009432,
      "language_loss": 0.76133275,
      "learning_rate": 2.4593506270014656e-06,
      "loss": 0.78350931,
      "num_input_tokens_seen": 79391735,
      "step": 3688,
      "time_per_iteration": 2.6422128677368164
    },
    {
      "auxiliary_loss_clip": 0.01197209,
      "auxiliary_loss_mlp": 0.01027982,
      "balance_loss_clip": 1.05088794,
      "balance_loss_mlp": 1.01933956,
      "epoch": 0.44357602356760656,
      "flos": 24169497528960.0,
      "grad_norm": 2.015988567659576,
      "language_loss": 0.81835473,
      "learning_rate": 2.45859244454187e-06,
      "loss": 0.84060669,
      "num_input_tokens_seen": 79411525,
      "step": 3689,
      "time_per_iteration": 2.6641440391540527
    },
    {
      "auxiliary_loss_clip": 0.01188789,
      "auxiliary_loss_mlp": 0.01027155,
      "balance_loss_clip": 1.05381477,
      "balance_loss_mlp": 1.0189414,
      "epoch": 0.44369626645824567,
      "flos": 22707717644160.0,
      "grad_norm": 1.6495499336218191,
      "language_loss": 0.66237736,
      "learning_rate": 2.4578341925141655e-06,
      "loss": 0.68453676,
      "num_input_tokens_seen": 79430740,
      "step": 3690,
      "time_per_iteration": 2.5738070011138916
    },
    {
      "auxiliary_loss_clip": 0.01200525,
      "auxiliary_loss_mlp": 0.01026892,
      "balance_loss_clip": 1.05604792,
      "balance_loss_mlp": 1.01798725,
      "epoch": 0.4438165093488847,
      "flos": 38030225420160.0,
      "grad_norm": 2.1994263064416844,
      "language_loss": 0.72621036,
      "learning_rate": 2.457075871033378e-06,
      "loss": 0.74848449,
      "num_input_tokens_seen": 79452615,
      "step": 3691,
      "time_per_iteration": 2.744318962097168
    },
    {
      "auxiliary_loss_clip": 0.01195055,
      "auxiliary_loss_mlp": 0.01024367,
      "balance_loss_clip": 1.05244887,
      "balance_loss_mlp": 1.01593876,
      "epoch": 0.44393675223952384,
      "flos": 15523213996800.0,
      "grad_norm": 2.456890813635064,
      "language_loss": 0.88423133,
      "learning_rate": 2.4563174802145445e-06,
      "loss": 0.9064256,
      "num_input_tokens_seen": 79469865,
      "step": 3692,
      "time_per_iteration": 2.673888683319092
    },
    {
      "auxiliary_loss_clip": 0.0109685,
      "auxiliary_loss_mlp": 0.01002001,
      "balance_loss_clip": 1.01990354,
      "balance_loss_mlp": 1.00107718,
      "epoch": 0.44405699513016295,
      "flos": 64574893779840.0,
      "grad_norm": 0.6312246772570848,
      "language_loss": 0.48551837,
      "learning_rate": 2.455559020172712e-06,
      "loss": 0.50650686,
      "num_input_tokens_seen": 79537220,
      "step": 3693,
      "time_per_iteration": 3.2595207691192627
    },
    {
      "auxiliary_loss_clip": 0.01201571,
      "auxiliary_loss_mlp": 0.01036738,
      "balance_loss_clip": 1.05380666,
      "balance_loss_mlp": 1.02799439,
      "epoch": 0.444177238020802,
      "flos": 23987394552960.0,
      "grad_norm": 1.717888359719349,
      "language_loss": 0.89904839,
      "learning_rate": 2.4548004910229385e-06,
      "loss": 0.92143154,
      "num_input_tokens_seen": 79554795,
      "step": 3694,
      "time_per_iteration": 2.738006353378296
    },
    {
      "auxiliary_loss_clip": 0.01194407,
      "auxiliary_loss_mlp": 0.01209172,
      "balance_loss_clip": 1.0556643,
      "balance_loss_mlp": 1.0007534,
      "epoch": 0.4442974809114411,
      "flos": 22563069575040.0,
      "grad_norm": 1.8958179548361551,
      "language_loss": 0.87361693,
      "learning_rate": 2.4540418928802913e-06,
      "loss": 0.89765275,
      "num_input_tokens_seen": 79573530,
      "step": 3695,
      "time_per_iteration": 2.665844678878784
    },
    {
      "auxiliary_loss_clip": 0.01195603,
      "auxiliary_loss_mlp": 0.0103175,
      "balance_loss_clip": 1.05333829,
      "balance_loss_mlp": 1.02243924,
      "epoch": 0.4444177238020802,
      "flos": 17675699483520.0,
      "grad_norm": 2.20218542046509,
      "language_loss": 0.66360998,
      "learning_rate": 2.4532832258598506e-06,
      "loss": 0.68588352,
      "num_input_tokens_seen": 79591360,
      "step": 3696,
      "time_per_iteration": 3.596374273300171
    },
    {
      "auxiliary_loss_clip": 0.01186651,
      "auxiliary_loss_mlp": 0.01026296,
      "balance_loss_clip": 1.05589843,
      "balance_loss_mlp": 1.01746845,
      "epoch": 0.4445379666927193,
      "flos": 28621594609920.0,
      "grad_norm": 1.8781262829429843,
      "language_loss": 0.80946207,
      "learning_rate": 2.4525244900767047e-06,
      "loss": 0.83159149,
      "num_input_tokens_seen": 79612175,
      "step": 3697,
      "time_per_iteration": 3.7054128646850586
    },
    {
      "auxiliary_loss_clip": 0.01088773,
      "auxiliary_loss_mlp": 0.01002658,
      "balance_loss_clip": 1.02402139,
      "balance_loss_mlp": 1.00182331,
      "epoch": 0.4446582095833584,
      "flos": 70487370115200.0,
      "grad_norm": 0.7681700016643958,
      "language_loss": 0.60504347,
      "learning_rate": 2.4517656856459536e-06,
      "loss": 0.62595779,
      "num_input_tokens_seen": 79678020,
      "step": 3698,
      "time_per_iteration": 3.2551376819610596
    },
    {
      "auxiliary_loss_clip": 0.01193012,
      "auxiliary_loss_mlp": 0.01032414,
      "balance_loss_clip": 1.05390024,
      "balance_loss_mlp": 1.02397954,
      "epoch": 0.4447784524739975,
      "flos": 26505199313280.0,
      "grad_norm": 1.7042467747886683,
      "language_loss": 0.68234688,
      "learning_rate": 2.4510068126827073e-06,
      "loss": 0.70460111,
      "num_input_tokens_seen": 79699020,
      "step": 3699,
      "time_per_iteration": 3.629183530807495
    },
    {
      "auxiliary_loss_clip": 0.01193115,
      "auxiliary_loss_mlp": 0.0103364,
      "balance_loss_clip": 1.05206442,
      "balance_loss_mlp": 1.02508044,
      "epoch": 0.44489869536463655,
      "flos": 11656209553920.0,
      "grad_norm": 3.1261094431625813,
      "language_loss": 0.81981111,
      "learning_rate": 2.450247871302086e-06,
      "loss": 0.84207863,
      "num_input_tokens_seen": 79716795,
      "step": 3700,
      "time_per_iteration": 2.660122871398926
    },
    {
      "auxiliary_loss_clip": 0.01196234,
      "auxiliary_loss_mlp": 0.01028486,
      "balance_loss_clip": 1.05516553,
      "balance_loss_mlp": 1.02017713,
      "epoch": 0.44501893825527566,
      "flos": 20448469958400.0,
      "grad_norm": 4.546961129485331,
      "language_loss": 0.83431238,
      "learning_rate": 2.44948886161922e-06,
      "loss": 0.85655957,
      "num_input_tokens_seen": 79735810,
      "step": 3701,
      "time_per_iteration": 2.5840649604797363
    },
    {
      "auxiliary_loss_clip": 0.01193914,
      "auxiliary_loss_mlp": 0.01026597,
      "balance_loss_clip": 1.05525982,
      "balance_loss_mlp": 1.01853275,
      "epoch": 0.4451391811459148,
      "flos": 18261079430400.0,
      "grad_norm": 2.0294484683309166,
      "language_loss": 0.84994525,
      "learning_rate": 2.4487297837492524e-06,
      "loss": 0.87215042,
      "num_input_tokens_seen": 79754975,
      "step": 3702,
      "time_per_iteration": 2.637495994567871
    },
    {
      "auxiliary_loss_clip": 0.0119371,
      "auxiliary_loss_mlp": 0.01027599,
      "balance_loss_clip": 1.05113232,
      "balance_loss_mlp": 1.01902187,
      "epoch": 0.44525942403655383,
      "flos": 16910155895040.0,
      "grad_norm": 2.488469181846599,
      "language_loss": 0.62413919,
      "learning_rate": 2.4479706378073323e-06,
      "loss": 0.64635223,
      "num_input_tokens_seen": 79773515,
      "step": 3703,
      "time_per_iteration": 2.6466033458709717
    },
    {
      "auxiliary_loss_clip": 0.01185057,
      "auxiliary_loss_mlp": 0.01029723,
      "balance_loss_clip": 1.04750848,
      "balance_loss_mlp": 1.02121758,
      "epoch": 0.44537966692719294,
      "flos": 23258838994560.0,
      "grad_norm": 1.5192081922223837,
      "language_loss": 0.83904672,
      "learning_rate": 2.447211423908623e-06,
      "loss": 0.86119449,
      "num_input_tokens_seen": 79793560,
      "step": 3704,
      "time_per_iteration": 2.7082481384277344
    },
    {
      "auxiliary_loss_clip": 0.01193611,
      "auxiliary_loss_mlp": 0.01029581,
      "balance_loss_clip": 1.05399919,
      "balance_loss_mlp": 1.02115858,
      "epoch": 0.445499909817832,
      "flos": 21724160457600.0,
      "grad_norm": 9.733787375514126,
      "language_loss": 0.75319231,
      "learning_rate": 2.4464521421682966e-06,
      "loss": 0.77542412,
      "num_input_tokens_seen": 79811150,
      "step": 3705,
      "time_per_iteration": 3.5882959365844727
    },
    {
      "auxiliary_loss_clip": 0.01185133,
      "auxiliary_loss_mlp": 0.01024216,
      "balance_loss_clip": 1.05449915,
      "balance_loss_mlp": 1.0163306,
      "epoch": 0.4456201527084711,
      "flos": 23987969170560.0,
      "grad_norm": 2.1849508697348132,
      "language_loss": 0.87694407,
      "learning_rate": 2.4456927927015345e-06,
      "loss": 0.8990376,
      "num_input_tokens_seen": 79832190,
      "step": 3706,
      "time_per_iteration": 2.670590877532959
    },
    {
      "auxiliary_loss_clip": 0.01199462,
      "auxiliary_loss_mlp": 0.0103037,
      "balance_loss_clip": 1.05540276,
      "balance_loss_mlp": 1.02079177,
      "epoch": 0.4457403955991102,
      "flos": 18807065136000.0,
      "grad_norm": 3.783685486600121,
      "language_loss": 0.7663821,
      "learning_rate": 2.4449333756235307e-06,
      "loss": 0.78868043,
      "num_input_tokens_seen": 79848905,
      "step": 3707,
      "time_per_iteration": 2.666323184967041
    },
    {
      "auxiliary_loss_clip": 0.01197741,
      "auxiliary_loss_mlp": 0.01034496,
      "balance_loss_clip": 1.05683327,
      "balance_loss_mlp": 1.02554917,
      "epoch": 0.4458606384897493,
      "flos": 19207756327680.0,
      "grad_norm": 2.247222862864269,
      "language_loss": 0.78646117,
      "learning_rate": 2.4441738910494876e-06,
      "loss": 0.80878353,
      "num_input_tokens_seen": 79863640,
      "step": 3708,
      "time_per_iteration": 2.628089189529419
    },
    {
      "auxiliary_loss_clip": 0.01200354,
      "auxiliary_loss_mlp": 0.01033871,
      "balance_loss_clip": 1.05193257,
      "balance_loss_mlp": 1.02543664,
      "epoch": 0.4459808813803884,
      "flos": 21361283308800.0,
      "grad_norm": 1.8046191141939683,
      "language_loss": 0.8231923,
      "learning_rate": 2.4434143390946176e-06,
      "loss": 0.84553456,
      "num_input_tokens_seen": 79882450,
      "step": 3709,
      "time_per_iteration": 2.753232717514038
    },
    {
      "auxiliary_loss_clip": 0.01194107,
      "auxiliary_loss_mlp": 0.01029442,
      "balance_loss_clip": 1.05073249,
      "balance_loss_mlp": 1.02085257,
      "epoch": 0.4461011242710275,
      "flos": 23288967527040.0,
      "grad_norm": 1.9006301832803514,
      "language_loss": 0.85920942,
      "learning_rate": 2.4426547198741457e-06,
      "loss": 0.88144487,
      "num_input_tokens_seen": 79900655,
      "step": 3710,
      "time_per_iteration": 2.696831226348877
    },
    {
      "auxiliary_loss_clip": 0.01197858,
      "auxiliary_loss_mlp": 0.01033899,
      "balance_loss_clip": 1.05248332,
      "balance_loss_mlp": 1.02535808,
      "epoch": 0.44622136716166655,
      "flos": 20193001453440.0,
      "grad_norm": 2.158295569330678,
      "language_loss": 0.74872732,
      "learning_rate": 2.441895033503305e-06,
      "loss": 0.77104491,
      "num_input_tokens_seen": 79918575,
      "step": 3711,
      "time_per_iteration": 2.6902074813842773
    },
    {
      "auxiliary_loss_clip": 0.01189725,
      "auxiliary_loss_mlp": 0.01029985,
      "balance_loss_clip": 1.05402565,
      "balance_loss_mlp": 1.0206449,
      "epoch": 0.44634161005230566,
      "flos": 21283033530240.0,
      "grad_norm": 1.6891225937346221,
      "language_loss": 0.82222831,
      "learning_rate": 2.4411352800973375e-06,
      "loss": 0.84442538,
      "num_input_tokens_seen": 79937010,
      "step": 3712,
      "time_per_iteration": 2.6898787021636963
    },
    {
      "auxiliary_loss_clip": 0.01191362,
      "auxiliary_loss_mlp": 0.01028873,
      "balance_loss_clip": 1.05000854,
      "balance_loss_mlp": 1.01967633,
      "epoch": 0.44646185294294477,
      "flos": 22929358515840.0,
      "grad_norm": 4.84751376783949,
      "language_loss": 0.75362742,
      "learning_rate": 2.4403754597715005e-06,
      "loss": 0.77582985,
      "num_input_tokens_seen": 79956455,
      "step": 3713,
      "time_per_iteration": 2.7033379077911377
    },
    {
      "auxiliary_loss_clip": 0.0119584,
      "auxiliary_loss_mlp": 0.01033783,
      "balance_loss_clip": 1.04860353,
      "balance_loss_mlp": 1.02391887,
      "epoch": 0.4465820958335838,
      "flos": 22637692080000.0,
      "grad_norm": 2.2075113888139004,
      "language_loss": 0.9256773,
      "learning_rate": 2.4396155726410553e-06,
      "loss": 0.94797361,
      "num_input_tokens_seen": 79975065,
      "step": 3714,
      "time_per_iteration": 2.664919376373291
    },
    {
      "auxiliary_loss_clip": 0.01196472,
      "auxiliary_loss_mlp": 0.01025334,
      "balance_loss_clip": 1.05340278,
      "balance_loss_mlp": 1.01663816,
      "epoch": 0.44670233872422294,
      "flos": 22672525294080.0,
      "grad_norm": 5.628542734398061,
      "language_loss": 0.91000247,
      "learning_rate": 2.438855618821278e-06,
      "loss": 0.93222058,
      "num_input_tokens_seen": 79990865,
      "step": 3715,
      "time_per_iteration": 2.598297357559204
    },
    {
      "auxiliary_loss_clip": 0.01181521,
      "auxiliary_loss_mlp": 0.01028998,
      "balance_loss_clip": 1.04934907,
      "balance_loss_mlp": 1.01994419,
      "epoch": 0.44682258161486205,
      "flos": 23582178247680.0,
      "grad_norm": 2.869270333261835,
      "language_loss": 0.67447507,
      "learning_rate": 2.4380955984274517e-06,
      "loss": 0.69658029,
      "num_input_tokens_seen": 80009520,
      "step": 3716,
      "time_per_iteration": 2.608450412750244
    },
    {
      "auxiliary_loss_clip": 0.01190293,
      "auxiliary_loss_mlp": 0.01035218,
      "balance_loss_clip": 1.05296302,
      "balance_loss_mlp": 1.02648652,
      "epoch": 0.4469428245055011,
      "flos": 26501356558080.0,
      "grad_norm": 4.453449415915546,
      "language_loss": 0.77763164,
      "learning_rate": 2.4373355115748716e-06,
      "loss": 0.79988676,
      "num_input_tokens_seen": 80030350,
      "step": 3717,
      "time_per_iteration": 2.656062602996826
    },
    {
      "auxiliary_loss_clip": 0.01188047,
      "auxiliary_loss_mlp": 0.01032464,
      "balance_loss_clip": 1.05358577,
      "balance_loss_mlp": 1.0235467,
      "epoch": 0.4470630673961402,
      "flos": 21504925797120.0,
      "grad_norm": 1.744550381109619,
      "language_loss": 0.72387886,
      "learning_rate": 2.436575358378842e-06,
      "loss": 0.74608392,
      "num_input_tokens_seen": 80049840,
      "step": 3718,
      "time_per_iteration": 2.6310200691223145
    },
    {
      "auxiliary_loss_clip": 0.01203341,
      "auxiliary_loss_mlp": 0.01028802,
      "balance_loss_clip": 1.05533838,
      "balance_loss_mlp": 1.01875234,
      "epoch": 0.44718331028677927,
      "flos": 16173986653440.0,
      "grad_norm": 2.558220534869131,
      "language_loss": 0.83216244,
      "learning_rate": 2.4358151389546782e-06,
      "loss": 0.8544839,
      "num_input_tokens_seen": 80066525,
      "step": 3719,
      "time_per_iteration": 2.621236562728882
    },
    {
      "auxiliary_loss_clip": 0.01189148,
      "auxiliary_loss_mlp": 0.01032087,
      "balance_loss_clip": 1.05643916,
      "balance_loss_mlp": 1.02318192,
      "epoch": 0.4473035531774184,
      "flos": 19681238430720.0,
      "grad_norm": 5.382193697728989,
      "language_loss": 0.75663865,
      "learning_rate": 2.4350548534177035e-06,
      "loss": 0.77885103,
      "num_input_tokens_seen": 80083355,
      "step": 3720,
      "time_per_iteration": 2.6168949604034424
    },
    {
      "auxiliary_loss_clip": 0.0119711,
      "auxiliary_loss_mlp": 0.01033777,
      "balance_loss_clip": 1.05237889,
      "balance_loss_mlp": 1.02546239,
      "epoch": 0.4474237960680575,
      "flos": 41427590515200.0,
      "grad_norm": 1.514889344427223,
      "language_loss": 0.66532183,
      "learning_rate": 2.434294501883254e-06,
      "loss": 0.68763077,
      "num_input_tokens_seen": 80106450,
      "step": 3721,
      "time_per_iteration": 2.8433661460876465
    },
    {
      "auxiliary_loss_clip": 0.01184737,
      "auxiliary_loss_mlp": 0.01029787,
      "balance_loss_clip": 1.04964852,
      "balance_loss_mlp": 1.02049446,
      "epoch": 0.44754403895869654,
      "flos": 22891328991360.0,
      "grad_norm": 3.058212976075888,
      "language_loss": 0.65553331,
      "learning_rate": 2.433534084466674e-06,
      "loss": 0.67767859,
      "num_input_tokens_seen": 80125670,
      "step": 3722,
      "time_per_iteration": 3.5923783779144287
    },
    {
      "auxiliary_loss_clip": 0.01185903,
      "auxiliary_loss_mlp": 0.01027705,
      "balance_loss_clip": 1.05567312,
      "balance_loss_mlp": 1.01920509,
      "epoch": 0.44766428184933565,
      "flos": 25630271832960.0,
      "grad_norm": 2.187452041666587,
      "language_loss": 0.70657212,
      "learning_rate": 2.4327736012833178e-06,
      "loss": 0.72870815,
      "num_input_tokens_seen": 80147390,
      "step": 3723,
      "time_per_iteration": 3.58229660987854
    },
    {
      "auxiliary_loss_clip": 0.01192561,
      "auxiliary_loss_mlp": 0.01032248,
      "balance_loss_clip": 1.05533767,
      "balance_loss_mlp": 1.0236764,
      "epoch": 0.44778452473997477,
      "flos": 20448972748800.0,
      "grad_norm": 4.862804001952474,
      "language_loss": 0.76549995,
      "learning_rate": 2.4320130524485506e-06,
      "loss": 0.78774804,
      "num_input_tokens_seen": 80166185,
      "step": 3724,
      "time_per_iteration": 2.648334264755249
    },
    {
      "auxiliary_loss_clip": 0.01187752,
      "auxiliary_loss_mlp": 0.01026645,
      "balance_loss_clip": 1.0556078,
      "balance_loss_mlp": 1.01901019,
      "epoch": 0.4479047676306138,
      "flos": 21975462984960.0,
      "grad_norm": 1.6647636271840502,
      "language_loss": 0.79780006,
      "learning_rate": 2.431252438077746e-06,
      "loss": 0.81994408,
      "num_input_tokens_seen": 80185685,
      "step": 3725,
      "time_per_iteration": 2.65596866607666
    },
    {
      "auxiliary_loss_clip": 0.01194903,
      "auxiliary_loss_mlp": 0.01209233,
      "balance_loss_clip": 1.053725,
      "balance_loss_mlp": 1.00082994,
      "epoch": 0.44802501052125293,
      "flos": 21467219495040.0,
      "grad_norm": 2.247833427092321,
      "language_loss": 0.76831889,
      "learning_rate": 2.4304917582862906e-06,
      "loss": 0.79236019,
      "num_input_tokens_seen": 80204865,
      "step": 3726,
      "time_per_iteration": 3.5324156284332275
    },
    {
      "auxiliary_loss_clip": 0.01188651,
      "auxiliary_loss_mlp": 0.01029747,
      "balance_loss_clip": 1.05599189,
      "balance_loss_mlp": 1.02124143,
      "epoch": 0.44814525341189204,
      "flos": 22126970551680.0,
      "grad_norm": 2.1340745436474005,
      "language_loss": 0.88045681,
      "learning_rate": 2.4297310131895774e-06,
      "loss": 0.90264082,
      "num_input_tokens_seen": 80223410,
      "step": 3727,
      "time_per_iteration": 2.6875712871551514
    },
    {
      "auxiliary_loss_clip": 0.01191761,
      "auxiliary_loss_mlp": 0.01030814,
      "balance_loss_clip": 1.05405664,
      "balance_loss_mlp": 1.02176619,
      "epoch": 0.4482654963025311,
      "flos": 16653933204480.0,
      "grad_norm": 3.3284036065046427,
      "language_loss": 0.74399352,
      "learning_rate": 2.4289702029030113e-06,
      "loss": 0.76621926,
      "num_input_tokens_seen": 80240880,
      "step": 3728,
      "time_per_iteration": 2.6138036251068115
    },
    {
      "auxiliary_loss_clip": 0.01193397,
      "auxiliary_loss_mlp": 0.01028536,
      "balance_loss_clip": 1.05837893,
      "balance_loss_mlp": 1.01981616,
      "epoch": 0.4483857391931702,
      "flos": 18841251905280.0,
      "grad_norm": 1.8972650497752235,
      "language_loss": 0.82950932,
      "learning_rate": 2.4282093275420057e-06,
      "loss": 0.85172868,
      "num_input_tokens_seen": 80259910,
      "step": 3729,
      "time_per_iteration": 2.606705665588379
    },
    {
      "auxiliary_loss_clip": 0.0119749,
      "auxiliary_loss_mlp": 0.01030536,
      "balance_loss_clip": 1.05689967,
      "balance_loss_mlp": 1.02225721,
      "epoch": 0.4485059820838093,
      "flos": 20372590477440.0,
      "grad_norm": 2.5490544407821116,
      "language_loss": 0.70957613,
      "learning_rate": 2.4274483872219863e-06,
      "loss": 0.73185647,
      "num_input_tokens_seen": 80277270,
      "step": 3730,
      "time_per_iteration": 2.578274965286255
    },
    {
      "auxiliary_loss_clip": 0.01189038,
      "auxiliary_loss_mlp": 0.01029992,
      "balance_loss_clip": 1.05404878,
      "balance_loss_mlp": 1.02147424,
      "epoch": 0.4486262249744484,
      "flos": 20047742853120.0,
      "grad_norm": 3.8091821650824573,
      "language_loss": 0.93701631,
      "learning_rate": 2.426687382058386e-06,
      "loss": 0.95920658,
      "num_input_tokens_seen": 80295550,
      "step": 3731,
      "time_per_iteration": 2.6997315883636475
    },
    {
      "auxiliary_loss_clip": 0.01091021,
      "auxiliary_loss_mlp": 0.01000926,
      "balance_loss_clip": 1.02609742,
      "balance_loss_mlp": 1.00001419,
      "epoch": 0.4487464678650875,
      "flos": 64595684776320.0,
      "grad_norm": 0.8654656492669949,
      "language_loss": 0.59835517,
      "learning_rate": 2.425926312166649e-06,
      "loss": 0.61927468,
      "num_input_tokens_seen": 80348425,
      "step": 3732,
      "time_per_iteration": 3.925583600997925
    },
    {
      "auxiliary_loss_clip": 0.01198238,
      "auxiliary_loss_mlp": 0.01025575,
      "balance_loss_clip": 1.05531931,
      "balance_loss_mlp": 1.01642537,
      "epoch": 0.4488667107557266,
      "flos": 20769798049920.0,
      "grad_norm": 2.5837890200354114,
      "language_loss": 0.73424095,
      "learning_rate": 2.42516517766223e-06,
      "loss": 0.75647914,
      "num_input_tokens_seen": 80366505,
      "step": 3733,
      "time_per_iteration": 2.576326370239258
    },
    {
      "auxiliary_loss_clip": 0.01190415,
      "auxiliary_loss_mlp": 0.01025375,
      "balance_loss_clip": 1.05906594,
      "balance_loss_mlp": 1.01656544,
      "epoch": 0.44898695364636565,
      "flos": 23951735326080.0,
      "grad_norm": 2.5522664008599616,
      "language_loss": 0.6828593,
      "learning_rate": 2.4244039786605907e-06,
      "loss": 0.70501715,
      "num_input_tokens_seen": 80387510,
      "step": 3734,
      "time_per_iteration": 2.628422498703003
    },
    {
      "auxiliary_loss_clip": 0.01198309,
      "auxiliary_loss_mlp": 0.01029123,
      "balance_loss_clip": 1.04895449,
      "balance_loss_mlp": 1.020087,
      "epoch": 0.44910719653700476,
      "flos": 18624351628800.0,
      "grad_norm": 2.7642319731342035,
      "language_loss": 0.82318282,
      "learning_rate": 2.4236427152772055e-06,
      "loss": 0.84545714,
      "num_input_tokens_seen": 80405915,
      "step": 3735,
      "time_per_iteration": 2.6216201782226562
    },
    {
      "auxiliary_loss_clip": 0.01091128,
      "auxiliary_loss_mlp": 0.01000687,
      "balance_loss_clip": 1.02041519,
      "balance_loss_mlp": 0.99977511,
      "epoch": 0.4492274394276438,
      "flos": 57033435749760.0,
      "grad_norm": 0.8434925706765969,
      "language_loss": 0.57409024,
      "learning_rate": 2.422881387627557e-06,
      "loss": 0.59500837,
      "num_input_tokens_seen": 80458365,
      "step": 3736,
      "time_per_iteration": 2.9706263542175293
    },
    {
      "auxiliary_loss_clip": 0.01197723,
      "auxiliary_loss_mlp": 0.01023931,
      "balance_loss_clip": 1.05575264,
      "balance_loss_mlp": 1.01519275,
      "epoch": 0.4493476823182829,
      "flos": 23254888498560.0,
      "grad_norm": 6.147995520537972,
      "language_loss": 0.77293915,
      "learning_rate": 2.422119995827139e-06,
      "loss": 0.7951557,
      "num_input_tokens_seen": 80478490,
      "step": 3737,
      "time_per_iteration": 2.6812400817871094
    },
    {
      "auxiliary_loss_clip": 0.01197877,
      "auxiliary_loss_mlp": 0.01024022,
      "balance_loss_clip": 1.05789483,
      "balance_loss_mlp": 1.01567149,
      "epoch": 0.44946792520892204,
      "flos": 15815131827840.0,
      "grad_norm": 3.0203075690737267,
      "language_loss": 0.74173784,
      "learning_rate": 2.4213585399914528e-06,
      "loss": 0.76395684,
      "num_input_tokens_seen": 80495695,
      "step": 3738,
      "time_per_iteration": 2.6513724327087402
    },
    {
      "auxiliary_loss_clip": 0.01191326,
      "auxiliary_loss_mlp": 0.01023277,
      "balance_loss_clip": 1.05568385,
      "balance_loss_mlp": 1.01510501,
      "epoch": 0.4495881680995611,
      "flos": 19610063631360.0,
      "grad_norm": 1.7652179393524718,
      "language_loss": 0.85315692,
      "learning_rate": 2.4205970202360113e-06,
      "loss": 0.87530291,
      "num_input_tokens_seen": 80515260,
      "step": 3739,
      "time_per_iteration": 2.5843582153320312
    },
    {
      "auxiliary_loss_clip": 0.01188044,
      "auxiliary_loss_mlp": 0.01027854,
      "balance_loss_clip": 1.04930973,
      "balance_loss_mlp": 1.01887798,
      "epoch": 0.4497084109902002,
      "flos": 26031465815040.0,
      "grad_norm": 1.9510842599682583,
      "language_loss": 0.77895272,
      "learning_rate": 2.4198354366763354e-06,
      "loss": 0.8011117,
      "num_input_tokens_seen": 80533900,
      "step": 3740,
      "time_per_iteration": 2.7058024406433105
    },
    {
      "auxiliary_loss_clip": 0.01198139,
      "auxiliary_loss_mlp": 0.0102761,
      "balance_loss_clip": 1.0547092,
      "balance_loss_mlp": 1.01892591,
      "epoch": 0.4498286538808393,
      "flos": 14793688771200.0,
      "grad_norm": 2.3573839696386822,
      "language_loss": 0.78563988,
      "learning_rate": 2.4190737894279587e-06,
      "loss": 0.80789733,
      "num_input_tokens_seen": 80551270,
      "step": 3741,
      "time_per_iteration": 2.5737388134002686
    },
    {
      "auxiliary_loss_clip": 0.01185375,
      "auxiliary_loss_mlp": 0.0102441,
      "balance_loss_clip": 1.04652333,
      "balance_loss_mlp": 1.01638138,
      "epoch": 0.44994889677147837,
      "flos": 15450171690240.0,
      "grad_norm": 2.0905233788107305,
      "language_loss": 0.80544531,
      "learning_rate": 2.4183120786064203e-06,
      "loss": 0.82754314,
      "num_input_tokens_seen": 80568145,
      "step": 3742,
      "time_per_iteration": 2.6196024417877197
    },
    {
      "auxiliary_loss_clip": 0.01191354,
      "auxiliary_loss_mlp": 0.01208898,
      "balance_loss_clip": 1.05716383,
      "balance_loss_mlp": 1.00090241,
      "epoch": 0.4500691396621175,
      "flos": 21798316085760.0,
      "grad_norm": 2.9244158738252253,
      "language_loss": 0.85716385,
      "learning_rate": 2.417550304327273e-06,
      "loss": 0.88116634,
      "num_input_tokens_seen": 80586185,
      "step": 3743,
      "time_per_iteration": 2.6339995861053467
    },
    {
      "auxiliary_loss_clip": 0.01192326,
      "auxiliary_loss_mlp": 0.01029321,
      "balance_loss_clip": 1.05780876,
      "balance_loss_mlp": 1.02025533,
      "epoch": 0.4501893825527566,
      "flos": 32382016421760.0,
      "grad_norm": 1.4932747732067342,
      "language_loss": 0.75829464,
      "learning_rate": 2.4167884667060763e-06,
      "loss": 0.78051108,
      "num_input_tokens_seen": 80608895,
      "step": 3744,
      "time_per_iteration": 2.6753523349761963
    },
    {
      "auxiliary_loss_clip": 0.01195216,
      "auxiliary_loss_mlp": 0.010304,
      "balance_loss_clip": 1.05327344,
      "balance_loss_mlp": 1.0215373,
      "epoch": 0.45030962544339564,
      "flos": 16544944362240.0,
      "grad_norm": 6.116479588426653,
      "language_loss": 0.87428194,
      "learning_rate": 2.4160265658584e-06,
      "loss": 0.89653808,
      "num_input_tokens_seen": 80623785,
      "step": 3745,
      "time_per_iteration": 2.5537519454956055
    },
    {
      "auxiliary_loss_clip": 0.01196259,
      "auxiliary_loss_mlp": 0.01026387,
      "balance_loss_clip": 1.05629718,
      "balance_loss_mlp": 1.01777983,
      "epoch": 0.45042986833403476,
      "flos": 19573039687680.0,
      "grad_norm": 2.645887510505581,
      "language_loss": 0.68822742,
      "learning_rate": 2.4152646018998253e-06,
      "loss": 0.71045387,
      "num_input_tokens_seen": 80642735,
      "step": 3746,
      "time_per_iteration": 2.5129287242889404
    },
    {
      "auxiliary_loss_clip": 0.01189514,
      "auxiliary_loss_mlp": 0.01030835,
      "balance_loss_clip": 1.05420232,
      "balance_loss_mlp": 1.02240729,
      "epoch": 0.45055011122467387,
      "flos": 23112467072640.0,
      "grad_norm": 2.6150862586828585,
      "language_loss": 0.72465241,
      "learning_rate": 2.4145025749459403e-06,
      "loss": 0.74685585,
      "num_input_tokens_seen": 80663760,
      "step": 3747,
      "time_per_iteration": 2.5889110565185547
    },
    {
      "auxiliary_loss_clip": 0.01201828,
      "auxiliary_loss_mlp": 0.01035871,
      "balance_loss_clip": 1.04782033,
      "balance_loss_mlp": 1.02644765,
      "epoch": 0.4506703541153129,
      "flos": 19934623946880.0,
      "grad_norm": 3.1839858208777763,
      "language_loss": 0.69823128,
      "learning_rate": 2.413740485112344e-06,
      "loss": 0.72060835,
      "num_input_tokens_seen": 80682100,
      "step": 3748,
      "time_per_iteration": 2.647794246673584
    },
    {
      "auxiliary_loss_clip": 0.01187926,
      "auxiliary_loss_mlp": 0.01026798,
      "balance_loss_clip": 1.05521214,
      "balance_loss_mlp": 1.01785719,
      "epoch": 0.45079059700595203,
      "flos": 19499530504320.0,
      "grad_norm": 1.7990183773421868,
      "language_loss": 0.82177621,
      "learning_rate": 2.412978332514646e-06,
      "loss": 0.84392345,
      "num_input_tokens_seen": 80700880,
      "step": 3749,
      "time_per_iteration": 3.599919557571411
    },
    {
      "auxiliary_loss_clip": 0.01196734,
      "auxiliary_loss_mlp": 0.01030782,
      "balance_loss_clip": 1.05418301,
      "balance_loss_mlp": 1.02113175,
      "epoch": 0.4509108398965911,
      "flos": 27636313570560.0,
      "grad_norm": 2.077888854467765,
      "language_loss": 0.724024,
      "learning_rate": 2.4122161172684623e-06,
      "loss": 0.74629909,
      "num_input_tokens_seen": 80721675,
      "step": 3750,
      "time_per_iteration": 3.642345666885376
    },
    {
      "auxiliary_loss_clip": 0.01197188,
      "auxiliary_loss_mlp": 0.01033924,
      "balance_loss_clip": 1.05449784,
      "balance_loss_mlp": 1.02489984,
      "epoch": 0.4510310827872302,
      "flos": 20995712640000.0,
      "grad_norm": 5.607260140574375,
      "language_loss": 0.84637117,
      "learning_rate": 2.4114538394894216e-06,
      "loss": 0.86868227,
      "num_input_tokens_seen": 80739315,
      "step": 3751,
      "time_per_iteration": 2.6490185260772705
    },
    {
      "auxiliary_loss_clip": 0.01191596,
      "auxiliary_loss_mlp": 0.01024264,
      "balance_loss_clip": 1.0498302,
      "balance_loss_mlp": 1.01596165,
      "epoch": 0.4511513256778693,
      "flos": 16216684945920.0,
      "grad_norm": 1.8729716992757393,
      "language_loss": 0.83317059,
      "learning_rate": 2.410691499293161e-06,
      "loss": 0.85532922,
      "num_input_tokens_seen": 80757470,
      "step": 3752,
      "time_per_iteration": 2.6946749687194824
    },
    {
      "auxiliary_loss_clip": 0.01189725,
      "auxiliary_loss_mlp": 0.01025987,
      "balance_loss_clip": 1.05355394,
      "balance_loss_mlp": 1.01729107,
      "epoch": 0.45127156856850836,
      "flos": 25186702780800.0,
      "grad_norm": 1.7229645959692437,
      "language_loss": 0.73985195,
      "learning_rate": 2.409929096795326e-06,
      "loss": 0.76200902,
      "num_input_tokens_seen": 80777840,
      "step": 3753,
      "time_per_iteration": 3.633788824081421
    },
    {
      "auxiliary_loss_clip": 0.01192105,
      "auxiliary_loss_mlp": 0.01028194,
      "balance_loss_clip": 1.05439472,
      "balance_loss_mlp": 1.01900315,
      "epoch": 0.4513918114591475,
      "flos": 20412523422720.0,
      "grad_norm": 2.5671406046276917,
      "language_loss": 0.79069358,
      "learning_rate": 2.409166632111573e-06,
      "loss": 0.81289661,
      "num_input_tokens_seen": 80795975,
      "step": 3754,
      "time_per_iteration": 2.6653385162353516
    },
    {
      "auxiliary_loss_clip": 0.012003,
      "auxiliary_loss_mlp": 0.01025024,
      "balance_loss_clip": 1.05609834,
      "balance_loss_mlp": 1.01592803,
      "epoch": 0.4515120543497866,
      "flos": 26648482665600.0,
      "grad_norm": 2.5191164123020697,
      "language_loss": 0.80354428,
      "learning_rate": 2.4084041053575674e-06,
      "loss": 0.8257975,
      "num_input_tokens_seen": 80815395,
      "step": 3755,
      "time_per_iteration": 2.6221489906311035
    },
    {
      "auxiliary_loss_clip": 0.01200795,
      "auxiliary_loss_mlp": 0.01029716,
      "balance_loss_clip": 1.05673766,
      "balance_loss_mlp": 1.0210371,
      "epoch": 0.45163229724042564,
      "flos": 20595093275520.0,
      "grad_norm": 1.8387141468956445,
      "language_loss": 0.72900319,
      "learning_rate": 2.4076415166489834e-06,
      "loss": 0.75130826,
      "num_input_tokens_seen": 80834805,
      "step": 3756,
      "time_per_iteration": 2.653773307800293
    },
    {
      "auxiliary_loss_clip": 0.01205413,
      "auxiliary_loss_mlp": 0.01031724,
      "balance_loss_clip": 1.05183339,
      "balance_loss_mlp": 1.02318263,
      "epoch": 0.45175254013106475,
      "flos": 21689004021120.0,
      "grad_norm": 1.670994754858505,
      "language_loss": 0.78948379,
      "learning_rate": 2.406878866101506e-06,
      "loss": 0.81185508,
      "num_input_tokens_seen": 80853770,
      "step": 3757,
      "time_per_iteration": 2.657332181930542
    },
    {
      "auxiliary_loss_clip": 0.0119067,
      "auxiliary_loss_mlp": 0.0102795,
      "balance_loss_clip": 1.05912507,
      "balance_loss_mlp": 1.0198493,
      "epoch": 0.45187278302170386,
      "flos": 18878850466560.0,
      "grad_norm": 2.249579360173845,
      "language_loss": 0.78462195,
      "learning_rate": 2.4061161538308273e-06,
      "loss": 0.80680811,
      "num_input_tokens_seen": 80870615,
      "step": 3758,
      "time_per_iteration": 2.5453574657440186
    },
    {
      "auxiliary_loss_clip": 0.01192025,
      "auxiliary_loss_mlp": 0.01026836,
      "balance_loss_clip": 1.05695713,
      "balance_loss_mlp": 1.01795483,
      "epoch": 0.4519930259123429,
      "flos": 18582479349120.0,
      "grad_norm": 2.0055409595133646,
      "language_loss": 0.89164603,
      "learning_rate": 2.4053533799526523e-06,
      "loss": 0.91383463,
      "num_input_tokens_seen": 80886335,
      "step": 3759,
      "time_per_iteration": 3.5629422664642334
    },
    {
      "auxiliary_loss_clip": 0.01189775,
      "auxiliary_loss_mlp": 0.01031037,
      "balance_loss_clip": 1.05463886,
      "balance_loss_mlp": 1.02250731,
      "epoch": 0.452113268802982,
      "flos": 25192377129600.0,
      "grad_norm": 2.147861835568894,
      "language_loss": 0.86473942,
      "learning_rate": 2.404590544582691e-06,
      "loss": 0.88694751,
      "num_input_tokens_seen": 80904570,
      "step": 3760,
      "time_per_iteration": 2.6193466186523438
    },
    {
      "auxiliary_loss_clip": 0.0120061,
      "auxiliary_loss_mlp": 0.01030265,
      "balance_loss_clip": 1.04519689,
      "balance_loss_mlp": 1.02135992,
      "epoch": 0.45223351169362114,
      "flos": 39378922312320.0,
      "grad_norm": 1.6920740051683665,
      "language_loss": 0.80693501,
      "learning_rate": 2.403827647836666e-06,
      "loss": 0.82924372,
      "num_input_tokens_seen": 80925125,
      "step": 3761,
      "time_per_iteration": 2.7888059616088867
    },
    {
      "auxiliary_loss_clip": 0.01190219,
      "auxiliary_loss_mlp": 0.01033012,
      "balance_loss_clip": 1.05556989,
      "balance_loss_mlp": 1.02358246,
      "epoch": 0.4523537545842602,
      "flos": 21582169994880.0,
      "grad_norm": 2.2216341604808028,
      "language_loss": 0.6943835,
      "learning_rate": 2.4030646898303075e-06,
      "loss": 0.71661586,
      "num_input_tokens_seen": 80946615,
      "step": 3762,
      "time_per_iteration": 2.55013370513916
    },
    {
      "auxiliary_loss_clip": 0.0119908,
      "auxiliary_loss_mlp": 0.01035637,
      "balance_loss_clip": 1.05487442,
      "balance_loss_mlp": 1.02682185,
      "epoch": 0.4524739974748993,
      "flos": 28439527547520.0,
      "grad_norm": 2.2760669730131986,
      "language_loss": 0.81926405,
      "learning_rate": 2.4023016706793566e-06,
      "loss": 0.84161127,
      "num_input_tokens_seen": 80966410,
      "step": 3763,
      "time_per_iteration": 2.7544844150543213
    },
    {
      "auxiliary_loss_clip": 0.01116288,
      "auxiliary_loss_mlp": 0.01009447,
      "balance_loss_clip": 1.02786171,
      "balance_loss_mlp": 1.00848126,
      "epoch": 0.4525942403655384,
      "flos": 61556492148480.0,
      "grad_norm": 0.8263767403912022,
      "language_loss": 0.56857651,
      "learning_rate": 2.401538590499561e-06,
      "loss": 0.58983386,
      "num_input_tokens_seen": 81026865,
      "step": 3764,
      "time_per_iteration": 3.244706392288208
    },
    {
      "auxiliary_loss_clip": 0.0119624,
      "auxiliary_loss_mlp": 0.01209091,
      "balance_loss_clip": 1.05722117,
      "balance_loss_mlp": 1.00081313,
      "epoch": 0.45271448325617747,
      "flos": 27529838680320.0,
      "grad_norm": 1.9572250995186917,
      "language_loss": 0.7199977,
      "learning_rate": 2.400775449406682e-06,
      "loss": 0.7440511,
      "num_input_tokens_seen": 81050060,
      "step": 3765,
      "time_per_iteration": 2.7059309482574463
    },
    {
      "auxiliary_loss_clip": 0.01192869,
      "auxiliary_loss_mlp": 0.01028089,
      "balance_loss_clip": 1.0542469,
      "balance_loss_mlp": 1.01963127,
      "epoch": 0.4528347261468166,
      "flos": 22452608275200.0,
      "grad_norm": 1.746166278586982,
      "language_loss": 0.72958297,
      "learning_rate": 2.400012247516485e-06,
      "loss": 0.75179255,
      "num_input_tokens_seen": 81070625,
      "step": 3766,
      "time_per_iteration": 2.6240644454956055
    },
    {
      "auxiliary_loss_clip": 0.01197055,
      "auxiliary_loss_mlp": 0.01027818,
      "balance_loss_clip": 1.04767036,
      "balance_loss_mlp": 1.01908541,
      "epoch": 0.45295496903745563,
      "flos": 21103875469440.0,
      "grad_norm": 2.004121717081381,
      "language_loss": 0.90044069,
      "learning_rate": 2.3992489849447484e-06,
      "loss": 0.92268944,
      "num_input_tokens_seen": 81089080,
      "step": 3767,
      "time_per_iteration": 2.648682117462158
    },
    {
      "auxiliary_loss_clip": 0.0120459,
      "auxiliary_loss_mlp": 0.01028798,
      "balance_loss_clip": 1.05437016,
      "balance_loss_mlp": 1.02035224,
      "epoch": 0.45307521192809475,
      "flos": 23221168606080.0,
      "grad_norm": 1.6426724626480376,
      "language_loss": 0.79120606,
      "learning_rate": 2.3984856618072584e-06,
      "loss": 0.81353998,
      "num_input_tokens_seen": 81109115,
      "step": 3768,
      "time_per_iteration": 2.6615138053894043
    },
    {
      "auxiliary_loss_clip": 0.01201063,
      "auxiliary_loss_mlp": 0.01030913,
      "balance_loss_clip": 1.05314744,
      "balance_loss_mlp": 1.02196026,
      "epoch": 0.45319545481873386,
      "flos": 15560094286080.0,
      "grad_norm": 2.272860523875624,
      "language_loss": 0.74378622,
      "learning_rate": 2.3977222782198098e-06,
      "loss": 0.76610595,
      "num_input_tokens_seen": 81127750,
      "step": 3769,
      "time_per_iteration": 2.6432316303253174
    },
    {
      "auxiliary_loss_clip": 0.01186694,
      "auxiliary_loss_mlp": 0.01032274,
      "balance_loss_clip": 1.04940987,
      "balance_loss_mlp": 1.0224036,
      "epoch": 0.4533156977093729,
      "flos": 21944759834880.0,
      "grad_norm": 2.0368755130004974,
      "language_loss": 0.75318706,
      "learning_rate": 2.3969588342982077e-06,
      "loss": 0.77537668,
      "num_input_tokens_seen": 81147125,
      "step": 3770,
      "time_per_iteration": 2.6914775371551514
    },
    {
      "auxiliary_loss_clip": 0.01192647,
      "auxiliary_loss_mlp": 0.01030054,
      "balance_loss_clip": 1.05786407,
      "balance_loss_mlp": 1.02129769,
      "epoch": 0.453435940600012,
      "flos": 24242180699520.0,
      "grad_norm": 1.6223308427908312,
      "language_loss": 0.72643441,
      "learning_rate": 2.396195330158267e-06,
      "loss": 0.74866146,
      "num_input_tokens_seen": 81167015,
      "step": 3771,
      "time_per_iteration": 2.606813669204712
    },
    {
      "auxiliary_loss_clip": 0.01189707,
      "auxiliary_loss_mlp": 0.01031239,
      "balance_loss_clip": 1.05620658,
      "balance_loss_mlp": 1.02266192,
      "epoch": 0.45355618349065113,
      "flos": 23440367352960.0,
      "grad_norm": 1.8381965739399062,
      "language_loss": 0.79345959,
      "learning_rate": 2.3954317659158094e-06,
      "loss": 0.81566906,
      "num_input_tokens_seen": 81187350,
      "step": 3772,
      "time_per_iteration": 2.603874444961548
    },
    {
      "auxiliary_loss_clip": 0.01091553,
      "auxiliary_loss_mlp": 0.01003905,
      "balance_loss_clip": 1.0267756,
      "balance_loss_mlp": 1.00299919,
      "epoch": 0.4536764263812902,
      "flos": 66903161448960.0,
      "grad_norm": 0.8954914741674982,
      "language_loss": 0.56897271,
      "learning_rate": 2.394668141686667e-06,
      "loss": 0.58992726,
      "num_input_tokens_seen": 81249315,
      "step": 3773,
      "time_per_iteration": 3.156777858734131
    },
    {
      "auxiliary_loss_clip": 0.01187205,
      "auxiliary_loss_mlp": 0.01029103,
      "balance_loss_clip": 1.05258608,
      "balance_loss_mlp": 1.0208118,
      "epoch": 0.4537966692719293,
      "flos": 42739766254080.0,
      "grad_norm": 4.842583489012086,
      "language_loss": 0.69766521,
      "learning_rate": 2.3939044575866813e-06,
      "loss": 0.71982825,
      "num_input_tokens_seen": 81272065,
      "step": 3774,
      "time_per_iteration": 2.80292010307312
    },
    {
      "auxiliary_loss_clip": 0.01190566,
      "auxiliary_loss_mlp": 0.01208752,
      "balance_loss_clip": 1.05253375,
      "balance_loss_mlp": 1.00074148,
      "epoch": 0.4539169121625684,
      "flos": 35549480517120.0,
      "grad_norm": 2.1878634462154767,
      "language_loss": 0.75711948,
      "learning_rate": 2.3931407137317024e-06,
      "loss": 0.78111267,
      "num_input_tokens_seen": 81292220,
      "step": 3775,
      "time_per_iteration": 3.8909449577331543
    },
    {
      "auxiliary_loss_clip": 0.01193972,
      "auxiliary_loss_mlp": 0.01032163,
      "balance_loss_clip": 1.04959345,
      "balance_loss_mlp": 1.02329338,
      "epoch": 0.45403715505320746,
      "flos": 18514716341760.0,
      "grad_norm": 1.9740774823289047,
      "language_loss": 0.85058314,
      "learning_rate": 2.3923769102375907e-06,
      "loss": 0.87284446,
      "num_input_tokens_seen": 81311085,
      "step": 3776,
      "time_per_iteration": 2.6623828411102295
    },
    {
      "auxiliary_loss_clip": 0.01195276,
      "auxiliary_loss_mlp": 0.01037162,
      "balance_loss_clip": 1.05022395,
      "balance_loss_mlp": 1.02798271,
      "epoch": 0.4541573979438466,
      "flos": 25045825639680.0,
      "grad_norm": 2.169465847266409,
      "language_loss": 0.78470063,
      "learning_rate": 2.391613047220213e-06,
      "loss": 0.80702502,
      "num_input_tokens_seen": 81330985,
      "step": 3777,
      "time_per_iteration": 3.548593282699585
    },
    {
      "auxiliary_loss_clip": 0.01203326,
      "auxiliary_loss_mlp": 0.01025464,
      "balance_loss_clip": 1.04936457,
      "balance_loss_mlp": 1.0167619,
      "epoch": 0.4542776408344857,
      "flos": 18332397884160.0,
      "grad_norm": 2.882197293945006,
      "language_loss": 0.79060328,
      "learning_rate": 2.390849124795447e-06,
      "loss": 0.81289119,
      "num_input_tokens_seen": 81346985,
      "step": 3778,
      "time_per_iteration": 2.650146484375
    },
    {
      "auxiliary_loss_clip": 0.01190979,
      "auxiliary_loss_mlp": 0.01024101,
      "balance_loss_clip": 1.0570296,
      "balance_loss_mlp": 1.01579225,
      "epoch": 0.45439788372512474,
      "flos": 20701173116160.0,
      "grad_norm": 1.8700777010201837,
      "language_loss": 0.84167373,
      "learning_rate": 2.3900851430791804e-06,
      "loss": 0.86382455,
      "num_input_tokens_seen": 81365005,
      "step": 3779,
      "time_per_iteration": 2.550320625305176
    },
    {
      "auxiliary_loss_clip": 0.01192759,
      "auxiliary_loss_mlp": 0.01031483,
      "balance_loss_clip": 1.05637515,
      "balance_loss_mlp": 1.02185047,
      "epoch": 0.45451812661576385,
      "flos": 22309432663680.0,
      "grad_norm": 7.2775490035553,
      "language_loss": 0.84709913,
      "learning_rate": 2.389321102187307e-06,
      "loss": 0.86934155,
      "num_input_tokens_seen": 81383785,
      "step": 3780,
      "time_per_iteration": 3.398507595062256
    },
    {
      "auxiliary_loss_clip": 0.01195464,
      "auxiliary_loss_mlp": 0.01209512,
      "balance_loss_clip": 1.05466723,
      "balance_loss_mlp": 1.00081587,
      "epoch": 0.4546383695064029,
      "flos": 21763303303680.0,
      "grad_norm": 1.9789966320975705,
      "language_loss": 0.81703115,
      "learning_rate": 2.3885570022357326e-06,
      "loss": 0.8410809,
      "num_input_tokens_seen": 81402915,
      "step": 3781,
      "time_per_iteration": 2.5728819370269775
    },
    {
      "auxiliary_loss_clip": 0.0111092,
      "auxiliary_loss_mlp": 0.01008234,
      "balance_loss_clip": 1.02304876,
      "balance_loss_mlp": 1.00738811,
      "epoch": 0.454758612397042,
      "flos": 64242755694720.0,
      "grad_norm": 0.8078374279458082,
      "language_loss": 0.60893762,
      "learning_rate": 2.38779284334037e-06,
      "loss": 0.63012922,
      "num_input_tokens_seen": 81467890,
      "step": 3782,
      "time_per_iteration": 3.2869763374328613
    },
    {
      "auxiliary_loss_clip": 0.01185967,
      "auxiliary_loss_mlp": 0.01028757,
      "balance_loss_clip": 1.04656041,
      "balance_loss_mlp": 1.02009344,
      "epoch": 0.4548788552876811,
      "flos": 27304175485440.0,
      "grad_norm": 2.2358767343060553,
      "language_loss": 0.7883749,
      "learning_rate": 2.387028625617141e-06,
      "loss": 0.8105222,
      "num_input_tokens_seen": 81487105,
      "step": 3783,
      "time_per_iteration": 2.6849775314331055
    },
    {
      "auxiliary_loss_clip": 0.01181064,
      "auxiliary_loss_mlp": 0.01030156,
      "balance_loss_clip": 1.05090749,
      "balance_loss_mlp": 1.02163839,
      "epoch": 0.4549990981783202,
      "flos": 22857142222080.0,
      "grad_norm": 1.85111857292241,
      "language_loss": 0.85014755,
      "learning_rate": 2.3862643491819766e-06,
      "loss": 0.87225974,
      "num_input_tokens_seen": 81505670,
      "step": 3784,
      "time_per_iteration": 2.6135928630828857
    },
    {
      "auxiliary_loss_clip": 0.01189499,
      "auxiliary_loss_mlp": 0.01028794,
      "balance_loss_clip": 1.05253172,
      "balance_loss_mlp": 1.02021122,
      "epoch": 0.4551193410689593,
      "flos": 23258587599360.0,
      "grad_norm": 1.861877031714775,
      "language_loss": 0.8485474,
      "learning_rate": 2.3855000141508186e-06,
      "loss": 0.87073028,
      "num_input_tokens_seen": 81525825,
      "step": 3785,
      "time_per_iteration": 2.6187877655029297
    },
    {
      "auxiliary_loss_clip": 0.01198285,
      "auxiliary_loss_mlp": 0.01031659,
      "balance_loss_clip": 1.05705619,
      "balance_loss_mlp": 1.02200317,
      "epoch": 0.4552395839595984,
      "flos": 20777519473920.0,
      "grad_norm": 2.139735125961937,
      "language_loss": 0.84114158,
      "learning_rate": 2.3847356206396143e-06,
      "loss": 0.86344105,
      "num_input_tokens_seen": 81543135,
      "step": 3786,
      "time_per_iteration": 3.5650038719177246
    },
    {
      "auxiliary_loss_clip": 0.01192027,
      "auxiliary_loss_mlp": 0.01028558,
      "balance_loss_clip": 1.05866623,
      "balance_loss_mlp": 1.01978469,
      "epoch": 0.45535982685023746,
      "flos": 23257510191360.0,
      "grad_norm": 1.8530828819342244,
      "language_loss": 0.78528738,
      "learning_rate": 2.3839711687643227e-06,
      "loss": 0.80749327,
      "num_input_tokens_seen": 81564360,
      "step": 3787,
      "time_per_iteration": 2.611114501953125
    },
    {
      "auxiliary_loss_clip": 0.01191351,
      "auxiliary_loss_mlp": 0.01032364,
      "balance_loss_clip": 1.05504429,
      "balance_loss_mlp": 1.02271998,
      "epoch": 0.45548006974087657,
      "flos": 19646117907840.0,
      "grad_norm": 2.4742257733786643,
      "language_loss": 0.74020731,
      "learning_rate": 2.38320665864091e-06,
      "loss": 0.7624445,
      "num_input_tokens_seen": 81583710,
      "step": 3788,
      "time_per_iteration": 2.5450243949890137
    },
    {
      "auxiliary_loss_clip": 0.01202131,
      "auxiliary_loss_mlp": 0.0103102,
      "balance_loss_clip": 1.04657865,
      "balance_loss_mlp": 1.02194238,
      "epoch": 0.4556003126315157,
      "flos": 20047778766720.0,
      "grad_norm": 3.4194618377605535,
      "language_loss": 0.82272124,
      "learning_rate": 2.3824420903853516e-06,
      "loss": 0.84505272,
      "num_input_tokens_seen": 81602175,
      "step": 3789,
      "time_per_iteration": 2.6610491275787354
    },
    {
      "auxiliary_loss_clip": 0.01191364,
      "auxiliary_loss_mlp": 0.01028156,
      "balance_loss_clip": 1.05582106,
      "balance_loss_mlp": 1.01924562,
      "epoch": 0.45572055552215474,
      "flos": 22959738443520.0,
      "grad_norm": 3.9983097675708468,
      "language_loss": 0.82302368,
      "learning_rate": 2.3816774641136324e-06,
      "loss": 0.8452189,
      "num_input_tokens_seen": 81619430,
      "step": 3790,
      "time_per_iteration": 2.6205122470855713
    },
    {
      "auxiliary_loss_clip": 0.01191454,
      "auxiliary_loss_mlp": 0.01208851,
      "balance_loss_clip": 1.0560143,
      "balance_loss_mlp": 1.00078773,
      "epoch": 0.45584079841279385,
      "flos": 33109925535360.0,
      "grad_norm": 1.649101124287088,
      "language_loss": 0.71454954,
      "learning_rate": 2.380912779941745e-06,
      "loss": 0.73855257,
      "num_input_tokens_seen": 81642550,
      "step": 3791,
      "time_per_iteration": 2.745239734649658
    },
    {
      "auxiliary_loss_clip": 0.01193876,
      "auxiliary_loss_mlp": 0.01035501,
      "balance_loss_clip": 1.05273652,
      "balance_loss_mlp": 1.02529097,
      "epoch": 0.45596104130343296,
      "flos": 27272179445760.0,
      "grad_norm": 2.55955255625896,
      "language_loss": 0.8316192,
      "learning_rate": 2.3801480379856918e-06,
      "loss": 0.85391295,
      "num_input_tokens_seen": 81664260,
      "step": 3792,
      "time_per_iteration": 2.644299268722534
    },
    {
      "auxiliary_loss_clip": 0.01198176,
      "auxiliary_loss_mlp": 0.0103431,
      "balance_loss_clip": 1.05644345,
      "balance_loss_mlp": 1.02595294,
      "epoch": 0.456081284194072,
      "flos": 21579799697280.0,
      "grad_norm": 1.71156833716836,
      "language_loss": 0.84008491,
      "learning_rate": 2.379383238361484e-06,
      "loss": 0.86240971,
      "num_input_tokens_seen": 81683620,
      "step": 3793,
      "time_per_iteration": 2.616506338119507
    },
    {
      "auxiliary_loss_clip": 0.0119091,
      "auxiliary_loss_mlp": 0.01031304,
      "balance_loss_clip": 1.05394697,
      "balance_loss_mlp": 1.02254784,
      "epoch": 0.4562015270847111,
      "flos": 35918822113920.0,
      "grad_norm": 1.9686032747873392,
      "language_loss": 0.80110252,
      "learning_rate": 2.3786183811851407e-06,
      "loss": 0.82332462,
      "num_input_tokens_seen": 81704325,
      "step": 3794,
      "time_per_iteration": 2.703350305557251
    },
    {
      "auxiliary_loss_clip": 0.01193537,
      "auxiliary_loss_mlp": 0.01027122,
      "balance_loss_clip": 1.06050432,
      "balance_loss_mlp": 1.01859248,
      "epoch": 0.45632176997535023,
      "flos": 13589783602560.0,
      "grad_norm": 1.737279600978444,
      "language_loss": 0.80170822,
      "learning_rate": 2.3778534665726892e-06,
      "loss": 0.82391483,
      "num_input_tokens_seen": 81721155,
      "step": 3795,
      "time_per_iteration": 2.627302885055542
    },
    {
      "auxiliary_loss_clip": 0.01183676,
      "auxiliary_loss_mlp": 0.01031553,
      "balance_loss_clip": 1.0558846,
      "balance_loss_mlp": 1.02323174,
      "epoch": 0.4564420128659893,
      "flos": 32635401937920.0,
      "grad_norm": 5.301130282683402,
      "language_loss": 0.72791505,
      "learning_rate": 2.377088494640168e-06,
      "loss": 0.75006729,
      "num_input_tokens_seen": 81742905,
      "step": 3796,
      "time_per_iteration": 2.6789331436157227
    },
    {
      "auxiliary_loss_clip": 0.01188215,
      "auxiliary_loss_mlp": 0.01030184,
      "balance_loss_clip": 1.05685616,
      "balance_loss_mlp": 1.02118397,
      "epoch": 0.4565622557566284,
      "flos": 20377690208640.0,
      "grad_norm": 1.8072994179458266,
      "language_loss": 0.78273737,
      "learning_rate": 2.3763234655036216e-06,
      "loss": 0.80492133,
      "num_input_tokens_seen": 81762105,
      "step": 3797,
      "time_per_iteration": 2.6286563873291016
    },
    {
      "auxiliary_loss_clip": 0.01194183,
      "auxiliary_loss_mlp": 0.01029393,
      "balance_loss_clip": 1.04836082,
      "balance_loss_mlp": 1.02067852,
      "epoch": 0.45668249864726745,
      "flos": 25374372364800.0,
      "grad_norm": 2.1162131487975815,
      "language_loss": 0.86937433,
      "learning_rate": 2.3755583792791046e-06,
      "loss": 0.89161009,
      "num_input_tokens_seen": 81781975,
      "step": 3798,
      "time_per_iteration": 2.746436357498169
    },
    {
      "auxiliary_loss_clip": 0.01194555,
      "auxiliary_loss_mlp": 0.01025542,
      "balance_loss_clip": 1.05625749,
      "balance_loss_mlp": 1.01715553,
      "epoch": 0.45680274153790656,
      "flos": 15559806977280.0,
      "grad_norm": 1.977031804105209,
      "language_loss": 0.74644876,
      "learning_rate": 2.3747932360826803e-06,
      "loss": 0.7686497,
      "num_input_tokens_seen": 81798905,
      "step": 3799,
      "time_per_iteration": 2.6035594940185547
    },
    {
      "auxiliary_loss_clip": 0.0118973,
      "auxiliary_loss_mlp": 0.01029351,
      "balance_loss_clip": 1.05518842,
      "balance_loss_mlp": 1.02014852,
      "epoch": 0.4569229844285457,
      "flos": 19792884879360.0,
      "grad_norm": 1.8900585478244818,
      "language_loss": 0.82215977,
      "learning_rate": 2.3740280360304205e-06,
      "loss": 0.84435058,
      "num_input_tokens_seen": 81816630,
      "step": 3800,
      "time_per_iteration": 2.6760060787200928
    },
    {
      "auxiliary_loss_clip": 0.011975,
      "auxiliary_loss_mlp": 0.01029246,
      "balance_loss_clip": 1.05558038,
      "balance_loss_mlp": 1.01997733,
      "epoch": 0.45704322731918473,
      "flos": 24093941270400.0,
      "grad_norm": 1.9805051471000765,
      "language_loss": 0.6811043,
      "learning_rate": 2.3732627792384038e-06,
      "loss": 0.70337176,
      "num_input_tokens_seen": 81837700,
      "step": 3801,
      "time_per_iteration": 2.6867127418518066
    },
    {
      "auxiliary_loss_clip": 0.01192021,
      "auxiliary_loss_mlp": 0.01031982,
      "balance_loss_clip": 1.05751526,
      "balance_loss_mlp": 1.02305305,
      "epoch": 0.45716347020982384,
      "flos": 31317803245440.0,
      "grad_norm": 2.16075358293453,
      "language_loss": 0.75455332,
      "learning_rate": 2.3724974658227207e-06,
      "loss": 0.77679324,
      "num_input_tokens_seen": 81858490,
      "step": 3802,
      "time_per_iteration": 3.5521225929260254
    },
    {
      "auxiliary_loss_clip": 0.011951,
      "auxiliary_loss_mlp": 0.01209272,
      "balance_loss_clip": 1.05629277,
      "balance_loss_mlp": 1.00078022,
      "epoch": 0.45728371310046295,
      "flos": 26501392471680.0,
      "grad_norm": 2.6834008642779836,
      "language_loss": 0.71278596,
      "learning_rate": 2.3717320958994687e-06,
      "loss": 0.73682964,
      "num_input_tokens_seen": 81876050,
      "step": 3803,
      "time_per_iteration": 2.653150796890259
    },
    {
      "auxiliary_loss_clip": 0.01195487,
      "auxiliary_loss_mlp": 0.01027442,
      "balance_loss_clip": 1.04748416,
      "balance_loss_mlp": 1.01896071,
      "epoch": 0.457403955991102,
      "flos": 17929408222080.0,
      "grad_norm": 1.8897052702949377,
      "language_loss": 0.70327079,
      "learning_rate": 2.3709666695847534e-06,
      "loss": 0.72550011,
      "num_input_tokens_seen": 81894230,
      "step": 3804,
      "time_per_iteration": 3.5526206493377686
    },
    {
      "auxiliary_loss_clip": 0.01192509,
      "auxiliary_loss_mlp": 0.01029411,
      "balance_loss_clip": 1.04876733,
      "balance_loss_mlp": 1.02101898,
      "epoch": 0.4575241988817411,
      "flos": 42230660837760.0,
      "grad_norm": 1.716414127363333,
      "language_loss": 0.70576715,
      "learning_rate": 2.370201186994689e-06,
      "loss": 0.7279864,
      "num_input_tokens_seen": 81917915,
      "step": 3805,
      "time_per_iteration": 2.8983771800994873
    },
    {
      "auxiliary_loss_clip": 0.01186642,
      "auxiliary_loss_mlp": 0.01028811,
      "balance_loss_clip": 1.05495024,
      "balance_loss_mlp": 1.02022839,
      "epoch": 0.45764444177238023,
      "flos": 30117309868800.0,
      "grad_norm": 2.1389993462254115,
      "language_loss": 0.69751704,
      "learning_rate": 2.369435648245399e-06,
      "loss": 0.71967149,
      "num_input_tokens_seen": 81938130,
      "step": 3806,
      "time_per_iteration": 3.6168620586395264
    },
    {
      "auxiliary_loss_clip": 0.01193516,
      "auxiliary_loss_mlp": 0.0103561,
      "balance_loss_clip": 1.05315995,
      "balance_loss_mlp": 1.02623391,
      "epoch": 0.4577646846630193,
      "flos": 24060293205120.0,
      "grad_norm": 1.7710974498288907,
      "language_loss": 0.85180551,
      "learning_rate": 2.368670053453015e-06,
      "loss": 0.87409675,
      "num_input_tokens_seen": 81959820,
      "step": 3807,
      "time_per_iteration": 2.662651300430298
    },
    {
      "auxiliary_loss_clip": 0.0120234,
      "auxiliary_loss_mlp": 0.01030516,
      "balance_loss_clip": 1.06110215,
      "balance_loss_mlp": 1.02088428,
      "epoch": 0.4578849275536584,
      "flos": 17418578952960.0,
      "grad_norm": 2.6073436751280767,
      "language_loss": 0.74461555,
      "learning_rate": 2.3679044027336757e-06,
      "loss": 0.76694405,
      "num_input_tokens_seen": 81975710,
      "step": 3808,
      "time_per_iteration": 2.5562751293182373
    },
    {
      "auxiliary_loss_clip": 0.01192607,
      "auxiliary_loss_mlp": 0.01026758,
      "balance_loss_clip": 1.05819416,
      "balance_loss_mlp": 1.01738214,
      "epoch": 0.4580051704442975,
      "flos": 13510169107200.0,
      "grad_norm": 3.022644051279784,
      "language_loss": 0.69405174,
      "learning_rate": 2.3671386962035326e-06,
      "loss": 0.71624541,
      "num_input_tokens_seen": 81993180,
      "step": 3809,
      "time_per_iteration": 2.627178430557251
    },
    {
      "auxiliary_loss_clip": 0.01193931,
      "auxiliary_loss_mlp": 0.01031354,
      "balance_loss_clip": 1.05653584,
      "balance_loss_mlp": 1.02246737,
      "epoch": 0.45812541333493656,
      "flos": 18037606965120.0,
      "grad_norm": 2.07664710283462,
      "language_loss": 0.69149834,
      "learning_rate": 2.3663729339787405e-06,
      "loss": 0.71375114,
      "num_input_tokens_seen": 82010115,
      "step": 3810,
      "time_per_iteration": 2.5467123985290527
    },
    {
      "auxiliary_loss_clip": 0.01192041,
      "auxiliary_loss_mlp": 0.01026402,
      "balance_loss_clip": 1.05751383,
      "balance_loss_mlp": 1.01733005,
      "epoch": 0.45824565622557567,
      "flos": 20222196232320.0,
      "grad_norm": 2.345745181959124,
      "language_loss": 0.73641455,
      "learning_rate": 2.365607116175466e-06,
      "loss": 0.75859904,
      "num_input_tokens_seen": 82025540,
      "step": 3811,
      "time_per_iteration": 2.5310616493225098
    },
    {
      "auxiliary_loss_clip": 0.01190832,
      "auxiliary_loss_mlp": 0.01026661,
      "balance_loss_clip": 1.05847764,
      "balance_loss_mlp": 1.017941,
      "epoch": 0.4583658991162148,
      "flos": 19864885691520.0,
      "grad_norm": 2.3707466889314928,
      "language_loss": 0.67296708,
      "learning_rate": 2.3648412429098825e-06,
      "loss": 0.69514203,
      "num_input_tokens_seen": 82043890,
      "step": 3812,
      "time_per_iteration": 2.5288779735565186
    },
    {
      "auxiliary_loss_clip": 0.01190214,
      "auxiliary_loss_mlp": 0.01032717,
      "balance_loss_clip": 1.05144966,
      "balance_loss_mlp": 1.02337074,
      "epoch": 0.45848614200685384,
      "flos": 21029935322880.0,
      "grad_norm": 1.8492528682656555,
      "language_loss": 0.81764698,
      "learning_rate": 2.364075314298172e-06,
      "loss": 0.83987629,
      "num_input_tokens_seen": 82061345,
      "step": 3813,
      "time_per_iteration": 3.540686845779419
    },
    {
      "auxiliary_loss_clip": 0.01200357,
      "auxiliary_loss_mlp": 0.0120926,
      "balance_loss_clip": 1.05981934,
      "balance_loss_mlp": 1.00067472,
      "epoch": 0.45860638489749295,
      "flos": 21069293650560.0,
      "grad_norm": 1.776825733142218,
      "language_loss": 0.70513636,
      "learning_rate": 2.3633093304565267e-06,
      "loss": 0.72923249,
      "num_input_tokens_seen": 82080400,
      "step": 3814,
      "time_per_iteration": 2.567105531692505
    },
    {
      "auxiliary_loss_clip": 0.01196818,
      "auxiliary_loss_mlp": 0.01032306,
      "balance_loss_clip": 1.06065261,
      "balance_loss_mlp": 1.02297151,
      "epoch": 0.458726627788132,
      "flos": 26833889692800.0,
      "grad_norm": 1.8231461677358045,
      "language_loss": 0.62963498,
      "learning_rate": 2.3625432915011443e-06,
      "loss": 0.65192616,
      "num_input_tokens_seen": 82102310,
      "step": 3815,
      "time_per_iteration": 2.5380687713623047
    },
    {
      "auxiliary_loss_clip": 0.01191736,
      "auxiliary_loss_mlp": 0.01032483,
      "balance_loss_clip": 1.05521142,
      "balance_loss_mlp": 1.02345932,
      "epoch": 0.4588468706787711,
      "flos": 24097927680000.0,
      "grad_norm": 1.772399101044945,
      "language_loss": 0.6564346,
      "learning_rate": 2.3617771975482334e-06,
      "loss": 0.67867678,
      "num_input_tokens_seen": 82121140,
      "step": 3816,
      "time_per_iteration": 2.5988826751708984
    },
    {
      "auxiliary_loss_clip": 0.01194806,
      "auxiliary_loss_mlp": 0.01029902,
      "balance_loss_clip": 1.05116713,
      "balance_loss_mlp": 1.02183151,
      "epoch": 0.4589671135694102,
      "flos": 17889331622400.0,
      "grad_norm": 1.7315474245513296,
      "language_loss": 0.74550653,
      "learning_rate": 2.3610110487140083e-06,
      "loss": 0.7677536,
      "num_input_tokens_seen": 82139575,
      "step": 3817,
      "time_per_iteration": 2.6201326847076416
    },
    {
      "auxiliary_loss_clip": 0.0119494,
      "auxiliary_loss_mlp": 0.01032028,
      "balance_loss_clip": 1.05571365,
      "balance_loss_mlp": 1.02296782,
      "epoch": 0.4590873564600493,
      "flos": 25626967781760.0,
      "grad_norm": 1.78888639575588,
      "language_loss": 0.80937833,
      "learning_rate": 2.360244845114695e-06,
      "loss": 0.83164799,
      "num_input_tokens_seen": 82159195,
      "step": 3818,
      "time_per_iteration": 2.690424919128418
    },
    {
      "auxiliary_loss_clip": 0.01191014,
      "auxiliary_loss_mlp": 0.01027827,
      "balance_loss_clip": 1.05704117,
      "balance_loss_mlp": 1.01874876,
      "epoch": 0.4592075993506884,
      "flos": 18514788168960.0,
      "grad_norm": 2.3863130836531425,
      "language_loss": 0.68887848,
      "learning_rate": 2.3594785868665245e-06,
      "loss": 0.71106684,
      "num_input_tokens_seen": 82175500,
      "step": 3819,
      "time_per_iteration": 2.5820071697235107
    },
    {
      "auxiliary_loss_clip": 0.01197723,
      "auxiliary_loss_mlp": 0.01209375,
      "balance_loss_clip": 1.05170727,
      "balance_loss_mlp": 1.00070024,
      "epoch": 0.4593278422413275,
      "flos": 20631111638400.0,
      "grad_norm": 2.2618469198504934,
      "language_loss": 0.80676341,
      "learning_rate": 2.3587122740857386e-06,
      "loss": 0.83083433,
      "num_input_tokens_seen": 82192600,
      "step": 3820,
      "time_per_iteration": 2.683814287185669
    },
    {
      "auxiliary_loss_clip": 0.0119224,
      "auxiliary_loss_mlp": 0.01029408,
      "balance_loss_clip": 1.05626476,
      "balance_loss_mlp": 1.02104557,
      "epoch": 0.45944808513196655,
      "flos": 21358517961600.0,
      "grad_norm": 1.5484066173102573,
      "language_loss": 0.77783263,
      "learning_rate": 2.357945906888586e-06,
      "loss": 0.80004907,
      "num_input_tokens_seen": 82212040,
      "step": 3821,
      "time_per_iteration": 2.7311341762542725
    },
    {
      "auxiliary_loss_clip": 0.01197207,
      "auxiliary_loss_mlp": 0.0103442,
      "balance_loss_clip": 1.05759311,
      "balance_loss_mlp": 1.02482998,
      "epoch": 0.45956832802260567,
      "flos": 21427789340160.0,
      "grad_norm": 2.2113116744527117,
      "language_loss": 0.79906952,
      "learning_rate": 2.357179485391324e-06,
      "loss": 0.8213858,
      "num_input_tokens_seen": 82229895,
      "step": 3822,
      "time_per_iteration": 2.690778970718384
    },
    {
      "auxiliary_loss_clip": 0.01189427,
      "auxiliary_loss_mlp": 0.01026667,
      "balance_loss_clip": 1.0580833,
      "balance_loss_mlp": 1.01781607,
      "epoch": 0.4596885709132448,
      "flos": 22382654538240.0,
      "grad_norm": 1.938450896965196,
      "language_loss": 0.86651474,
      "learning_rate": 2.3564130097102173e-06,
      "loss": 0.88867557,
      "num_input_tokens_seen": 82249550,
      "step": 3823,
      "time_per_iteration": 2.6090903282165527
    },
    {
      "auxiliary_loss_clip": 0.01188834,
      "auxiliary_loss_mlp": 0.01029266,
      "balance_loss_clip": 1.05712676,
      "balance_loss_mlp": 1.02005744,
      "epoch": 0.45980881380388383,
      "flos": 28981957806720.0,
      "grad_norm": 1.717281091169215,
      "language_loss": 0.75036395,
      "learning_rate": 2.355646479961541e-06,
      "loss": 0.77254498,
      "num_input_tokens_seen": 82268860,
      "step": 3824,
      "time_per_iteration": 2.7362284660339355
    },
    {
      "auxiliary_loss_clip": 0.01192467,
      "auxiliary_loss_mlp": 0.01026714,
      "balance_loss_clip": 1.05928361,
      "balance_loss_mlp": 1.01712406,
      "epoch": 0.45992905669452294,
      "flos": 33396599980800.0,
      "grad_norm": 1.845492131675665,
      "language_loss": 0.71468717,
      "learning_rate": 2.354879896261576e-06,
      "loss": 0.73687899,
      "num_input_tokens_seen": 82289070,
      "step": 3825,
      "time_per_iteration": 2.7184224128723145
    },
    {
      "auxiliary_loss_clip": 0.01195469,
      "auxiliary_loss_mlp": 0.01032976,
      "balance_loss_clip": 1.05494046,
      "balance_loss_mlp": 1.02432692,
      "epoch": 0.46004929958516205,
      "flos": 36318184502400.0,
      "grad_norm": 4.665208400909502,
      "language_loss": 0.56878328,
      "learning_rate": 2.3541132587266133e-06,
      "loss": 0.59106779,
      "num_input_tokens_seen": 82311790,
      "step": 3826,
      "time_per_iteration": 2.7864692211151123
    },
    {
      "auxiliary_loss_clip": 0.01205407,
      "auxiliary_loss_mlp": 0.01028301,
      "balance_loss_clip": 1.05469656,
      "balance_loss_mlp": 1.01941991,
      "epoch": 0.4601695424758011,
      "flos": 17238451224960.0,
      "grad_norm": 1.924574049104019,
      "language_loss": 0.69300616,
      "learning_rate": 2.3533465674729515e-06,
      "loss": 0.71534318,
      "num_input_tokens_seen": 82329020,
      "step": 3827,
      "time_per_iteration": 2.676816463470459
    },
    {
      "auxiliary_loss_clip": 0.01193724,
      "auxiliary_loss_mlp": 0.01035371,
      "balance_loss_clip": 1.06000483,
      "balance_loss_mlp": 1.02569699,
      "epoch": 0.4602897853664402,
      "flos": 15888425529600.0,
      "grad_norm": 4.805653202044809,
      "language_loss": 0.73062301,
      "learning_rate": 2.352579822616895e-06,
      "loss": 0.75291395,
      "num_input_tokens_seen": 82346455,
      "step": 3828,
      "time_per_iteration": 2.5936200618743896
    },
    {
      "auxiliary_loss_clip": 0.01199481,
      "auxiliary_loss_mlp": 0.01027643,
      "balance_loss_clip": 1.05551064,
      "balance_loss_mlp": 1.01886916,
      "epoch": 0.4604100282570793,
      "flos": 25412617370880.0,
      "grad_norm": 1.6304927059872587,
      "language_loss": 0.77809799,
      "learning_rate": 2.351813024274761e-06,
      "loss": 0.80036926,
      "num_input_tokens_seen": 82367810,
      "step": 3829,
      "time_per_iteration": 3.586663246154785
    },
    {
      "auxiliary_loss_clip": 0.01203312,
      "auxiliary_loss_mlp": 0.01033334,
      "balance_loss_clip": 1.05529046,
      "balance_loss_mlp": 1.02443528,
      "epoch": 0.4605302711477184,
      "flos": 27630711048960.0,
      "grad_norm": 1.7803726233467199,
      "language_loss": 0.73661256,
      "learning_rate": 2.3510461725628693e-06,
      "loss": 0.75897902,
      "num_input_tokens_seen": 82388275,
      "step": 3830,
      "time_per_iteration": 3.638617515563965
    },
    {
      "auxiliary_loss_clip": 0.01202128,
      "auxiliary_loss_mlp": 0.01030048,
      "balance_loss_clip": 1.05439878,
      "balance_loss_mlp": 1.02106595,
      "epoch": 0.4606505140383575,
      "flos": 23839657914240.0,
      "grad_norm": 2.1789727542471837,
      "language_loss": 0.71107006,
      "learning_rate": 2.350279267597554e-06,
      "loss": 0.73339188,
      "num_input_tokens_seen": 82408915,
      "step": 3831,
      "time_per_iteration": 2.714078903198242
    },
    {
      "auxiliary_loss_clip": 0.01195955,
      "auxiliary_loss_mlp": 0.01033169,
      "balance_loss_clip": 1.05948687,
      "balance_loss_mlp": 1.02401924,
      "epoch": 0.46077075692899655,
      "flos": 16107013745280.0,
      "grad_norm": 2.178178829225682,
      "language_loss": 0.82798946,
      "learning_rate": 2.3495123094951515e-06,
      "loss": 0.8502807,
      "num_input_tokens_seen": 82427260,
      "step": 3832,
      "time_per_iteration": 2.5626771450042725
    },
    {
      "auxiliary_loss_clip": 0.01189894,
      "auxiliary_loss_mlp": 0.01024501,
      "balance_loss_clip": 1.05573297,
      "balance_loss_mlp": 1.01527429,
      "epoch": 0.46089099981963566,
      "flos": 48798147634560.0,
      "grad_norm": 4.3587856739714805,
      "language_loss": 0.7644341,
      "learning_rate": 2.34874529837201e-06,
      "loss": 0.78657806,
      "num_input_tokens_seen": 82450805,
      "step": 3833,
      "time_per_iteration": 3.778935432434082
    },
    {
      "auxiliary_loss_clip": 0.01198932,
      "auxiliary_loss_mlp": 0.01026577,
      "balance_loss_clip": 1.04834831,
      "balance_loss_mlp": 1.01764178,
      "epoch": 0.46101124271027477,
      "flos": 19099234362240.0,
      "grad_norm": 1.787693345427471,
      "language_loss": 0.79102385,
      "learning_rate": 2.347978234344483e-06,
      "loss": 0.81327891,
      "num_input_tokens_seen": 82467010,
      "step": 3834,
      "time_per_iteration": 2.664963960647583
    },
    {
      "auxiliary_loss_clip": 0.01200148,
      "auxiliary_loss_mlp": 0.01036028,
      "balance_loss_clip": 1.06030726,
      "balance_loss_mlp": 1.02594256,
      "epoch": 0.4611314856009138,
      "flos": 39347931853440.0,
      "grad_norm": 1.7332226946246785,
      "language_loss": 0.69147933,
      "learning_rate": 2.347211117528935e-06,
      "loss": 0.71384108,
      "num_input_tokens_seen": 82489310,
      "step": 3835,
      "time_per_iteration": 2.7338345050811768
    },
    {
      "auxiliary_loss_clip": 0.01209707,
      "auxiliary_loss_mlp": 0.01035948,
      "balance_loss_clip": 1.06011045,
      "balance_loss_mlp": 1.02659619,
      "epoch": 0.46125172849155294,
      "flos": 20810772489600.0,
      "grad_norm": 1.9197400095194952,
      "language_loss": 0.71921515,
      "learning_rate": 2.3464439480417374e-06,
      "loss": 0.74167174,
      "num_input_tokens_seen": 82508830,
      "step": 3836,
      "time_per_iteration": 2.7020132541656494
    },
    {
      "auxiliary_loss_clip": 0.01196651,
      "auxiliary_loss_mlp": 0.01036145,
      "balance_loss_clip": 1.05814981,
      "balance_loss_mlp": 1.02689457,
      "epoch": 0.46137197138219205,
      "flos": 17930808852480.0,
      "grad_norm": 7.0890653112487465,
      "language_loss": 0.77764732,
      "learning_rate": 2.3456767259992676e-06,
      "loss": 0.79997528,
      "num_input_tokens_seen": 82526475,
      "step": 3837,
      "time_per_iteration": 2.5766072273254395
    },
    {
      "auxiliary_loss_clip": 0.01193332,
      "auxiliary_loss_mlp": 0.01209432,
      "balance_loss_clip": 1.05838203,
      "balance_loss_mlp": 1.00054002,
      "epoch": 0.4614922142728311,
      "flos": 16836610798080.0,
      "grad_norm": 2.246165940076469,
      "language_loss": 0.88915217,
      "learning_rate": 2.3449094515179135e-06,
      "loss": 0.91317981,
      "num_input_tokens_seen": 82543935,
      "step": 3838,
      "time_per_iteration": 2.5697782039642334
    },
    {
      "auxiliary_loss_clip": 0.01201517,
      "auxiliary_loss_mlp": 0.01028768,
      "balance_loss_clip": 1.05547452,
      "balance_loss_mlp": 1.01948118,
      "epoch": 0.4616124571634702,
      "flos": 26614906427520.0,
      "grad_norm": 1.7168975237037436,
      "language_loss": 0.81932819,
      "learning_rate": 2.34414212471407e-06,
      "loss": 0.84163105,
      "num_input_tokens_seen": 82563730,
      "step": 3839,
      "time_per_iteration": 3.6817498207092285
    },
    {
      "auxiliary_loss_clip": 0.01201702,
      "auxiliary_loss_mlp": 0.01028092,
      "balance_loss_clip": 1.05818915,
      "balance_loss_mlp": 1.01887679,
      "epoch": 0.4617327000541093,
      "flos": 20340127560960.0,
      "grad_norm": 2.0635693655379175,
      "language_loss": 0.72974885,
      "learning_rate": 2.3433747457041394e-06,
      "loss": 0.7520467,
      "num_input_tokens_seen": 82582435,
      "step": 3840,
      "time_per_iteration": 2.5916929244995117
    },
    {
      "auxiliary_loss_clip": 0.01199869,
      "auxiliary_loss_mlp": 0.01030439,
      "balance_loss_clip": 1.0554018,
      "balance_loss_mlp": 1.02086091,
      "epoch": 0.4618529429447484,
      "flos": 29570749545600.0,
      "grad_norm": 1.7686087939103048,
      "language_loss": 0.85063577,
      "learning_rate": 2.342607314604533e-06,
      "loss": 0.87293881,
      "num_input_tokens_seen": 82602185,
      "step": 3841,
      "time_per_iteration": 2.693587064743042
    },
    {
      "auxiliary_loss_clip": 0.01193744,
      "auxiliary_loss_mlp": 0.01029066,
      "balance_loss_clip": 1.05864978,
      "balance_loss_mlp": 1.02020288,
      "epoch": 0.4619731858353875,
      "flos": 19787030962560.0,
      "grad_norm": 2.070534583480895,
      "language_loss": 0.84273052,
      "learning_rate": 2.3418398315316694e-06,
      "loss": 0.86495858,
      "num_input_tokens_seen": 82620005,
      "step": 3842,
      "time_per_iteration": 2.5899739265441895
    },
    {
      "auxiliary_loss_clip": 0.01194561,
      "auxiliary_loss_mlp": 0.01037036,
      "balance_loss_clip": 1.06230831,
      "balance_loss_mlp": 1.02780962,
      "epoch": 0.4620934287260266,
      "flos": 18951138587520.0,
      "grad_norm": 2.8347100008643595,
      "language_loss": 0.78888381,
      "learning_rate": 2.3410722966019755e-06,
      "loss": 0.81119978,
      "num_input_tokens_seen": 82635120,
      "step": 3843,
      "time_per_iteration": 2.593966007232666
    },
    {
      "auxiliary_loss_clip": 0.01192318,
      "auxiliary_loss_mlp": 0.0102927,
      "balance_loss_clip": 1.05631053,
      "balance_loss_mlp": 1.02003098,
      "epoch": 0.46221367161666566,
      "flos": 37341674634240.0,
      "grad_norm": 5.579226705021719,
      "language_loss": 0.65769464,
      "learning_rate": 2.3403047099318848e-06,
      "loss": 0.67991054,
      "num_input_tokens_seen": 82659190,
      "step": 3844,
      "time_per_iteration": 2.697540521621704
    },
    {
      "auxiliary_loss_clip": 0.01196105,
      "auxiliary_loss_mlp": 0.0102861,
      "balance_loss_clip": 1.05038285,
      "balance_loss_mlp": 1.01996148,
      "epoch": 0.46233391450730477,
      "flos": 14428549065600.0,
      "grad_norm": 4.1853789552083445,
      "language_loss": 0.75332367,
      "learning_rate": 2.3395370716378405e-06,
      "loss": 0.77557087,
      "num_input_tokens_seen": 82676635,
      "step": 3845,
      "time_per_iteration": 2.6788265705108643
    },
    {
      "auxiliary_loss_clip": 0.01198735,
      "auxiliary_loss_mlp": 0.01034538,
      "balance_loss_clip": 1.05804515,
      "balance_loss_mlp": 1.02574587,
      "epoch": 0.4624541573979438,
      "flos": 22493044010880.0,
      "grad_norm": 2.3443481332160725,
      "language_loss": 0.72780812,
      "learning_rate": 2.338769381836292e-06,
      "loss": 0.75014079,
      "num_input_tokens_seen": 82696245,
      "step": 3846,
      "time_per_iteration": 2.5986859798431396
    },
    {
      "auxiliary_loss_clip": 0.0119213,
      "auxiliary_loss_mlp": 0.01031983,
      "balance_loss_clip": 1.05423021,
      "balance_loss_mlp": 1.02316105,
      "epoch": 0.46257440028858293,
      "flos": 14465070218880.0,
      "grad_norm": 2.3515813569258475,
      "language_loss": 0.73036063,
      "learning_rate": 2.3380016406436984e-06,
      "loss": 0.75260168,
      "num_input_tokens_seen": 82713725,
      "step": 3847,
      "time_per_iteration": 2.6213595867156982
    },
    {
      "auxiliary_loss_clip": 0.0119736,
      "auxiliary_loss_mlp": 0.01035193,
      "balance_loss_clip": 1.05337465,
      "balance_loss_mlp": 1.02542329,
      "epoch": 0.46269464317922204,
      "flos": 23332204523520.0,
      "grad_norm": 2.532347026614873,
      "language_loss": 0.81536007,
      "learning_rate": 2.337233848176524e-06,
      "loss": 0.83768559,
      "num_input_tokens_seen": 82731495,
      "step": 3848,
      "time_per_iteration": 2.679832696914673
    },
    {
      "auxiliary_loss_clip": 0.01193604,
      "auxiliary_loss_mlp": 0.01030541,
      "balance_loss_clip": 1.05051255,
      "balance_loss_mlp": 1.02163553,
      "epoch": 0.4628148860698611,
      "flos": 18552027594240.0,
      "grad_norm": 3.0740349007764185,
      "language_loss": 0.83425963,
      "learning_rate": 2.3364660045512435e-06,
      "loss": 0.8565011,
      "num_input_tokens_seen": 82750255,
      "step": 3849,
      "time_per_iteration": 2.6066296100616455
    },
    {
      "auxiliary_loss_clip": 0.01101691,
      "auxiliary_loss_mlp": 0.01007699,
      "balance_loss_clip": 1.02558804,
      "balance_loss_mlp": 1.00659645,
      "epoch": 0.4629351289605002,
      "flos": 70667569670400.0,
      "grad_norm": 0.7529722267944294,
      "language_loss": 0.58184361,
      "learning_rate": 2.335698109884337e-06,
      "loss": 0.60293746,
      "num_input_tokens_seen": 82815460,
      "step": 3850,
      "time_per_iteration": 3.26486873626709
    },
    {
      "auxiliary_loss_clip": 0.01114271,
      "auxiliary_loss_mlp": 0.0100249,
      "balance_loss_clip": 1.02886403,
      "balance_loss_mlp": 1.00130355,
      "epoch": 0.4630553718511393,
      "flos": 59687200465920.0,
      "grad_norm": 0.7864125065714336,
      "language_loss": 0.59847379,
      "learning_rate": 2.334930164292294e-06,
      "loss": 0.61964136,
      "num_input_tokens_seen": 82878010,
      "step": 3851,
      "time_per_iteration": 3.3290374279022217
    },
    {
      "auxiliary_loss_clip": 0.01195486,
      "auxiliary_loss_mlp": 0.01029306,
      "balance_loss_clip": 1.05149007,
      "balance_loss_mlp": 1.02098513,
      "epoch": 0.4631756147417784,
      "flos": 15960605909760.0,
      "grad_norm": 3.2560324628455017,
      "language_loss": 0.80144143,
      "learning_rate": 2.334162167891612e-06,
      "loss": 0.82368934,
      "num_input_tokens_seen": 82895275,
      "step": 3852,
      "time_per_iteration": 2.7570676803588867
    },
    {
      "auxiliary_loss_clip": 0.01198479,
      "auxiliary_loss_mlp": 0.01035496,
      "balance_loss_clip": 1.0546962,
      "balance_loss_mlp": 1.02598882,
      "epoch": 0.4632958576324175,
      "flos": 16472907636480.0,
      "grad_norm": 2.288923939902309,
      "language_loss": 0.74818045,
      "learning_rate": 2.333394120798795e-06,
      "loss": 0.77052021,
      "num_input_tokens_seen": 82914010,
      "step": 3853,
      "time_per_iteration": 2.590324640274048
    },
    {
      "auxiliary_loss_clip": 0.01194895,
      "auxiliary_loss_mlp": 0.0102626,
      "balance_loss_clip": 1.05273485,
      "balance_loss_mlp": 1.0165627,
      "epoch": 0.4634161005230566,
      "flos": 22346492520960.0,
      "grad_norm": 2.437717724992394,
      "language_loss": 0.71892619,
      "learning_rate": 2.3326260231303545e-06,
      "loss": 0.74113774,
      "num_input_tokens_seen": 82932610,
      "step": 3854,
      "time_per_iteration": 2.703979253768921
    },
    {
      "auxiliary_loss_clip": 0.01191038,
      "auxiliary_loss_mlp": 0.01024379,
      "balance_loss_clip": 1.061005,
      "balance_loss_mlp": 1.01558721,
      "epoch": 0.46353634341369565,
      "flos": 15742233175680.0,
      "grad_norm": 1.6633129318221072,
      "language_loss": 0.86879599,
      "learning_rate": 2.331857875002811e-06,
      "loss": 0.89095014,
      "num_input_tokens_seen": 82951210,
      "step": 3855,
      "time_per_iteration": 3.52626371383667
    },
    {
      "auxiliary_loss_clip": 0.01198541,
      "auxiliary_loss_mlp": 0.0103544,
      "balance_loss_clip": 1.05872345,
      "balance_loss_mlp": 1.02688026,
      "epoch": 0.46365658630433476,
      "flos": 28329820433280.0,
      "grad_norm": 1.9406180063023153,
      "language_loss": 0.76475126,
      "learning_rate": 2.3310896765326916e-06,
      "loss": 0.78709102,
      "num_input_tokens_seen": 82972210,
      "step": 3856,
      "time_per_iteration": 3.6509315967559814
    },
    {
      "auxiliary_loss_clip": 0.01195162,
      "auxiliary_loss_mlp": 0.01034737,
      "balance_loss_clip": 1.05382609,
      "balance_loss_mlp": 1.02494383,
      "epoch": 0.46377682919497387,
      "flos": 24608074590720.0,
      "grad_norm": 1.4941646922765526,
      "language_loss": 0.84082437,
      "learning_rate": 2.330321427836531e-06,
      "loss": 0.8631233,
      "num_input_tokens_seen": 82994080,
      "step": 3857,
      "time_per_iteration": 2.7160699367523193
    },
    {
      "auxiliary_loss_clip": 0.01188756,
      "auxiliary_loss_mlp": 0.01024955,
      "balance_loss_clip": 1.05562687,
      "balance_loss_mlp": 1.01515639,
      "epoch": 0.4638970720856129,
      "flos": 19060953442560.0,
      "grad_norm": 1.9678320433487562,
      "language_loss": 0.82603168,
      "learning_rate": 2.3295531290308733e-06,
      "loss": 0.84816885,
      "num_input_tokens_seen": 83012230,
      "step": 3858,
      "time_per_iteration": 2.596970558166504
    },
    {
      "auxiliary_loss_clip": 0.01192421,
      "auxiliary_loss_mlp": 0.01209399,
      "balance_loss_clip": 1.05850804,
      "balance_loss_mlp": 1.00053072,
      "epoch": 0.46401731497625204,
      "flos": 18471012468480.0,
      "grad_norm": 2.5479008925106776,
      "language_loss": 0.75875854,
      "learning_rate": 2.3287847802322678e-06,
      "loss": 0.78277683,
      "num_input_tokens_seen": 83027800,
      "step": 3859,
      "time_per_iteration": 2.5770864486694336
    },
    {
      "auxiliary_loss_clip": 0.01202558,
      "auxiliary_loss_mlp": 0.01027047,
      "balance_loss_clip": 1.05736303,
      "balance_loss_mlp": 1.01773643,
      "epoch": 0.4641375578668911,
      "flos": 26067053214720.0,
      "grad_norm": 12.370334288792503,
      "language_loss": 0.84075272,
      "learning_rate": 2.3280163815572723e-06,
      "loss": 0.86304879,
      "num_input_tokens_seen": 83048395,
      "step": 3860,
      "time_per_iteration": 3.620699644088745
    },
    {
      "auxiliary_loss_clip": 0.01187723,
      "auxiliary_loss_mlp": 0.01025765,
      "balance_loss_clip": 1.05386066,
      "balance_loss_mlp": 1.01690173,
      "epoch": 0.4642578007575302,
      "flos": 19570382081280.0,
      "grad_norm": 2.134744766460678,
      "language_loss": 0.7690829,
      "learning_rate": 2.3272479331224522e-06,
      "loss": 0.7912178,
      "num_input_tokens_seen": 83065825,
      "step": 3861,
      "time_per_iteration": 2.6072707176208496
    },
    {
      "auxiliary_loss_clip": 0.01190748,
      "auxiliary_loss_mlp": 0.01027266,
      "balance_loss_clip": 1.05701387,
      "balance_loss_mlp": 1.01853967,
      "epoch": 0.4643780436481693,
      "flos": 28186249772160.0,
      "grad_norm": 2.2147256293418223,
      "language_loss": 0.78449035,
      "learning_rate": 2.3264794350443817e-06,
      "loss": 0.80667043,
      "num_input_tokens_seen": 83087920,
      "step": 3862,
      "time_per_iteration": 2.6577377319335938
    },
    {
      "auxiliary_loss_clip": 0.01192289,
      "auxiliary_loss_mlp": 0.0102702,
      "balance_loss_clip": 1.05221856,
      "balance_loss_mlp": 1.0180614,
      "epoch": 0.46449828653880837,
      "flos": 25375270204800.0,
      "grad_norm": 3.327110896680155,
      "language_loss": 0.78980517,
      "learning_rate": 2.3257108874396396e-06,
      "loss": 0.81199837,
      "num_input_tokens_seen": 83109015,
      "step": 3863,
      "time_per_iteration": 2.681955575942993
    },
    {
      "auxiliary_loss_clip": 0.01192799,
      "auxiliary_loss_mlp": 0.0103576,
      "balance_loss_clip": 1.05275393,
      "balance_loss_mlp": 1.02628839,
      "epoch": 0.4646185294294475,
      "flos": 16034330574720.0,
      "grad_norm": 2.5769252580687048,
      "language_loss": 0.73984098,
      "learning_rate": 2.3249422904248152e-06,
      "loss": 0.76212656,
      "num_input_tokens_seen": 83127450,
      "step": 3864,
      "time_per_iteration": 2.692171335220337
    },
    {
      "auxiliary_loss_clip": 0.01195251,
      "auxiliary_loss_mlp": 0.01032963,
      "balance_loss_clip": 1.0560813,
      "balance_loss_mlp": 1.02440369,
      "epoch": 0.4647387723200866,
      "flos": 26363101109760.0,
      "grad_norm": 1.473483708597874,
      "language_loss": 0.87357795,
      "learning_rate": 2.324173644116504e-06,
      "loss": 0.89586008,
      "num_input_tokens_seen": 83150300,
      "step": 3865,
      "time_per_iteration": 2.749675750732422
    },
    {
      "auxiliary_loss_clip": 0.01191017,
      "auxiliary_loss_mlp": 0.0102917,
      "balance_loss_clip": 1.0584867,
      "balance_loss_mlp": 1.02059317,
      "epoch": 0.46485901521072565,
      "flos": 27160209774720.0,
      "grad_norm": 2.0887435899356146,
      "language_loss": 0.8135277,
      "learning_rate": 2.3234049486313087e-06,
      "loss": 0.8357296,
      "num_input_tokens_seen": 83171750,
      "step": 3866,
      "time_per_iteration": 3.5669219493865967
    },
    {
      "auxiliary_loss_clip": 0.01191664,
      "auxiliary_loss_mlp": 0.01029563,
      "balance_loss_clip": 1.05723047,
      "balance_loss_mlp": 1.02133787,
      "epoch": 0.46497925810136476,
      "flos": 24279851088000.0,
      "grad_norm": 2.008754721593099,
      "language_loss": 0.76094902,
      "learning_rate": 2.322636204085839e-06,
      "loss": 0.78316128,
      "num_input_tokens_seen": 83191820,
      "step": 3867,
      "time_per_iteration": 2.7588417530059814
    },
    {
      "auxiliary_loss_clip": 0.01185729,
      "auxiliary_loss_mlp": 0.01034949,
      "balance_loss_clip": 1.05140591,
      "balance_loss_mlp": 1.02568007,
      "epoch": 0.46509950099200387,
      "flos": 16253134272000.0,
      "grad_norm": 2.2500294364815625,
      "language_loss": 0.79264593,
      "learning_rate": 2.3218674105967143e-06,
      "loss": 0.81485271,
      "num_input_tokens_seen": 83210085,
      "step": 3868,
      "time_per_iteration": 2.698734998703003
    },
    {
      "auxiliary_loss_clip": 0.01188647,
      "auxiliary_loss_mlp": 0.01031084,
      "balance_loss_clip": 1.05295718,
      "balance_loss_mlp": 1.02249479,
      "epoch": 0.4652197438826429,
      "flos": 23442270773760.0,
      "grad_norm": 1.7164098436832413,
      "language_loss": 0.83306944,
      "learning_rate": 2.3210985682805593e-06,
      "loss": 0.85526681,
      "num_input_tokens_seen": 83231865,
      "step": 3869,
      "time_per_iteration": 2.6887741088867188
    },
    {
      "auxiliary_loss_clip": 0.01192395,
      "auxiliary_loss_mlp": 0.01027813,
      "balance_loss_clip": 1.05990863,
      "balance_loss_mlp": 1.01903343,
      "epoch": 0.46533998677328203,
      "flos": 16216397637120.0,
      "grad_norm": 3.7553778683935044,
      "language_loss": 0.68059391,
      "learning_rate": 2.320329677254007e-06,
      "loss": 0.70279598,
      "num_input_tokens_seen": 83249195,
      "step": 3870,
      "time_per_iteration": 2.536663055419922
    },
    {
      "auxiliary_loss_clip": 0.0118964,
      "auxiliary_loss_mlp": 0.01027615,
      "balance_loss_clip": 1.05806553,
      "balance_loss_mlp": 1.01866889,
      "epoch": 0.46546022966392114,
      "flos": 21141869080320.0,
      "grad_norm": 2.881180269804317,
      "language_loss": 0.73165894,
      "learning_rate": 2.319560737633697e-06,
      "loss": 0.75383151,
      "num_input_tokens_seen": 83267915,
      "step": 3871,
      "time_per_iteration": 2.64172101020813
    },
    {
      "auxiliary_loss_clip": 0.01201362,
      "auxiliary_loss_mlp": 0.01028096,
      "balance_loss_clip": 1.05095661,
      "balance_loss_mlp": 1.01869059,
      "epoch": 0.4655804725545602,
      "flos": 41171942442240.0,
      "grad_norm": 2.8355537414469687,
      "language_loss": 0.67958844,
      "learning_rate": 2.3187917495362775e-06,
      "loss": 0.70188308,
      "num_input_tokens_seen": 83292325,
      "step": 3872,
      "time_per_iteration": 2.814552068710327
    },
    {
      "auxiliary_loss_clip": 0.01197622,
      "auxiliary_loss_mlp": 0.01037177,
      "balance_loss_clip": 1.05230021,
      "balance_loss_mlp": 1.02791429,
      "epoch": 0.4657007154451993,
      "flos": 19570956698880.0,
      "grad_norm": 3.7134442880877105,
      "language_loss": 0.76410103,
      "learning_rate": 2.318022713078403e-06,
      "loss": 0.78644896,
      "num_input_tokens_seen": 83306905,
      "step": 3873,
      "time_per_iteration": 2.612169027328491
    },
    {
      "auxiliary_loss_clip": 0.01192516,
      "auxiliary_loss_mlp": 0.01034248,
      "balance_loss_clip": 1.0541774,
      "balance_loss_mlp": 1.02523541,
      "epoch": 0.4658209583358384,
      "flos": 15517826956800.0,
      "grad_norm": 2.2174184639700734,
      "language_loss": 0.85137546,
      "learning_rate": 2.3172536283767354e-06,
      "loss": 0.87364304,
      "num_input_tokens_seen": 83320665,
      "step": 3874,
      "time_per_iteration": 2.575909376144409
    },
    {
      "auxiliary_loss_clip": 0.0119483,
      "auxiliary_loss_mlp": 0.01025621,
      "balance_loss_clip": 1.0548737,
      "balance_loss_mlp": 1.01592326,
      "epoch": 0.4659412012264775,
      "flos": 14903180403840.0,
      "grad_norm": 2.018581540660249,
      "language_loss": 0.81098509,
      "learning_rate": 2.3164844955479447e-06,
      "loss": 0.83318955,
      "num_input_tokens_seen": 83336475,
      "step": 3875,
      "time_per_iteration": 2.6515464782714844
    },
    {
      "auxiliary_loss_clip": 0.01209722,
      "auxiliary_loss_mlp": 0.01029358,
      "balance_loss_clip": 1.05338478,
      "balance_loss_mlp": 1.0200597,
      "epoch": 0.4660614441171166,
      "flos": 24425612478720.0,
      "grad_norm": 1.9945251869860068,
      "language_loss": 0.70687461,
      "learning_rate": 2.3157153147087082e-06,
      "loss": 0.72926545,
      "num_input_tokens_seen": 83358365,
      "step": 3876,
      "time_per_iteration": 2.7358758449554443
    },
    {
      "auxiliary_loss_clip": 0.0120792,
      "auxiliary_loss_mlp": 0.01028931,
      "balance_loss_clip": 1.05526507,
      "balance_loss_mlp": 1.02000773,
      "epoch": 0.46618168700775564,
      "flos": 22091095843200.0,
      "grad_norm": 1.7405430985363965,
      "language_loss": 0.83473146,
      "learning_rate": 2.314946085975709e-06,
      "loss": 0.85709995,
      "num_input_tokens_seen": 83377345,
      "step": 3877,
      "time_per_iteration": 2.6826233863830566
    },
    {
      "auxiliary_loss_clip": 0.01186669,
      "auxiliary_loss_mlp": 0.0103264,
      "balance_loss_clip": 1.05418706,
      "balance_loss_mlp": 1.0239141,
      "epoch": 0.46630192989839475,
      "flos": 26176975810560.0,
      "grad_norm": 1.7714206652565092,
      "language_loss": 0.82331455,
      "learning_rate": 2.3141768094656393e-06,
      "loss": 0.84550762,
      "num_input_tokens_seen": 83395920,
      "step": 3878,
      "time_per_iteration": 2.7259645462036133
    },
    {
      "auxiliary_loss_clip": 0.01211356,
      "auxiliary_loss_mlp": 0.01028451,
      "balance_loss_clip": 1.04771423,
      "balance_loss_mlp": 1.02012444,
      "epoch": 0.46642217278903386,
      "flos": 11509622150400.0,
      "grad_norm": 3.3498394934065083,
      "language_loss": 0.82668948,
      "learning_rate": 2.3134074852951966e-06,
      "loss": 0.84908754,
      "num_input_tokens_seen": 83412510,
      "step": 3879,
      "time_per_iteration": 2.7114927768707275
    },
    {
      "auxiliary_loss_clip": 0.01188737,
      "auxiliary_loss_mlp": 0.01028548,
      "balance_loss_clip": 1.04535127,
      "balance_loss_mlp": 1.02033448,
      "epoch": 0.4665424156796729,
      "flos": 32306819299200.0,
      "grad_norm": 4.9908067150559985,
      "language_loss": 0.77609044,
      "learning_rate": 2.312638113581088e-06,
      "loss": 0.79826331,
      "num_input_tokens_seen": 83432995,
      "step": 3880,
      "time_per_iteration": 2.774785041809082
    },
    {
      "auxiliary_loss_clip": 0.01190793,
      "auxiliary_loss_mlp": 0.01029438,
      "balance_loss_clip": 1.05382729,
      "balance_loss_mlp": 1.02053308,
      "epoch": 0.46666265857031203,
      "flos": 18436179254400.0,
      "grad_norm": 2.5380922349730697,
      "language_loss": 0.77635908,
      "learning_rate": 2.311868694440027e-06,
      "loss": 0.79856139,
      "num_input_tokens_seen": 83447415,
      "step": 3881,
      "time_per_iteration": 2.58858060836792
    },
    {
      "auxiliary_loss_clip": 0.0109025,
      "auxiliary_loss_mlp": 0.01002541,
      "balance_loss_clip": 1.02570748,
      "balance_loss_mlp": 1.00146198,
      "epoch": 0.46678290146095114,
      "flos": 68438989221120.0,
      "grad_norm": 0.7330414377092915,
      "language_loss": 0.62444896,
      "learning_rate": 2.3110992279887323e-06,
      "loss": 0.64537686,
      "num_input_tokens_seen": 83519340,
      "step": 3882,
      "time_per_iteration": 4.192864418029785
    },
    {
      "auxiliary_loss_clip": 0.01202193,
      "auxiliary_loss_mlp": 0.01029772,
      "balance_loss_clip": 1.0554539,
      "balance_loss_mlp": 1.02072442,
      "epoch": 0.4669031443515902,
      "flos": 17712507945600.0,
      "grad_norm": 2.18423231124009,
      "language_loss": 0.84773469,
      "learning_rate": 2.310329714343932e-06,
      "loss": 0.87005436,
      "num_input_tokens_seen": 83535490,
      "step": 3883,
      "time_per_iteration": 3.5615270137786865
    },
    {
      "auxiliary_loss_clip": 0.01189031,
      "auxiliary_loss_mlp": 0.01022935,
      "balance_loss_clip": 1.05403996,
      "balance_loss_mlp": 1.01433456,
      "epoch": 0.4670233872422293,
      "flos": 23947748916480.0,
      "grad_norm": 2.0233657441289146,
      "language_loss": 0.81827563,
      "learning_rate": 2.309560153622361e-06,
      "loss": 0.84039527,
      "num_input_tokens_seen": 83552400,
      "step": 3884,
      "time_per_iteration": 2.6039886474609375
    },
    {
      "auxiliary_loss_clip": 0.01195815,
      "auxiliary_loss_mlp": 0.01029283,
      "balance_loss_clip": 1.05357051,
      "balance_loss_mlp": 1.02002692,
      "epoch": 0.4671436301328684,
      "flos": 28111268131200.0,
      "grad_norm": 2.259314016121811,
      "language_loss": 0.74719656,
      "learning_rate": 2.3087905459407602e-06,
      "loss": 0.76944751,
      "num_input_tokens_seen": 83571340,
      "step": 3885,
      "time_per_iteration": 2.7244813442230225
    },
    {
      "auxiliary_loss_clip": 0.010966,
      "auxiliary_loss_mlp": 0.0100279,
      "balance_loss_clip": 1.02446246,
      "balance_loss_mlp": 1.00174665,
      "epoch": 0.46726387302350747,
      "flos": 69369684566400.0,
      "grad_norm": 0.7906951468868747,
      "language_loss": 0.62893319,
      "learning_rate": 2.3080208914158795e-06,
      "loss": 0.64992708,
      "num_input_tokens_seen": 83634340,
      "step": 3886,
      "time_per_iteration": 3.165342092514038
    },
    {
      "auxiliary_loss_clip": 0.01194206,
      "auxiliary_loss_mlp": 0.01026784,
      "balance_loss_clip": 1.0579561,
      "balance_loss_mlp": 1.01846945,
      "epoch": 0.4673841159141466,
      "flos": 25519666878720.0,
      "grad_norm": 2.714681717711558,
      "language_loss": 0.72151577,
      "learning_rate": 2.3072511901644753e-06,
      "loss": 0.74372566,
      "num_input_tokens_seen": 83653410,
      "step": 3887,
      "time_per_iteration": 3.6787824630737305
    },
    {
      "auxiliary_loss_clip": 0.01189386,
      "auxiliary_loss_mlp": 0.01023174,
      "balance_loss_clip": 1.05942559,
      "balance_loss_mlp": 1.01469851,
      "epoch": 0.4675043588047857,
      "flos": 24499265316480.0,
      "grad_norm": 1.8540636741053886,
      "language_loss": 0.80656356,
      "learning_rate": 2.306481442303309e-06,
      "loss": 0.8286891,
      "num_input_tokens_seen": 83672985,
      "step": 3888,
      "time_per_iteration": 2.718308448791504
    },
    {
      "auxiliary_loss_clip": 0.01194202,
      "auxiliary_loss_mlp": 0.01028209,
      "balance_loss_clip": 1.0564816,
      "balance_loss_mlp": 1.01900578,
      "epoch": 0.46762460169542475,
      "flos": 20960771685120.0,
      "grad_norm": 2.2668572852083924,
      "language_loss": 0.73285407,
      "learning_rate": 2.3057116479491515e-06,
      "loss": 0.7550782,
      "num_input_tokens_seen": 83692395,
      "step": 3889,
      "time_per_iteration": 2.585232734680176
    },
    {
      "auxiliary_loss_clip": 0.01187418,
      "auxiliary_loss_mlp": 0.01028472,
      "balance_loss_clip": 1.05387807,
      "balance_loss_mlp": 1.01980531,
      "epoch": 0.46774484458606386,
      "flos": 19171666137600.0,
      "grad_norm": 1.8880351119982643,
      "language_loss": 0.76314926,
      "learning_rate": 2.30494180721878e-06,
      "loss": 0.78530812,
      "num_input_tokens_seen": 83709735,
      "step": 3890,
      "time_per_iteration": 2.634981155395508
    },
    {
      "auxiliary_loss_clip": 0.01190929,
      "auxiliary_loss_mlp": 0.01031172,
      "balance_loss_clip": 1.05644917,
      "balance_loss_mlp": 1.02289343,
      "epoch": 0.4678650874767029,
      "flos": 17967689141760.0,
      "grad_norm": 2.299705354769348,
      "language_loss": 0.89795244,
      "learning_rate": 2.3041719202289794e-06,
      "loss": 0.92017347,
      "num_input_tokens_seen": 83725910,
      "step": 3891,
      "time_per_iteration": 2.592317819595337
    },
    {
      "auxiliary_loss_clip": 0.01191641,
      "auxiliary_loss_mlp": 0.01027602,
      "balance_loss_clip": 1.05720186,
      "balance_loss_mlp": 1.01897693,
      "epoch": 0.467985330367342,
      "flos": 21360816432000.0,
      "grad_norm": 2.746214626698971,
      "language_loss": 0.80604076,
      "learning_rate": 2.30340198709654e-06,
      "loss": 0.82823318,
      "num_input_tokens_seen": 83745745,
      "step": 3892,
      "time_per_iteration": 2.663865566253662
    },
    {
      "auxiliary_loss_clip": 0.01198732,
      "auxiliary_loss_mlp": 0.01029973,
      "balance_loss_clip": 1.05321527,
      "balance_loss_mlp": 1.02149129,
      "epoch": 0.46810557325798113,
      "flos": 20521835487360.0,
      "grad_norm": 3.2401307539241317,
      "language_loss": 0.74681902,
      "learning_rate": 2.3026320079382605e-06,
      "loss": 0.76910603,
      "num_input_tokens_seen": 83762680,
      "step": 3893,
      "time_per_iteration": 3.535461902618408
    },
    {
      "auxiliary_loss_clip": 0.01188777,
      "auxiliary_loss_mlp": 0.01026897,
      "balance_loss_clip": 1.05887222,
      "balance_loss_mlp": 1.01797986,
      "epoch": 0.4682258161486202,
      "flos": 30117848572800.0,
      "grad_norm": 2.253784705457258,
      "language_loss": 0.76269186,
      "learning_rate": 2.3018619828709454e-06,
      "loss": 0.78484857,
      "num_input_tokens_seen": 83784220,
      "step": 3894,
      "time_per_iteration": 2.6937320232391357
    },
    {
      "auxiliary_loss_clip": 0.01187001,
      "auxiliary_loss_mlp": 0.01208901,
      "balance_loss_clip": 1.05808461,
      "balance_loss_mlp": 1.0006175,
      "epoch": 0.4683460590392593,
      "flos": 25293357239040.0,
      "grad_norm": 1.9484312597304858,
      "language_loss": 0.82218146,
      "learning_rate": 2.3010919120114084e-06,
      "loss": 0.8461405,
      "num_input_tokens_seen": 83800750,
      "step": 3895,
      "time_per_iteration": 2.6700448989868164
    },
    {
      "auxiliary_loss_clip": 0.01187522,
      "auxiliary_loss_mlp": 0.01032853,
      "balance_loss_clip": 1.0523262,
      "balance_loss_mlp": 1.0239538,
      "epoch": 0.4684663019298984,
      "flos": 15368330551680.0,
      "grad_norm": 2.171342424986156,
      "language_loss": 0.66403699,
      "learning_rate": 2.3003217954764672e-06,
      "loss": 0.68624079,
      "num_input_tokens_seen": 83815455,
      "step": 3896,
      "time_per_iteration": 2.5563526153564453
    },
    {
      "auxiliary_loss_clip": 0.0119206,
      "auxiliary_loss_mlp": 0.01024426,
      "balance_loss_clip": 1.05235529,
      "balance_loss_mlp": 1.01543725,
      "epoch": 0.46858654482053747,
      "flos": 27778842737280.0,
      "grad_norm": 2.1552198696385227,
      "language_loss": 0.79459614,
      "learning_rate": 2.299551633382949e-06,
      "loss": 0.81676102,
      "num_input_tokens_seen": 83835765,
      "step": 3897,
      "time_per_iteration": 2.6640126705169678
    },
    {
      "auxiliary_loss_clip": 0.0118667,
      "auxiliary_loss_mlp": 0.01032142,
      "balance_loss_clip": 1.05236936,
      "balance_loss_mlp": 1.02335644,
      "epoch": 0.4687067877111766,
      "flos": 18040623707520.0,
      "grad_norm": 5.316004419688426,
      "language_loss": 0.85940039,
      "learning_rate": 2.2987814258476854e-06,
      "loss": 0.88158858,
      "num_input_tokens_seen": 83853565,
      "step": 3898,
      "time_per_iteration": 2.684967517852783
    },
    {
      "auxiliary_loss_clip": 0.01197389,
      "auxiliary_loss_mlp": 0.01026923,
      "balance_loss_clip": 1.04594779,
      "balance_loss_mlp": 1.01794672,
      "epoch": 0.4688270306018157,
      "flos": 16977380198400.0,
      "grad_norm": 3.27371697838461,
      "language_loss": 0.68138909,
      "learning_rate": 2.2980111729875177e-06,
      "loss": 0.70363224,
      "num_input_tokens_seen": 83869815,
      "step": 3899,
      "time_per_iteration": 2.634925365447998
    },
    {
      "auxiliary_loss_clip": 0.0118663,
      "auxiliary_loss_mlp": 0.01035775,
      "balance_loss_clip": 1.05360198,
      "balance_loss_mlp": 1.0272274,
      "epoch": 0.46894727349245474,
      "flos": 17821640442240.0,
      "grad_norm": 1.7400370235270843,
      "language_loss": 0.82279086,
      "learning_rate": 2.2972408749192917e-06,
      "loss": 0.84501493,
      "num_input_tokens_seen": 83887545,
      "step": 3900,
      "time_per_iteration": 2.6113696098327637
    },
    {
      "auxiliary_loss_clip": 0.01191057,
      "auxiliary_loss_mlp": 0.0120857,
      "balance_loss_clip": 1.05721748,
      "balance_loss_mlp": 1.00059056,
      "epoch": 0.46906751638309385,
      "flos": 21471349559040.0,
      "grad_norm": 2.0733757198618115,
      "language_loss": 0.67024636,
      "learning_rate": 2.296470531759861e-06,
      "loss": 0.6942426,
      "num_input_tokens_seen": 83905645,
      "step": 3901,
      "time_per_iteration": 2.7063496112823486
    },
    {
      "auxiliary_loss_clip": 0.01191272,
      "auxiliary_loss_mlp": 0.01028019,
      "balance_loss_clip": 1.05103993,
      "balance_loss_mlp": 1.01854777,
      "epoch": 0.46918775927373296,
      "flos": 20337829090560.0,
      "grad_norm": 2.512485403709425,
      "language_loss": 0.79677641,
      "learning_rate": 2.2957001436260866e-06,
      "loss": 0.81896925,
      "num_input_tokens_seen": 83922705,
      "step": 3902,
      "time_per_iteration": 2.6816275119781494
    },
    {
      "auxiliary_loss_clip": 0.01191143,
      "auxiliary_loss_mlp": 0.01029209,
      "balance_loss_clip": 1.05528235,
      "balance_loss_mlp": 1.02075696,
      "epoch": 0.469308002164372,
      "flos": 18403249461120.0,
      "grad_norm": 1.5677613044899352,
      "language_loss": 0.72807097,
      "learning_rate": 2.294929710634836e-06,
      "loss": 0.75027454,
      "num_input_tokens_seen": 83940795,
      "step": 3903,
      "time_per_iteration": 2.645106792449951
    },
    {
      "auxiliary_loss_clip": 0.01189794,
      "auxiliary_loss_mlp": 0.01034898,
      "balance_loss_clip": 1.05458474,
      "balance_loss_mlp": 1.02561712,
      "epoch": 0.46942824505501113,
      "flos": 37962067363200.0,
      "grad_norm": 1.7485526723545668,
      "language_loss": 0.61122608,
      "learning_rate": 2.2941592329029823e-06,
      "loss": 0.63347298,
      "num_input_tokens_seen": 83961900,
      "step": 3904,
      "time_per_iteration": 2.7207703590393066
    },
    {
      "auxiliary_loss_clip": 0.0118776,
      "auxiliary_loss_mlp": 0.01033846,
      "balance_loss_clip": 1.05527568,
      "balance_loss_mlp": 1.02410686,
      "epoch": 0.46954848794565024,
      "flos": 21872507627520.0,
      "grad_norm": 1.9609079364287014,
      "language_loss": 0.79177904,
      "learning_rate": 2.2933887105474067e-06,
      "loss": 0.81399506,
      "num_input_tokens_seen": 83980075,
      "step": 3905,
      "time_per_iteration": 2.5720503330230713
    },
    {
      "auxiliary_loss_clip": 0.01188939,
      "auxiliary_loss_mlp": 0.0102985,
      "balance_loss_clip": 1.05724812,
      "balance_loss_mlp": 1.02150559,
      "epoch": 0.4696687308362893,
      "flos": 22016545165440.0,
      "grad_norm": 1.614164886286123,
      "language_loss": 0.81446284,
      "learning_rate": 2.2926181436849974e-06,
      "loss": 0.83665073,
      "num_input_tokens_seen": 83999430,
      "step": 3906,
      "time_per_iteration": 2.5193440914154053
    },
    {
      "auxiliary_loss_clip": 0.01192359,
      "auxiliary_loss_mlp": 0.01031542,
      "balance_loss_clip": 1.05870247,
      "balance_loss_mlp": 1.02222598,
      "epoch": 0.4697889737269284,
      "flos": 21613663244160.0,
      "grad_norm": 1.596685002000908,
      "language_loss": 0.72615683,
      "learning_rate": 2.2918475324326478e-06,
      "loss": 0.7483958,
      "num_input_tokens_seen": 84019150,
      "step": 3907,
      "time_per_iteration": 2.5451300144195557
    },
    {
      "auxiliary_loss_clip": 0.01196843,
      "auxiliary_loss_mlp": 0.01209197,
      "balance_loss_clip": 1.0590049,
      "balance_loss_mlp": 1.00067663,
      "epoch": 0.46990921661756746,
      "flos": 25228323665280.0,
      "grad_norm": 2.1047929500067335,
      "language_loss": 0.91372156,
      "learning_rate": 2.2910768769072603e-06,
      "loss": 0.93778193,
      "num_input_tokens_seen": 84037930,
      "step": 3908,
      "time_per_iteration": 2.6421656608581543
    },
    {
      "auxiliary_loss_clip": 0.01187222,
      "auxiliary_loss_mlp": 0.01033034,
      "balance_loss_clip": 1.05620706,
      "balance_loss_mlp": 1.02440882,
      "epoch": 0.47002945950820657,
      "flos": 13844031045120.0,
      "grad_norm": 3.765529379245149,
      "language_loss": 0.75857353,
      "learning_rate": 2.2903061772257417e-06,
      "loss": 0.78077614,
      "num_input_tokens_seen": 84055915,
      "step": 3909,
      "time_per_iteration": 3.675478935241699
    },
    {
      "auxiliary_loss_clip": 0.01190869,
      "auxiliary_loss_mlp": 0.01028137,
      "balance_loss_clip": 1.0569998,
      "balance_loss_mlp": 1.01949465,
      "epoch": 0.4701497023988457,
      "flos": 26247001374720.0,
      "grad_norm": 1.6327291137361204,
      "language_loss": 0.7888701,
      "learning_rate": 2.289535433505007e-06,
      "loss": 0.81106013,
      "num_input_tokens_seen": 84077270,
      "step": 3910,
      "time_per_iteration": 3.7161741256713867
    },
    {
      "auxiliary_loss_clip": 0.01197512,
      "auxiliary_loss_mlp": 0.01028407,
      "balance_loss_clip": 1.05520391,
      "balance_loss_mlp": 1.01955557,
      "epoch": 0.47026994528948474,
      "flos": 25629517647360.0,
      "grad_norm": 1.8317002452033762,
      "language_loss": 0.63867366,
      "learning_rate": 2.2887646458619767e-06,
      "loss": 0.66093284,
      "num_input_tokens_seen": 84098635,
      "step": 3911,
      "time_per_iteration": 2.6581084728240967
    },
    {
      "auxiliary_loss_clip": 0.01203469,
      "auxiliary_loss_mlp": 0.01034558,
      "balance_loss_clip": 1.05505621,
      "balance_loss_mlp": 1.02522397,
      "epoch": 0.47039018818012385,
      "flos": 20554406144640.0,
      "grad_norm": 1.866277403197009,
      "language_loss": 0.76819158,
      "learning_rate": 2.2879938144135797e-06,
      "loss": 0.79057187,
      "num_input_tokens_seen": 84114740,
      "step": 3912,
      "time_per_iteration": 2.6342222690582275
    },
    {
      "auxiliary_loss_clip": 0.01195559,
      "auxiliary_loss_mlp": 0.0120842,
      "balance_loss_clip": 1.05325222,
      "balance_loss_mlp": 1.00069094,
      "epoch": 0.47051043107076296,
      "flos": 21577249831680.0,
      "grad_norm": 1.6122099309205697,
      "language_loss": 0.74865162,
      "learning_rate": 2.2872229392767496e-06,
      "loss": 0.77269137,
      "num_input_tokens_seen": 84134845,
      "step": 3913,
      "time_per_iteration": 2.669698715209961
    },
    {
      "auxiliary_loss_clip": 0.01197843,
      "auxiliary_loss_mlp": 0.01028586,
      "balance_loss_clip": 1.05931187,
      "balance_loss_mlp": 1.02024746,
      "epoch": 0.470630673961402,
      "flos": 18953185662720.0,
      "grad_norm": 1.791502287716049,
      "language_loss": 0.75081038,
      "learning_rate": 2.286452020568428e-06,
      "loss": 0.77307457,
      "num_input_tokens_seen": 84152920,
      "step": 3914,
      "time_per_iteration": 3.5233397483825684
    },
    {
      "auxiliary_loss_clip": 0.01196045,
      "auxiliary_loss_mlp": 0.01028478,
      "balance_loss_clip": 1.05955112,
      "balance_loss_mlp": 1.01876807,
      "epoch": 0.4707509168520411,
      "flos": 19938969492480.0,
      "grad_norm": 2.7451704229039655,
      "language_loss": 0.73572618,
      "learning_rate": 2.2856810584055637e-06,
      "loss": 0.75797141,
      "num_input_tokens_seen": 84170455,
      "step": 3915,
      "time_per_iteration": 2.699270725250244
    },
    {
      "auxiliary_loss_clip": 0.01191697,
      "auxiliary_loss_mlp": 0.01025259,
      "balance_loss_clip": 1.05571508,
      "balance_loss_mlp": 1.01672983,
      "epoch": 0.47087115974268023,
      "flos": 40118754741120.0,
      "grad_norm": 1.7231356977250425,
      "language_loss": 0.67606306,
      "learning_rate": 2.2849100529051085e-06,
      "loss": 0.69823259,
      "num_input_tokens_seen": 84197390,
      "step": 3916,
      "time_per_iteration": 2.78524112701416
    },
    {
      "auxiliary_loss_clip": 0.01189174,
      "auxiliary_loss_mlp": 0.01029178,
      "balance_loss_clip": 1.05917275,
      "balance_loss_mlp": 1.0200702,
      "epoch": 0.4709914026333193,
      "flos": 13552723745280.0,
      "grad_norm": 2.6510082119637426,
      "language_loss": 0.7981863,
      "learning_rate": 2.284139004184026e-06,
      "loss": 0.82036984,
      "num_input_tokens_seen": 84214620,
      "step": 3917,
      "time_per_iteration": 2.6247551441192627
    },
    {
      "auxiliary_loss_clip": 0.01192519,
      "auxiliary_loss_mlp": 0.01027646,
      "balance_loss_clip": 1.05945539,
      "balance_loss_mlp": 1.01878238,
      "epoch": 0.4711116455239584,
      "flos": 19974628719360.0,
      "grad_norm": 2.1125793269097053,
      "language_loss": 0.74318713,
      "learning_rate": 2.2833679123592814e-06,
      "loss": 0.76538885,
      "num_input_tokens_seen": 84231880,
      "step": 3918,
      "time_per_iteration": 2.5423710346221924
    },
    {
      "auxiliary_loss_clip": 0.01192591,
      "auxiliary_loss_mlp": 0.01031323,
      "balance_loss_clip": 1.05431151,
      "balance_loss_mlp": 1.02183437,
      "epoch": 0.4712318884145975,
      "flos": 32124824064000.0,
      "grad_norm": 2.779259047023982,
      "language_loss": 0.63457268,
      "learning_rate": 2.2825967775478508e-06,
      "loss": 0.65681183,
      "num_input_tokens_seen": 84252980,
      "step": 3919,
      "time_per_iteration": 3.5844390392303467
    },
    {
      "auxiliary_loss_clip": 0.01188626,
      "auxiliary_loss_mlp": 0.01030737,
      "balance_loss_clip": 1.05618644,
      "balance_loss_mlp": 1.02174819,
      "epoch": 0.47135213130523657,
      "flos": 20047850593920.0,
      "grad_norm": 2.9876782963686916,
      "language_loss": 0.83417463,
      "learning_rate": 2.2818255998667135e-06,
      "loss": 0.8563683,
      "num_input_tokens_seen": 84271490,
      "step": 3920,
      "time_per_iteration": 2.6781933307647705
    },
    {
      "auxiliary_loss_clip": 0.01192698,
      "auxiliary_loss_mlp": 0.01024774,
      "balance_loss_clip": 1.05824327,
      "balance_loss_mlp": 1.01677501,
      "epoch": 0.4714723741958757,
      "flos": 19426990988160.0,
      "grad_norm": 2.536956979461819,
      "language_loss": 0.79097593,
      "learning_rate": 2.2810543794328566e-06,
      "loss": 0.81315064,
      "num_input_tokens_seen": 84290525,
      "step": 3921,
      "time_per_iteration": 2.5847995281219482
    },
    {
      "auxiliary_loss_clip": 0.01194997,
      "auxiliary_loss_mlp": 0.01033725,
      "balance_loss_clip": 1.05585361,
      "balance_loss_mlp": 1.0251298,
      "epoch": 0.4715926170865148,
      "flos": 20373883367040.0,
      "grad_norm": 2.036639324148936,
      "language_loss": 0.82246578,
      "learning_rate": 2.2802831163632735e-06,
      "loss": 0.84475303,
      "num_input_tokens_seen": 84309245,
      "step": 3922,
      "time_per_iteration": 2.6103978157043457
    },
    {
      "auxiliary_loss_clip": 0.01202093,
      "auxiliary_loss_mlp": 0.01029352,
      "balance_loss_clip": 1.04986668,
      "balance_loss_mlp": 1.02025616,
      "epoch": 0.47171285997715384,
      "flos": 22672884430080.0,
      "grad_norm": 1.8312344780946246,
      "language_loss": 0.74457753,
      "learning_rate": 2.279511810774965e-06,
      "loss": 0.76689196,
      "num_input_tokens_seen": 84330775,
      "step": 3923,
      "time_per_iteration": 2.6714887619018555
    },
    {
      "auxiliary_loss_clip": 0.0119103,
      "auxiliary_loss_mlp": 0.01033892,
      "balance_loss_clip": 1.05897629,
      "balance_loss_mlp": 1.02482057,
      "epoch": 0.47183310286779295,
      "flos": 21105419754240.0,
      "grad_norm": 1.8851440224256983,
      "language_loss": 0.71380019,
      "learning_rate": 2.2787404627849364e-06,
      "loss": 0.73604947,
      "num_input_tokens_seen": 84349985,
      "step": 3924,
      "time_per_iteration": 2.6255321502685547
    },
    {
      "auxiliary_loss_clip": 0.01190912,
      "auxiliary_loss_mlp": 0.01026381,
      "balance_loss_clip": 1.05296731,
      "balance_loss_mlp": 1.01782203,
      "epoch": 0.471953345758432,
      "flos": 21726566668800.0,
      "grad_norm": 3.390067669685369,
      "language_loss": 0.78938508,
      "learning_rate": 2.277969072510202e-06,
      "loss": 0.81155801,
      "num_input_tokens_seen": 84368965,
      "step": 3925,
      "time_per_iteration": 2.601102828979492
    },
    {
      "auxiliary_loss_clip": 0.01194869,
      "auxiliary_loss_mlp": 0.01026051,
      "balance_loss_clip": 1.05667877,
      "balance_loss_mlp": 1.01765847,
      "epoch": 0.4720735886490711,
      "flos": 19861078849920.0,
      "grad_norm": 1.6903093853451439,
      "language_loss": 0.81354046,
      "learning_rate": 2.2771976400677803e-06,
      "loss": 0.83574963,
      "num_input_tokens_seen": 84387795,
      "step": 3926,
      "time_per_iteration": 2.7000648975372314
    },
    {
      "auxiliary_loss_clip": 0.01185217,
      "auxiliary_loss_mlp": 0.01026869,
      "balance_loss_clip": 1.04649162,
      "balance_loss_mlp": 1.01860762,
      "epoch": 0.47219383153971023,
      "flos": 19171809792000.0,
      "grad_norm": 1.785481516355754,
      "language_loss": 0.79107428,
      "learning_rate": 2.2764261655746965e-06,
      "loss": 0.81319511,
      "num_input_tokens_seen": 84405290,
      "step": 3927,
      "time_per_iteration": 2.6500635147094727
    },
    {
      "auxiliary_loss_clip": 0.01193702,
      "auxiliary_loss_mlp": 0.01028083,
      "balance_loss_clip": 1.05294847,
      "balance_loss_mlp": 1.01917839,
      "epoch": 0.4723140744303493,
      "flos": 23224005780480.0,
      "grad_norm": 1.6072055537327181,
      "language_loss": 0.75930297,
      "learning_rate": 2.2756546491479832e-06,
      "loss": 0.78152078,
      "num_input_tokens_seen": 84426205,
      "step": 3928,
      "time_per_iteration": 2.697951316833496
    },
    {
      "auxiliary_loss_clip": 0.01190894,
      "auxiliary_loss_mlp": 0.01208607,
      "balance_loss_clip": 1.05727124,
      "balance_loss_mlp": 1.00068462,
      "epoch": 0.4724343173209884,
      "flos": 18223265387520.0,
      "grad_norm": 3.4472083119913246,
      "language_loss": 0.80694032,
      "learning_rate": 2.274883090904679e-06,
      "loss": 0.83093524,
      "num_input_tokens_seen": 84443970,
      "step": 3929,
      "time_per_iteration": 2.54648494720459
    },
    {
      "auxiliary_loss_clip": 0.0119309,
      "auxiliary_loss_mlp": 0.01029876,
      "balance_loss_clip": 1.06080055,
      "balance_loss_mlp": 1.02107882,
      "epoch": 0.4725545602116275,
      "flos": 21251037490560.0,
      "grad_norm": 2.2524564177349924,
      "language_loss": 0.67816389,
      "learning_rate": 2.2741114909618283e-06,
      "loss": 0.7003935,
      "num_input_tokens_seen": 84459865,
      "step": 3930,
      "time_per_iteration": 2.634568452835083
    },
    {
      "auxiliary_loss_clip": 0.01197603,
      "auxiliary_loss_mlp": 0.010267,
      "balance_loss_clip": 1.05292368,
      "balance_loss_mlp": 1.01797366,
      "epoch": 0.47267480310226656,
      "flos": 21434002392960.0,
      "grad_norm": 1.7997498664260305,
      "language_loss": 0.71927381,
      "learning_rate": 2.2733398494364828e-06,
      "loss": 0.74151683,
      "num_input_tokens_seen": 84479110,
      "step": 3931,
      "time_per_iteration": 2.652130365371704
    },
    {
      "auxiliary_loss_clip": 0.0118604,
      "auxiliary_loss_mlp": 0.0102592,
      "balance_loss_clip": 1.05455613,
      "balance_loss_mlp": 1.0176115,
      "epoch": 0.47279504599290567,
      "flos": 18770508069120.0,
      "grad_norm": 2.1884556679356293,
      "language_loss": 0.84659147,
      "learning_rate": 2.272568166445699e-06,
      "loss": 0.86871111,
      "num_input_tokens_seen": 84497675,
      "step": 3932,
      "time_per_iteration": 2.692744731903076
    },
    {
      "auxiliary_loss_clip": 0.01192584,
      "auxiliary_loss_mlp": 0.01022334,
      "balance_loss_clip": 1.0552032,
      "balance_loss_mlp": 1.01333952,
      "epoch": 0.4729152888835448,
      "flos": 21105742976640.0,
      "grad_norm": 2.3261752924572616,
      "language_loss": 0.64701754,
      "learning_rate": 2.271796442106541e-06,
      "loss": 0.66916674,
      "num_input_tokens_seen": 84517030,
      "step": 3933,
      "time_per_iteration": 2.6229090690612793
    },
    {
      "auxiliary_loss_clip": 0.01107185,
      "auxiliary_loss_mlp": 0.01001043,
      "balance_loss_clip": 1.02483821,
      "balance_loss_mlp": 1.00011945,
      "epoch": 0.47303553177418384,
      "flos": 70201877840640.0,
      "grad_norm": 0.7964093270844307,
      "language_loss": 0.56550658,
      "learning_rate": 2.271024676536079e-06,
      "loss": 0.58658886,
      "num_input_tokens_seen": 84577290,
      "step": 3934,
      "time_per_iteration": 3.1942672729492188
    },
    {
      "auxiliary_loss_clip": 0.01201121,
      "auxiliary_loss_mlp": 0.01032584,
      "balance_loss_clip": 1.05959177,
      "balance_loss_mlp": 1.0231967,
      "epoch": 0.47315577466482295,
      "flos": 22455122227200.0,
      "grad_norm": 2.1242540541459762,
      "language_loss": 0.73100328,
      "learning_rate": 2.2702528698513894e-06,
      "loss": 0.75334036,
      "num_input_tokens_seen": 84598415,
      "step": 3935,
      "time_per_iteration": 2.708333969116211
    },
    {
      "auxiliary_loss_clip": 0.01194548,
      "auxiliary_loss_mlp": 0.01031152,
      "balance_loss_clip": 1.05271077,
      "balance_loss_mlp": 1.02211618,
      "epoch": 0.47327601755546206,
      "flos": 24352857480960.0,
      "grad_norm": 1.9332763159930348,
      "language_loss": 0.78385526,
      "learning_rate": 2.269481022169554e-06,
      "loss": 0.80611229,
      "num_input_tokens_seen": 84617010,
      "step": 3936,
      "time_per_iteration": 3.6128740310668945
    },
    {
      "auxiliary_loss_clip": 0.01201027,
      "auxiliary_loss_mlp": 0.01027911,
      "balance_loss_clip": 1.05412149,
      "balance_loss_mlp": 1.01879179,
      "epoch": 0.4733962604461011,
      "flos": 22926772736640.0,
      "grad_norm": 3.197958961193085,
      "language_loss": 0.80716956,
      "learning_rate": 2.2687091336076614e-06,
      "loss": 0.82945895,
      "num_input_tokens_seen": 84636350,
      "step": 3937,
      "time_per_iteration": 3.548173427581787
    },
    {
      "auxiliary_loss_clip": 0.01190571,
      "auxiliary_loss_mlp": 0.01029976,
      "balance_loss_clip": 1.05668163,
      "balance_loss_mlp": 1.02121997,
      "epoch": 0.4735165033367402,
      "flos": 18327369980160.0,
      "grad_norm": 2.1134700448362147,
      "language_loss": 0.80184507,
      "learning_rate": 2.267937204282807e-06,
      "loss": 0.82405049,
      "num_input_tokens_seen": 84653490,
      "step": 3938,
      "time_per_iteration": 2.652312755584717
    },
    {
      "auxiliary_loss_clip": 0.0120094,
      "auxiliary_loss_mlp": 0.01028814,
      "balance_loss_clip": 1.06002355,
      "balance_loss_mlp": 1.01976001,
      "epoch": 0.4736367462273793,
      "flos": 23037018554880.0,
      "grad_norm": 2.3779313127797264,
      "language_loss": 0.79168952,
      "learning_rate": 2.2671652343120926e-06,
      "loss": 0.81398708,
      "num_input_tokens_seen": 84673965,
      "step": 3939,
      "time_per_iteration": 2.6252267360687256
    },
    {
      "auxiliary_loss_clip": 0.01190332,
      "auxiliary_loss_mlp": 0.01029115,
      "balance_loss_clip": 1.06027913,
      "balance_loss_mlp": 1.02073503,
      "epoch": 0.4737569891180184,
      "flos": 25374336451200.0,
      "grad_norm": 1.6568244814242927,
      "language_loss": 0.80314243,
      "learning_rate": 2.2663932238126236e-06,
      "loss": 0.82533687,
      "num_input_tokens_seen": 84692525,
      "step": 3940,
      "time_per_iteration": 2.6637911796569824
    },
    {
      "auxiliary_loss_clip": 0.01191076,
      "auxiliary_loss_mlp": 0.01027725,
      "balance_loss_clip": 1.05454099,
      "balance_loss_mlp": 1.01879573,
      "epoch": 0.4738772320086575,
      "flos": 25849326925440.0,
      "grad_norm": 1.524406857004345,
      "language_loss": 0.80222905,
      "learning_rate": 2.265621172901515e-06,
      "loss": 0.82441711,
      "num_input_tokens_seen": 84715640,
      "step": 3941,
      "time_per_iteration": 3.603285789489746
    },
    {
      "auxiliary_loss_clip": 0.01194661,
      "auxiliary_loss_mlp": 0.01035741,
      "balance_loss_clip": 1.06311786,
      "balance_loss_mlp": 1.026443,
      "epoch": 0.47399747489929656,
      "flos": 27564420499200.0,
      "grad_norm": 2.2220570221288836,
      "language_loss": 0.71271765,
      "learning_rate": 2.2648490816958854e-06,
      "loss": 0.73502171,
      "num_input_tokens_seen": 84736635,
      "step": 3942,
      "time_per_iteration": 2.6265969276428223
    },
    {
      "auxiliary_loss_clip": 0.01190206,
      "auxiliary_loss_mlp": 0.01034412,
      "balance_loss_clip": 1.05342603,
      "balance_loss_mlp": 1.02431488,
      "epoch": 0.47411771778993567,
      "flos": 24863650836480.0,
      "grad_norm": 2.2243635288473422,
      "language_loss": 0.72839957,
      "learning_rate": 2.264076950312861e-06,
      "loss": 0.75064576,
      "num_input_tokens_seen": 84755445,
      "step": 3943,
      "time_per_iteration": 2.656461715698242
    },
    {
      "auxiliary_loss_clip": 0.01198793,
      "auxiliary_loss_mlp": 0.01032646,
      "balance_loss_clip": 1.05541039,
      "balance_loss_mlp": 1.02321076,
      "epoch": 0.4742379606805748,
      "flos": 22748009725440.0,
      "grad_norm": 6.166798944477077,
      "language_loss": 0.8269797,
      "learning_rate": 2.2633047788695727e-06,
      "loss": 0.84929407,
      "num_input_tokens_seen": 84775750,
      "step": 3944,
      "time_per_iteration": 2.6471433639526367
    },
    {
      "auxiliary_loss_clip": 0.01195317,
      "auxiliary_loss_mlp": 0.01030677,
      "balance_loss_clip": 1.05716848,
      "balance_loss_mlp": 1.02243352,
      "epoch": 0.47435820357121383,
      "flos": 19681130689920.0,
      "grad_norm": 2.2333431947535463,
      "language_loss": 0.63805646,
      "learning_rate": 2.262532567483159e-06,
      "loss": 0.66031635,
      "num_input_tokens_seen": 84794310,
      "step": 3945,
      "time_per_iteration": 2.653316020965576
    },
    {
      "auxiliary_loss_clip": 0.01193392,
      "auxiliary_loss_mlp": 0.01209592,
      "balance_loss_clip": 1.06045651,
      "balance_loss_mlp": 1.00082564,
      "epoch": 0.47447844646185294,
      "flos": 25228718714880.0,
      "grad_norm": 3.140550100828239,
      "language_loss": 0.80258381,
      "learning_rate": 2.2617603162707635e-06,
      "loss": 0.82661366,
      "num_input_tokens_seen": 84814720,
      "step": 3946,
      "time_per_iteration": 3.525972843170166
    },
    {
      "auxiliary_loss_clip": 0.01190648,
      "auxiliary_loss_mlp": 0.01029155,
      "balance_loss_clip": 1.05992544,
      "balance_loss_mlp": 1.02062559,
      "epoch": 0.47459868935249205,
      "flos": 24570619683840.0,
      "grad_norm": 1.7141381400839275,
      "language_loss": 0.82940775,
      "learning_rate": 2.2609880253495363e-06,
      "loss": 0.85160577,
      "num_input_tokens_seen": 84834355,
      "step": 3947,
      "time_per_iteration": 2.6515092849731445
    },
    {
      "auxiliary_loss_clip": 0.01205866,
      "auxiliary_loss_mlp": 0.01033571,
      "balance_loss_clip": 1.05281174,
      "balance_loss_mlp": 1.02482677,
      "epoch": 0.4747189322431311,
      "flos": 20558500295040.0,
      "grad_norm": 2.1984464868136553,
      "language_loss": 0.86811829,
      "learning_rate": 2.260215694836633e-06,
      "loss": 0.89051265,
      "num_input_tokens_seen": 84853530,
      "step": 3948,
      "time_per_iteration": 2.6397359371185303
    },
    {
      "auxiliary_loss_clip": 0.01200366,
      "auxiliary_loss_mlp": 0.01208802,
      "balance_loss_clip": 1.0490284,
      "balance_loss_mlp": 1.00073719,
      "epoch": 0.4748391751337702,
      "flos": 25995231970560.0,
      "grad_norm": 2.2384975697083638,
      "language_loss": 0.64902693,
      "learning_rate": 2.2594433248492157e-06,
      "loss": 0.67311859,
      "num_input_tokens_seen": 84872505,
      "step": 3949,
      "time_per_iteration": 2.793980121612549
    },
    {
      "auxiliary_loss_clip": 0.01195209,
      "auxiliary_loss_mlp": 0.01030611,
      "balance_loss_clip": 1.05539298,
      "balance_loss_mlp": 1.02221239,
      "epoch": 0.47495941802440933,
      "flos": 22821052032000.0,
      "grad_norm": 5.213032636864402,
      "language_loss": 0.8046639,
      "learning_rate": 2.2586709155044527e-06,
      "loss": 0.82692212,
      "num_input_tokens_seen": 84893105,
      "step": 3950,
      "time_per_iteration": 2.623253345489502
    },
    {
      "auxiliary_loss_clip": 0.01190594,
      "auxiliary_loss_mlp": 0.01026613,
      "balance_loss_clip": 1.05892992,
      "balance_loss_mlp": 1.01773739,
      "epoch": 0.4750796609150484,
      "flos": 27891782075520.0,
      "grad_norm": 1.560556188819023,
      "language_loss": 0.76028264,
      "learning_rate": 2.2578984669195167e-06,
      "loss": 0.78245473,
      "num_input_tokens_seen": 84914070,
      "step": 3951,
      "time_per_iteration": 2.74652361869812
    },
    {
      "auxiliary_loss_clip": 0.01189951,
      "auxiliary_loss_mlp": 0.01027345,
      "balance_loss_clip": 1.05443037,
      "balance_loss_mlp": 1.01919687,
      "epoch": 0.4751999038056875,
      "flos": 35660085471360.0,
      "grad_norm": 2.3995856313285007,
      "language_loss": 0.67907608,
      "learning_rate": 2.2571259792115887e-06,
      "loss": 0.701249,
      "num_input_tokens_seen": 84935290,
      "step": 3952,
      "time_per_iteration": 2.723339319229126
    },
    {
      "auxiliary_loss_clip": 0.01185008,
      "auxiliary_loss_mlp": 0.0102952,
      "balance_loss_clip": 1.05461168,
      "balance_loss_mlp": 1.02149749,
      "epoch": 0.4753201466963266,
      "flos": 22090880361600.0,
      "grad_norm": 1.8955184936677671,
      "language_loss": 0.79758847,
      "learning_rate": 2.2563534524978544e-06,
      "loss": 0.81973374,
      "num_input_tokens_seen": 84952760,
      "step": 3953,
      "time_per_iteration": 2.628328323364258
    },
    {
      "auxiliary_loss_clip": 0.01192259,
      "auxiliary_loss_mlp": 0.01024212,
      "balance_loss_clip": 1.05634987,
      "balance_loss_mlp": 1.01643932,
      "epoch": 0.47544038958696566,
      "flos": 30190854965760.0,
      "grad_norm": 2.356981317075702,
      "language_loss": 0.70763373,
      "learning_rate": 2.2555808868955052e-06,
      "loss": 0.72979844,
      "num_input_tokens_seen": 84974890,
      "step": 3954,
      "time_per_iteration": 2.9186933040618896
    },
    {
      "auxiliary_loss_clip": 0.01199403,
      "auxiliary_loss_mlp": 0.01025938,
      "balance_loss_clip": 1.05298853,
      "balance_loss_mlp": 1.01716375,
      "epoch": 0.47556063247760477,
      "flos": 23472219738240.0,
      "grad_norm": 2.8064278466902692,
      "language_loss": 0.73984218,
      "learning_rate": 2.254808282521738e-06,
      "loss": 0.76209557,
      "num_input_tokens_seen": 84993640,
      "step": 3955,
      "time_per_iteration": 2.67692494392395
    },
    {
      "auxiliary_loss_clip": 0.01198652,
      "auxiliary_loss_mlp": 0.0120924,
      "balance_loss_clip": 1.05234814,
      "balance_loss_mlp": 1.00084841,
      "epoch": 0.4756808753682438,
      "flos": 25155209531520.0,
      "grad_norm": 1.7518650760538237,
      "language_loss": 0.81357288,
      "learning_rate": 2.2540356394937573e-06,
      "loss": 0.83765179,
      "num_input_tokens_seen": 85012340,
      "step": 3956,
      "time_per_iteration": 2.7111756801605225
    },
    {
      "auxiliary_loss_clip": 0.012029,
      "auxiliary_loss_mlp": 0.01028884,
      "balance_loss_clip": 1.05345035,
      "balance_loss_mlp": 1.01991916,
      "epoch": 0.47580111825888294,
      "flos": 15669729573120.0,
      "grad_norm": 2.5248282709634666,
      "language_loss": 0.84046316,
      "learning_rate": 2.253262957928772e-06,
      "loss": 0.86278105,
      "num_input_tokens_seen": 85029225,
      "step": 3957,
      "time_per_iteration": 2.6233649253845215
    },
    {
      "auxiliary_loss_clip": 0.01187741,
      "auxiliary_loss_mlp": 0.01025571,
      "balance_loss_clip": 1.05263424,
      "balance_loss_mlp": 1.01679718,
      "epoch": 0.47592136114952205,
      "flos": 17636556637440.0,
      "grad_norm": 3.40312602367979,
      "language_loss": 0.7250756,
      "learning_rate": 2.2524902379439976e-06,
      "loss": 0.74720871,
      "num_input_tokens_seen": 85047895,
      "step": 3958,
      "time_per_iteration": 2.5840322971343994
    },
    {
      "auxiliary_loss_clip": 0.01116593,
      "auxiliary_loss_mlp": 0.01002089,
      "balance_loss_clip": 1.02917624,
      "balance_loss_mlp": 1.00100994,
      "epoch": 0.4760416040401611,
      "flos": 61417159292160.0,
      "grad_norm": 0.7562984676554046,
      "language_loss": 0.63684511,
      "learning_rate": 2.251717479656655e-06,
      "loss": 0.65803194,
      "num_input_tokens_seen": 85112690,
      "step": 3959,
      "time_per_iteration": 3.272463321685791
    },
    {
      "auxiliary_loss_clip": 0.01190956,
      "auxiliary_loss_mlp": 0.0102845,
      "balance_loss_clip": 1.0589509,
      "balance_loss_mlp": 1.01932502,
      "epoch": 0.4761618469308002,
      "flos": 18405871153920.0,
      "grad_norm": 1.8059445390949165,
      "language_loss": 0.76026332,
      "learning_rate": 2.2509446831839704e-06,
      "loss": 0.78245735,
      "num_input_tokens_seen": 85132130,
      "step": 3960,
      "time_per_iteration": 2.9650826454162598
    },
    {
      "auxiliary_loss_clip": 0.01194323,
      "auxiliary_loss_mlp": 0.01031273,
      "balance_loss_clip": 1.05296516,
      "balance_loss_mlp": 1.02224326,
      "epoch": 0.4762820898214393,
      "flos": 18040911016320.0,
      "grad_norm": 2.3681202637295407,
      "language_loss": 0.82236409,
      "learning_rate": 2.250171848643177e-06,
      "loss": 0.84462011,
      "num_input_tokens_seen": 85149420,
      "step": 3961,
      "time_per_iteration": 2.6556782722473145
    },
    {
      "auxiliary_loss_clip": 0.0119334,
      "auxiliary_loss_mlp": 0.01030403,
      "balance_loss_clip": 1.05792534,
      "balance_loss_mlp": 1.02251744,
      "epoch": 0.4764023327120784,
      "flos": 19318253541120.0,
      "grad_norm": 2.3522895021003847,
      "language_loss": 0.85879135,
      "learning_rate": 2.249398976151513e-06,
      "loss": 0.88102877,
      "num_input_tokens_seen": 85166970,
      "step": 3962,
      "time_per_iteration": 3.525852680206299
    },
    {
      "auxiliary_loss_clip": 0.01189051,
      "auxiliary_loss_mlp": 0.01030637,
      "balance_loss_clip": 1.05915833,
      "balance_loss_mlp": 1.02227485,
      "epoch": 0.4765225756027175,
      "flos": 22747255539840.0,
      "grad_norm": 3.1938826734501626,
      "language_loss": 0.78729606,
      "learning_rate": 2.248626065826223e-06,
      "loss": 0.80949295,
      "num_input_tokens_seen": 85185175,
      "step": 3963,
      "time_per_iteration": 2.6238996982574463
    },
    {
      "auxiliary_loss_clip": 0.01087761,
      "auxiliary_loss_mlp": 0.01000622,
      "balance_loss_clip": 1.02463675,
      "balance_loss_mlp": 0.99971008,
      "epoch": 0.4766428184933566,
      "flos": 65933392106880.0,
      "grad_norm": 0.7646968191467808,
      "language_loss": 0.62565219,
      "learning_rate": 2.2478531177845564e-06,
      "loss": 0.64653611,
      "num_input_tokens_seen": 85246170,
      "step": 3964,
      "time_per_iteration": 4.027915716171265
    },
    {
      "auxiliary_loss_clip": 0.01195443,
      "auxiliary_loss_mlp": 0.01027071,
      "balance_loss_clip": 1.05578196,
      "balance_loss_mlp": 1.01914012,
      "epoch": 0.47676306138399566,
      "flos": 24136495908480.0,
      "grad_norm": 1.7171097549038516,
      "language_loss": 0.84992671,
      "learning_rate": 2.247080132143769e-06,
      "loss": 0.87215185,
      "num_input_tokens_seen": 85268525,
      "step": 3965,
      "time_per_iteration": 2.7054495811462402
    },
    {
      "auxiliary_loss_clip": 0.01191215,
      "auxiliary_loss_mlp": 0.01025767,
      "balance_loss_clip": 1.04674959,
      "balance_loss_mlp": 1.01656997,
      "epoch": 0.47688330427463477,
      "flos": 12604322995200.0,
      "grad_norm": 3.042985133160927,
      "language_loss": 0.69278669,
      "learning_rate": 2.246307109021121e-06,
      "loss": 0.71495652,
      "num_input_tokens_seen": 85285930,
      "step": 3966,
      "time_per_iteration": 2.66654634475708
    },
    {
      "auxiliary_loss_clip": 0.01188557,
      "auxiliary_loss_mlp": 0.01029203,
      "balance_loss_clip": 1.05124187,
      "balance_loss_mlp": 1.02078104,
      "epoch": 0.4770035471652739,
      "flos": 21390585828480.0,
      "grad_norm": 1.8842273802100682,
      "language_loss": 0.82108212,
      "learning_rate": 2.2455340485338817e-06,
      "loss": 0.84325969,
      "num_input_tokens_seen": 85303565,
      "step": 3967,
      "time_per_iteration": 2.645031452178955
    },
    {
      "auxiliary_loss_clip": 0.01192226,
      "auxiliary_loss_mlp": 0.01026361,
      "balance_loss_clip": 1.05578303,
      "balance_loss_mlp": 1.01792669,
      "epoch": 0.47712379005591293,
      "flos": 25156251025920.0,
      "grad_norm": 2.1638001056987166,
      "language_loss": 0.68101156,
      "learning_rate": 2.244760950799322e-06,
      "loss": 0.70319748,
      "num_input_tokens_seen": 85321835,
      "step": 3968,
      "time_per_iteration": 3.660325288772583
    },
    {
      "auxiliary_loss_clip": 0.01181933,
      "auxiliary_loss_mlp": 0.01024352,
      "balance_loss_clip": 1.05115259,
      "balance_loss_mlp": 1.01592398,
      "epoch": 0.47724403294655204,
      "flos": 22054323294720.0,
      "grad_norm": 1.9945585166520412,
      "language_loss": 0.72981691,
      "learning_rate": 2.2439878159347203e-06,
      "loss": 0.75187969,
      "num_input_tokens_seen": 85341260,
      "step": 3969,
      "time_per_iteration": 2.6646852493286133
    },
    {
      "auxiliary_loss_clip": 0.01086661,
      "auxiliary_loss_mlp": 0.01002667,
      "balance_loss_clip": 1.02365291,
      "balance_loss_mlp": 1.0017972,
      "epoch": 0.4773642758371911,
      "flos": 70229387658240.0,
      "grad_norm": 0.7286287215459291,
      "language_loss": 0.55206311,
      "learning_rate": 2.2432146440573616e-06,
      "loss": 0.57295644,
      "num_input_tokens_seen": 85407220,
      "step": 3970,
      "time_per_iteration": 3.2202577590942383
    },
    {
      "auxiliary_loss_clip": 0.01190533,
      "auxiliary_loss_mlp": 0.01024928,
      "balance_loss_clip": 1.05417299,
      "balance_loss_mlp": 1.01644647,
      "epoch": 0.4774845187278302,
      "flos": 23548602009600.0,
      "grad_norm": 9.203502053747796,
      "language_loss": 0.66393322,
      "learning_rate": 2.242441435284534e-06,
      "loss": 0.68608785,
      "num_input_tokens_seen": 85426095,
      "step": 3971,
      "time_per_iteration": 2.6952154636383057
    },
    {
      "auxiliary_loss_clip": 0.01193132,
      "auxiliary_loss_mlp": 0.01031096,
      "balance_loss_clip": 1.05762529,
      "balance_loss_mlp": 1.02115345,
      "epoch": 0.4776047616184693,
      "flos": 23075371301760.0,
      "grad_norm": 2.2647403015135814,
      "language_loss": 0.8527506,
      "learning_rate": 2.2416681897335337e-06,
      "loss": 0.87499291,
      "num_input_tokens_seen": 85444245,
      "step": 3972,
      "time_per_iteration": 2.642953634262085
    },
    {
      "auxiliary_loss_clip": 0.01196575,
      "auxiliary_loss_mlp": 0.01030694,
      "balance_loss_clip": 1.05239439,
      "balance_loss_mlp": 1.02172923,
      "epoch": 0.4777250045091084,
      "flos": 31898119374720.0,
      "grad_norm": 1.9795511895745266,
      "language_loss": 0.67658669,
      "learning_rate": 2.240894907521661e-06,
      "loss": 0.69885945,
      "num_input_tokens_seen": 85463325,
      "step": 3973,
      "time_per_iteration": 3.804938554763794
    },
    {
      "auxiliary_loss_clip": 0.01190891,
      "auxiliary_loss_mlp": 0.01024152,
      "balance_loss_clip": 1.05360532,
      "balance_loss_mlp": 1.01596808,
      "epoch": 0.4778452473997475,
      "flos": 24278163148800.0,
      "grad_norm": 1.8735848857570503,
      "language_loss": 0.63819319,
      "learning_rate": 2.240121588766223e-06,
      "loss": 0.66034365,
      "num_input_tokens_seen": 85483375,
      "step": 3974,
      "time_per_iteration": 2.6199722290039062
    },
    {
      "auxiliary_loss_clip": 0.01185036,
      "auxiliary_loss_mlp": 0.01031831,
      "balance_loss_clip": 1.05084026,
      "balance_loss_mlp": 1.02311659,
      "epoch": 0.4779654902903866,
      "flos": 31575031516800.0,
      "grad_norm": 1.8487866710157024,
      "language_loss": 0.71278238,
      "learning_rate": 2.239348233584531e-06,
      "loss": 0.73495102,
      "num_input_tokens_seen": 85504230,
      "step": 3975,
      "time_per_iteration": 2.823641300201416
    },
    {
      "auxiliary_loss_clip": 0.01190422,
      "auxiliary_loss_mlp": 0.01031837,
      "balance_loss_clip": 1.05591488,
      "balance_loss_mlp": 1.0232718,
      "epoch": 0.47808573318102565,
      "flos": 19500428344320.0,
      "grad_norm": 7.629968055507731,
      "language_loss": 0.8046428,
      "learning_rate": 2.2385748420939013e-06,
      "loss": 0.82686543,
      "num_input_tokens_seen": 85523425,
      "step": 3976,
      "time_per_iteration": 2.664623498916626
    },
    {
      "auxiliary_loss_clip": 0.01187375,
      "auxiliary_loss_mlp": 0.01029114,
      "balance_loss_clip": 1.06086349,
      "balance_loss_mlp": 1.020895,
      "epoch": 0.47820597607166476,
      "flos": 22601135013120.0,
      "grad_norm": 2.0258513514396275,
      "language_loss": 0.72599757,
      "learning_rate": 2.2378014144116583e-06,
      "loss": 0.74816239,
      "num_input_tokens_seen": 85542235,
      "step": 3977,
      "time_per_iteration": 2.6529924869537354
    },
    {
      "auxiliary_loss_clip": 0.01191181,
      "auxiliary_loss_mlp": 0.01032761,
      "balance_loss_clip": 1.05858433,
      "balance_loss_mlp": 1.02496529,
      "epoch": 0.4783262189623039,
      "flos": 23003011353600.0,
      "grad_norm": 2.0591871590749085,
      "language_loss": 0.80176222,
      "learning_rate": 2.23702795065513e-06,
      "loss": 0.82400173,
      "num_input_tokens_seen": 85561815,
      "step": 3978,
      "time_per_iteration": 2.622908353805542
    },
    {
      "auxiliary_loss_clip": 0.01095602,
      "auxiliary_loss_mlp": 0.01001288,
      "balance_loss_clip": 1.02479577,
      "balance_loss_mlp": 1.00045907,
      "epoch": 0.47844646185294293,
      "flos": 49772801226240.0,
      "grad_norm": 0.9845432560750308,
      "language_loss": 0.67445803,
      "learning_rate": 2.2362544509416493e-06,
      "loss": 0.69542694,
      "num_input_tokens_seen": 85613930,
      "step": 3979,
      "time_per_iteration": 3.031841516494751
    },
    {
      "auxiliary_loss_clip": 0.01184373,
      "auxiliary_loss_mlp": 0.01032921,
      "balance_loss_clip": 1.05132341,
      "balance_loss_mlp": 1.02448153,
      "epoch": 0.47856670474358204,
      "flos": 20229558520320.0,
      "grad_norm": 2.2945200807355737,
      "language_loss": 0.82713461,
      "learning_rate": 2.2354809153885572e-06,
      "loss": 0.84930748,
      "num_input_tokens_seen": 85631000,
      "step": 3980,
      "time_per_iteration": 2.666505813598633
    },
    {
      "auxiliary_loss_clip": 0.0118714,
      "auxiliary_loss_mlp": 0.01030603,
      "balance_loss_clip": 1.05245852,
      "balance_loss_mlp": 1.02209699,
      "epoch": 0.47868694763422115,
      "flos": 20990936131200.0,
      "grad_norm": 2.4389760951902693,
      "language_loss": 0.82959157,
      "learning_rate": 2.234707344113197e-06,
      "loss": 0.85176897,
      "num_input_tokens_seen": 85649095,
      "step": 3981,
      "time_per_iteration": 2.5911965370178223
    },
    {
      "auxiliary_loss_clip": 0.0118435,
      "auxiliary_loss_mlp": 0.01025444,
      "balance_loss_clip": 1.05685735,
      "balance_loss_mlp": 1.01715875,
      "epoch": 0.4788071905248602,
      "flos": 19026551191680.0,
      "grad_norm": 1.905965848407359,
      "language_loss": 0.77880937,
      "learning_rate": 2.233933737232919e-06,
      "loss": 0.80090737,
      "num_input_tokens_seen": 85666875,
      "step": 3982,
      "time_per_iteration": 2.583115339279175
    },
    {
      "auxiliary_loss_clip": 0.01188195,
      "auxiliary_loss_mlp": 0.0120796,
      "balance_loss_clip": 1.0484612,
      "balance_loss_mlp": 1.00062752,
      "epoch": 0.4789274334154993,
      "flos": 23002221254400.0,
      "grad_norm": 2.4752478003657576,
      "language_loss": 0.78293133,
      "learning_rate": 2.2331600948650793e-06,
      "loss": 0.80689281,
      "num_input_tokens_seen": 85687020,
      "step": 3983,
      "time_per_iteration": 2.681597948074341
    },
    {
      "auxiliary_loss_clip": 0.01183166,
      "auxiliary_loss_mlp": 0.01209108,
      "balance_loss_clip": 1.05250072,
      "balance_loss_mlp": 1.00062835,
      "epoch": 0.4790476763061384,
      "flos": 23075586783360.0,
      "grad_norm": 1.5066063380292471,
      "language_loss": 0.79913652,
      "learning_rate": 2.2323864171270386e-06,
      "loss": 0.8230592,
      "num_input_tokens_seen": 85708290,
      "step": 3984,
      "time_per_iteration": 2.717200994491577
    },
    {
      "auxiliary_loss_clip": 0.01195462,
      "auxiliary_loss_mlp": 0.01026279,
      "balance_loss_clip": 1.04998469,
      "balance_loss_mlp": 1.01688588,
      "epoch": 0.4791679191967775,
      "flos": 21179288073600.0,
      "grad_norm": 1.809360300440244,
      "language_loss": 0.72425747,
      "learning_rate": 2.231612704136164e-06,
      "loss": 0.74647486,
      "num_input_tokens_seen": 85728660,
      "step": 3985,
      "time_per_iteration": 2.6468923091888428
    },
    {
      "auxiliary_loss_clip": 0.01186113,
      "auxiliary_loss_mlp": 0.01030279,
      "balance_loss_clip": 1.05430651,
      "balance_loss_mlp": 1.02157104,
      "epoch": 0.4792881620874166,
      "flos": 22301495758080.0,
      "grad_norm": 4.564845603346577,
      "language_loss": 0.75161493,
      "learning_rate": 2.2308389560098253e-06,
      "loss": 0.77377886,
      "num_input_tokens_seen": 85745035,
      "step": 3986,
      "time_per_iteration": 2.564127206802368
    },
    {
      "auxiliary_loss_clip": 0.01196913,
      "auxiliary_loss_mlp": 0.0102878,
      "balance_loss_clip": 1.05420256,
      "balance_loss_mlp": 1.0197736,
      "epoch": 0.47940840497805565,
      "flos": 17420877423360.0,
      "grad_norm": 3.7052686244681947,
      "language_loss": 0.76996511,
      "learning_rate": 2.2300651728654008e-06,
      "loss": 0.79222202,
      "num_input_tokens_seen": 85760295,
      "step": 3987,
      "time_per_iteration": 2.5994932651519775
    },
    {
      "auxiliary_loss_clip": 0.01089552,
      "auxiliary_loss_mlp": 0.01199347,
      "balance_loss_clip": 1.02336979,
      "balance_loss_mlp": 1.00015795,
      "epoch": 0.47952864786869476,
      "flos": 65358175708800.0,
      "grad_norm": 0.7372019380105879,
      "language_loss": 0.60145342,
      "learning_rate": 2.229291354820272e-06,
      "loss": 0.62434244,
      "num_input_tokens_seen": 85821305,
      "step": 3988,
      "time_per_iteration": 3.204211473464966
    },
    {
      "auxiliary_loss_clip": 0.01185959,
      "auxiliary_loss_mlp": 0.01032137,
      "balance_loss_clip": 1.05242407,
      "balance_loss_mlp": 1.02346444,
      "epoch": 0.47964889075933387,
      "flos": 16799802336000.0,
      "grad_norm": 2.2224306487275705,
      "language_loss": 0.76335597,
      "learning_rate": 2.228517501991828e-06,
      "loss": 0.785537,
      "num_input_tokens_seen": 85840105,
      "step": 3989,
      "time_per_iteration": 3.585076332092285
    },
    {
      "auxiliary_loss_clip": 0.0109682,
      "auxiliary_loss_mlp": 0.01004013,
      "balance_loss_clip": 1.02373338,
      "balance_loss_mlp": 1.00314856,
      "epoch": 0.4797691336499729,
      "flos": 70079244808320.0,
      "grad_norm": 0.8099624726825561,
      "language_loss": 0.61025524,
      "learning_rate": 2.22774361449746e-06,
      "loss": 0.63126355,
      "num_input_tokens_seen": 85896585,
      "step": 3990,
      "time_per_iteration": 3.2445313930511475
    },
    {
      "auxiliary_loss_clip": 0.01199176,
      "auxiliary_loss_mlp": 0.01026995,
      "balance_loss_clip": 1.05013716,
      "balance_loss_mlp": 1.01819777,
      "epoch": 0.47988937654061203,
      "flos": 18953329317120.0,
      "grad_norm": 2.9168367714680725,
      "language_loss": 0.70780289,
      "learning_rate": 2.2269696924545668e-06,
      "loss": 0.73006463,
      "num_input_tokens_seen": 85914415,
      "step": 3991,
      "time_per_iteration": 3.632908582687378
    },
    {
      "auxiliary_loss_clip": 0.01195854,
      "auxiliary_loss_mlp": 0.01032255,
      "balance_loss_clip": 1.05427992,
      "balance_loss_mlp": 1.02382135,
      "epoch": 0.48000961943125114,
      "flos": 14461981649280.0,
      "grad_norm": 2.791830591137234,
      "language_loss": 0.78684592,
      "learning_rate": 2.2261957359805523e-06,
      "loss": 0.80912697,
      "num_input_tokens_seen": 85931650,
      "step": 3992,
      "time_per_iteration": 2.659581422805786
    },
    {
      "auxiliary_loss_clip": 0.01187527,
      "auxiliary_loss_mlp": 0.01025435,
      "balance_loss_clip": 1.05672443,
      "balance_loss_mlp": 1.0167923,
      "epoch": 0.4801298623218902,
      "flos": 27051149105280.0,
      "grad_norm": 1.836010917377567,
      "language_loss": 0.73735219,
      "learning_rate": 2.225421745192823e-06,
      "loss": 0.75948185,
      "num_input_tokens_seen": 85951805,
      "step": 3993,
      "time_per_iteration": 2.6219515800476074
    },
    {
      "auxiliary_loss_clip": 0.01189408,
      "auxiliary_loss_mlp": 0.01027925,
      "balance_loss_clip": 1.05623484,
      "balance_loss_mlp": 1.01905584,
      "epoch": 0.4802501052125293,
      "flos": 26355236031360.0,
      "grad_norm": 3.5923849097326874,
      "language_loss": 0.78321481,
      "learning_rate": 2.2246477202087955e-06,
      "loss": 0.80538815,
      "num_input_tokens_seen": 85972485,
      "step": 3994,
      "time_per_iteration": 3.591461181640625
    },
    {
      "auxiliary_loss_clip": 0.01192757,
      "auxiliary_loss_mlp": 0.01029736,
      "balance_loss_clip": 1.05308294,
      "balance_loss_mlp": 1.02197587,
      "epoch": 0.4803703481031684,
      "flos": 20993916960000.0,
      "grad_norm": 1.6752885955759518,
      "language_loss": 0.83033311,
      "learning_rate": 2.223873661145887e-06,
      "loss": 0.85255808,
      "num_input_tokens_seen": 85992540,
      "step": 3995,
      "time_per_iteration": 2.6644294261932373
    },
    {
      "auxiliary_loss_clip": 0.01190591,
      "auxiliary_loss_mlp": 0.01208477,
      "balance_loss_clip": 1.05782688,
      "balance_loss_mlp": 1.00060987,
      "epoch": 0.4804905909938075,
      "flos": 20703722981760.0,
      "grad_norm": 1.600872289989705,
      "language_loss": 0.71355331,
      "learning_rate": 2.2230995681215226e-06,
      "loss": 0.73754394,
      "num_input_tokens_seen": 86012065,
      "step": 3996,
      "time_per_iteration": 2.6861014366149902
    },
    {
      "auxiliary_loss_clip": 0.01191209,
      "auxiliary_loss_mlp": 0.01025177,
      "balance_loss_clip": 1.05303264,
      "balance_loss_mlp": 1.0169698,
      "epoch": 0.4806108338844466,
      "flos": 16654831044480.0,
      "grad_norm": 2.0760154257559136,
      "language_loss": 0.7828486,
      "learning_rate": 2.2223254412531305e-06,
      "loss": 0.80501246,
      "num_input_tokens_seen": 86029435,
      "step": 3997,
      "time_per_iteration": 2.702674388885498
    },
    {
      "auxiliary_loss_clip": 0.01178115,
      "auxiliary_loss_mlp": 0.01025696,
      "balance_loss_clip": 1.04946852,
      "balance_loss_mlp": 1.01751852,
      "epoch": 0.4807310767750857,
      "flos": 20011329440640.0,
      "grad_norm": 2.109624312561939,
      "language_loss": 0.82668912,
      "learning_rate": 2.221551280658146e-06,
      "loss": 0.84872723,
      "num_input_tokens_seen": 86048495,
      "step": 3998,
      "time_per_iteration": 2.6831536293029785
    },
    {
      "auxiliary_loss_clip": 0.0119219,
      "auxiliary_loss_mlp": 0.01027336,
      "balance_loss_clip": 1.05044985,
      "balance_loss_mlp": 1.01924133,
      "epoch": 0.48085131966572475,
      "flos": 23185257984000.0,
      "grad_norm": 1.6086788661198705,
      "language_loss": 0.74089098,
      "learning_rate": 2.2207770864540085e-06,
      "loss": 0.76308626,
      "num_input_tokens_seen": 86067470,
      "step": 3999,
      "time_per_iteration": 3.760246515274048
    },
    {
      "auxiliary_loss_clip": 0.01182158,
      "auxiliary_loss_mlp": 0.01029997,
      "balance_loss_clip": 1.05122769,
      "balance_loss_mlp": 1.02119935,
      "epoch": 0.48097156255636386,
      "flos": 20558643949440.0,
      "grad_norm": 1.863981422646489,
      "language_loss": 0.73234916,
      "learning_rate": 2.220002858758162e-06,
      "loss": 0.75447077,
      "num_input_tokens_seen": 86085460,
      "step": 4000,
      "time_per_iteration": 2.60097599029541
    },
    {
      "auxiliary_loss_clip": 0.01093167,
      "auxiliary_loss_mlp": 0.01002266,
      "balance_loss_clip": 1.02312899,
      "balance_loss_mlp": 1.00143147,
      "epoch": 0.481091805447003,
      "flos": 70511608817280.0,
      "grad_norm": 0.8877904008327884,
      "language_loss": 0.60795188,
      "learning_rate": 2.2192285976880573e-06,
      "loss": 0.62890613,
      "num_input_tokens_seen": 86149715,
      "step": 4001,
      "time_per_iteration": 3.1432220935821533
    },
    {
      "auxiliary_loss_clip": 0.01196759,
      "auxiliary_loss_mlp": 0.01207946,
      "balance_loss_clip": 1.05106306,
      "balance_loss_mlp": 1.00068259,
      "epoch": 0.48121204833764203,
      "flos": 36428214839040.0,
      "grad_norm": 1.6581276825640952,
      "language_loss": 0.80396056,
      "learning_rate": 2.2184543033611485e-06,
      "loss": 0.82800758,
      "num_input_tokens_seen": 86170795,
      "step": 4002,
      "time_per_iteration": 2.687091112136841
    },
    {
      "auxiliary_loss_clip": 0.01189345,
      "auxiliary_loss_mlp": 0.01028836,
      "balance_loss_clip": 1.05375266,
      "balance_loss_mlp": 1.02074802,
      "epoch": 0.48133229122828114,
      "flos": 27490264871040.0,
      "grad_norm": 3.715755278742156,
      "language_loss": 0.81867367,
      "learning_rate": 2.2176799758948957e-06,
      "loss": 0.84085548,
      "num_input_tokens_seen": 86190955,
      "step": 4003,
      "time_per_iteration": 2.586533308029175
    },
    {
      "auxiliary_loss_clip": 0.01186167,
      "auxiliary_loss_mlp": 0.01033523,
      "balance_loss_clip": 1.0536691,
      "balance_loss_mlp": 1.02537465,
      "epoch": 0.4814525341189202,
      "flos": 43072802179200.0,
      "grad_norm": 2.1111588326362405,
      "language_loss": 0.72818494,
      "learning_rate": 2.2169056154067635e-06,
      "loss": 0.75038183,
      "num_input_tokens_seen": 86214875,
      "step": 4004,
      "time_per_iteration": 2.836237907409668
    },
    {
      "auxiliary_loss_clip": 0.01191096,
      "auxiliary_loss_mlp": 0.01208682,
      "balance_loss_clip": 1.05723095,
      "balance_loss_mlp": 1.00061154,
      "epoch": 0.4815727770095593,
      "flos": 24236901400320.0,
      "grad_norm": 1.8457209079416659,
      "language_loss": 0.82462174,
      "learning_rate": 2.216131222014222e-06,
      "loss": 0.84861952,
      "num_input_tokens_seen": 86232950,
      "step": 4005,
      "time_per_iteration": 2.6529300212860107
    },
    {
      "auxiliary_loss_clip": 0.01184479,
      "auxiliary_loss_mlp": 0.01030281,
      "balance_loss_clip": 1.04947388,
      "balance_loss_mlp": 1.02184105,
      "epoch": 0.4816930199001984,
      "flos": 18113630100480.0,
      "grad_norm": 2.3459023458688577,
      "language_loss": 0.80513179,
      "learning_rate": 2.2153567958347455e-06,
      "loss": 0.82727939,
      "num_input_tokens_seen": 86249160,
      "step": 4006,
      "time_per_iteration": 2.696390151977539
    },
    {
      "auxiliary_loss_clip": 0.01191897,
      "auxiliary_loss_mlp": 0.01025575,
      "balance_loss_clip": 1.05620039,
      "balance_loss_mlp": 1.01665807,
      "epoch": 0.48181326279083747,
      "flos": 17274720983040.0,
      "grad_norm": 2.3860845880418275,
      "language_loss": 0.80155957,
      "learning_rate": 2.214582336985815e-06,
      "loss": 0.82373428,
      "num_input_tokens_seen": 86267060,
      "step": 4007,
      "time_per_iteration": 2.6439781188964844
    },
    {
      "auxiliary_loss_clip": 0.01179989,
      "auxiliary_loss_mlp": 0.01029297,
      "balance_loss_clip": 1.05052269,
      "balance_loss_mlp": 1.02008224,
      "epoch": 0.4819335056814766,
      "flos": 14903252231040.0,
      "grad_norm": 4.93078756346886,
      "language_loss": 0.66650534,
      "learning_rate": 2.2138078455849142e-06,
      "loss": 0.68859816,
      "num_input_tokens_seen": 86285055,
      "step": 4008,
      "time_per_iteration": 2.6586575508117676
    },
    {
      "auxiliary_loss_clip": 0.01193786,
      "auxiliary_loss_mlp": 0.01024256,
      "balance_loss_clip": 1.05532861,
      "balance_loss_mlp": 1.01608372,
      "epoch": 0.4820537485721157,
      "flos": 19244888012160.0,
      "grad_norm": 2.054924856267997,
      "language_loss": 0.78765345,
      "learning_rate": 2.2130333217495334e-06,
      "loss": 0.80983388,
      "num_input_tokens_seen": 86304225,
      "step": 4009,
      "time_per_iteration": 2.588754653930664
    },
    {
      "auxiliary_loss_clip": 0.01185304,
      "auxiliary_loss_mlp": 0.01028097,
      "balance_loss_clip": 1.05151701,
      "balance_loss_mlp": 1.01946592,
      "epoch": 0.48217399146275475,
      "flos": 16033791870720.0,
      "grad_norm": 3.7323384433526194,
      "language_loss": 0.68004358,
      "learning_rate": 2.2122587655971665e-06,
      "loss": 0.70217764,
      "num_input_tokens_seen": 86319170,
      "step": 4010,
      "time_per_iteration": 2.6492977142333984
    },
    {
      "auxiliary_loss_clip": 0.0119016,
      "auxiliary_loss_mlp": 0.01029911,
      "balance_loss_clip": 1.05212355,
      "balance_loss_mlp": 1.02201283,
      "epoch": 0.48229423435339386,
      "flos": 24134197438080.0,
      "grad_norm": 1.5731313332609036,
      "language_loss": 0.64207029,
      "learning_rate": 2.211484177245314e-06,
      "loss": 0.664271,
      "num_input_tokens_seen": 86338760,
      "step": 4011,
      "time_per_iteration": 2.655385971069336
    },
    {
      "auxiliary_loss_clip": 0.01186711,
      "auxiliary_loss_mlp": 0.01032658,
      "balance_loss_clip": 1.05574226,
      "balance_loss_mlp": 1.02412319,
      "epoch": 0.48241447724403297,
      "flos": 23805435231360.0,
      "grad_norm": 8.453080959605906,
      "language_loss": 0.72525728,
      "learning_rate": 2.21070955681148e-06,
      "loss": 0.74745089,
      "num_input_tokens_seen": 86357865,
      "step": 4012,
      "time_per_iteration": 2.6945388317108154
    },
    {
      "auxiliary_loss_clip": 0.01182494,
      "auxiliary_loss_mlp": 0.01026575,
      "balance_loss_clip": 1.04930031,
      "balance_loss_mlp": 1.01841521,
      "epoch": 0.482534720134672,
      "flos": 23110312256640.0,
      "grad_norm": 2.112845566881724,
      "language_loss": 0.78545564,
      "learning_rate": 2.209934904413174e-06,
      "loss": 0.80754632,
      "num_input_tokens_seen": 86379470,
      "step": 4013,
      "time_per_iteration": 2.71563720703125
    },
    {
      "auxiliary_loss_clip": 0.01191065,
      "auxiliary_loss_mlp": 0.0102875,
      "balance_loss_clip": 1.04044938,
      "balance_loss_mlp": 1.01960111,
      "epoch": 0.48265496302531113,
      "flos": 20923819568640.0,
      "grad_norm": 2.498364684316547,
      "language_loss": 0.71751463,
      "learning_rate": 2.2091602201679095e-06,
      "loss": 0.73971272,
      "num_input_tokens_seen": 86399080,
      "step": 4014,
      "time_per_iteration": 2.759709358215332
    },
    {
      "auxiliary_loss_clip": 0.01198517,
      "auxiliary_loss_mlp": 0.0102776,
      "balance_loss_clip": 1.05394709,
      "balance_loss_mlp": 1.01962364,
      "epoch": 0.48277520591595025,
      "flos": 15231152511360.0,
      "grad_norm": 3.3024555536260407,
      "language_loss": 0.83284104,
      "learning_rate": 2.208385504193206e-06,
      "loss": 0.85510373,
      "num_input_tokens_seen": 86416580,
      "step": 4015,
      "time_per_iteration": 2.713184356689453
    },
    {
      "auxiliary_loss_clip": 0.01185406,
      "auxiliary_loss_mlp": 0.01022458,
      "balance_loss_clip": 1.05511904,
      "balance_loss_mlp": 1.01425624,
      "epoch": 0.4828954488065893,
      "flos": 17858664385920.0,
      "grad_norm": 3.4437020137734584,
      "language_loss": 0.81293833,
      "learning_rate": 2.2076107566065873e-06,
      "loss": 0.83501697,
      "num_input_tokens_seen": 86434365,
      "step": 4016,
      "time_per_iteration": 3.5478594303131104
    },
    {
      "auxiliary_loss_clip": 0.01191297,
      "auxiliary_loss_mlp": 0.01029716,
      "balance_loss_clip": 1.05504596,
      "balance_loss_mlp": 1.02222347,
      "epoch": 0.4830156916972284,
      "flos": 32087405070720.0,
      "grad_norm": 2.138340183179873,
      "language_loss": 0.75927329,
      "learning_rate": 2.2068359775255816e-06,
      "loss": 0.78148341,
      "num_input_tokens_seen": 86452675,
      "step": 4017,
      "time_per_iteration": 2.768653154373169
    },
    {
      "auxiliary_loss_clip": 0.01188895,
      "auxiliary_loss_mlp": 0.01027439,
      "balance_loss_clip": 1.04815149,
      "balance_loss_mlp": 1.01905227,
      "epoch": 0.48313593458786747,
      "flos": 21871717528320.0,
      "grad_norm": 2.4433461577050313,
      "language_loss": 0.78108215,
      "learning_rate": 2.206061167067723e-06,
      "loss": 0.80324543,
      "num_input_tokens_seen": 86470785,
      "step": 4018,
      "time_per_iteration": 3.5888307094573975
    },
    {
      "auxiliary_loss_clip": 0.01190064,
      "auxiliary_loss_mlp": 0.01026221,
      "balance_loss_clip": 1.04758525,
      "balance_loss_mlp": 1.01671386,
      "epoch": 0.4832561774785066,
      "flos": 22601206840320.0,
      "grad_norm": 2.2767313220208996,
      "language_loss": 0.79562438,
      "learning_rate": 2.205286325350549e-06,
      "loss": 0.81778717,
      "num_input_tokens_seen": 86489850,
      "step": 4019,
      "time_per_iteration": 2.7376484870910645
    },
    {
      "auxiliary_loss_clip": 0.01193268,
      "auxiliary_loss_mlp": 0.01026298,
      "balance_loss_clip": 1.04949212,
      "balance_loss_mlp": 1.01837015,
      "epoch": 0.4833764203691457,
      "flos": 13437342282240.0,
      "grad_norm": 2.2593844617402685,
      "language_loss": 0.72656655,
      "learning_rate": 2.204511452491603e-06,
      "loss": 0.74876219,
      "num_input_tokens_seen": 86506475,
      "step": 4020,
      "time_per_iteration": 3.61767578125
    },
    {
      "auxiliary_loss_clip": 0.01182299,
      "auxiliary_loss_mlp": 0.01028983,
      "balance_loss_clip": 1.05671465,
      "balance_loss_mlp": 1.02065647,
      "epoch": 0.48349666325978474,
      "flos": 44128036955520.0,
      "grad_norm": 2.092680789992139,
      "language_loss": 0.75151181,
      "learning_rate": 2.2037365486084316e-06,
      "loss": 0.77362466,
      "num_input_tokens_seen": 86529715,
      "step": 4021,
      "time_per_iteration": 2.7834322452545166
    },
    {
      "auxiliary_loss_clip": 0.0119953,
      "auxiliary_loss_mlp": 0.01030718,
      "balance_loss_clip": 1.04809785,
      "balance_loss_mlp": 1.02225423,
      "epoch": 0.48361690615042385,
      "flos": 26028377245440.0,
      "grad_norm": 1.8645371582213373,
      "language_loss": 0.779333,
      "learning_rate": 2.2029616138185886e-06,
      "loss": 0.8016355,
      "num_input_tokens_seen": 86548715,
      "step": 4022,
      "time_per_iteration": 2.7419381141662598
    },
    {
      "auxiliary_loss_clip": 0.01190405,
      "auxiliary_loss_mlp": 0.01030111,
      "balance_loss_clip": 1.05358648,
      "balance_loss_mlp": 1.02199841,
      "epoch": 0.48373714904106296,
      "flos": 22273306560000.0,
      "grad_norm": 1.6519112274577605,
      "language_loss": 0.82716763,
      "learning_rate": 2.202186648239629e-06,
      "loss": 0.84937274,
      "num_input_tokens_seen": 86568650,
      "step": 4023,
      "time_per_iteration": 2.668642997741699
    },
    {
      "auxiliary_loss_clip": 0.01186877,
      "auxiliary_loss_mlp": 0.01026379,
      "balance_loss_clip": 1.05512726,
      "balance_loss_mlp": 1.01840377,
      "epoch": 0.483857391931702,
      "flos": 28292293699200.0,
      "grad_norm": 1.6901306719840723,
      "language_loss": 0.71516877,
      "learning_rate": 2.201411651989117e-06,
      "loss": 0.73730135,
      "num_input_tokens_seen": 86590630,
      "step": 4024,
      "time_per_iteration": 2.6933038234710693
    },
    {
      "auxiliary_loss_clip": 0.01189525,
      "auxiliary_loss_mlp": 0.01208531,
      "balance_loss_clip": 1.05279708,
      "balance_loss_mlp": 1.00058985,
      "epoch": 0.48397763482234113,
      "flos": 27418048577280.0,
      "grad_norm": 1.9968898480607633,
      "language_loss": 0.77946597,
      "learning_rate": 2.2006366251846167e-06,
      "loss": 0.80344653,
      "num_input_tokens_seen": 86611270,
      "step": 4025,
      "time_per_iteration": 2.6311182975769043
    },
    {
      "auxiliary_loss_clip": 0.01186925,
      "auxiliary_loss_mlp": 0.01024987,
      "balance_loss_clip": 1.05254436,
      "balance_loss_mlp": 1.01748919,
      "epoch": 0.48409787771298024,
      "flos": 16797252470400.0,
      "grad_norm": 1.876450150990945,
      "language_loss": 0.75428927,
      "learning_rate": 2.1998615679436997e-06,
      "loss": 0.77640837,
      "num_input_tokens_seen": 86628810,
      "step": 4026,
      "time_per_iteration": 3.5931739807128906
    },
    {
      "auxiliary_loss_clip": 0.0119918,
      "auxiliary_loss_mlp": 0.01028341,
      "balance_loss_clip": 1.05477822,
      "balance_loss_mlp": 1.01904249,
      "epoch": 0.4842181206036193,
      "flos": 25083496028160.0,
      "grad_norm": 2.401900413205711,
      "language_loss": 0.77504069,
      "learning_rate": 2.199086480383942e-06,
      "loss": 0.7973159,
      "num_input_tokens_seen": 86648185,
      "step": 4027,
      "time_per_iteration": 2.713560104370117
    },
    {
      "auxiliary_loss_clip": 0.01202327,
      "auxiliary_loss_mlp": 0.01031463,
      "balance_loss_clip": 1.05431128,
      "balance_loss_mlp": 1.02203357,
      "epoch": 0.4843383634942584,
      "flos": 30372311496960.0,
      "grad_norm": 2.760970941992646,
      "language_loss": 0.67861313,
      "learning_rate": 2.1983113626229234e-06,
      "loss": 0.70095098,
      "num_input_tokens_seen": 86667435,
      "step": 4028,
      "time_per_iteration": 2.7072362899780273
    },
    {
      "auxiliary_loss_clip": 0.01188121,
      "auxiliary_loss_mlp": 0.01208491,
      "balance_loss_clip": 1.05027199,
      "balance_loss_mlp": 1.00066543,
      "epoch": 0.4844586063848975,
      "flos": 20413564917120.0,
      "grad_norm": 1.9579989789676564,
      "language_loss": 0.78532416,
      "learning_rate": 2.1975362147782293e-06,
      "loss": 0.80929029,
      "num_input_tokens_seen": 86686630,
      "step": 4029,
      "time_per_iteration": 2.7125775814056396
    },
    {
      "auxiliary_loss_clip": 0.01097481,
      "auxiliary_loss_mlp": 0.01002397,
      "balance_loss_clip": 1.02820277,
      "balance_loss_mlp": 1.00149083,
      "epoch": 0.48457884927553657,
      "flos": 70303722854400.0,
      "grad_norm": 0.6873475465256618,
      "language_loss": 0.54070389,
      "learning_rate": 2.196761036967448e-06,
      "loss": 0.56170267,
      "num_input_tokens_seen": 86754595,
      "step": 4030,
      "time_per_iteration": 3.360790967941284
    },
    {
      "auxiliary_loss_clip": 0.01182854,
      "auxiliary_loss_mlp": 0.01022618,
      "balance_loss_clip": 1.05221927,
      "balance_loss_mlp": 1.01460671,
      "epoch": 0.4846990921661757,
      "flos": 19934516206080.0,
      "grad_norm": 2.0094706569417866,
      "language_loss": 0.77318394,
      "learning_rate": 2.1959858293081743e-06,
      "loss": 0.79523867,
      "num_input_tokens_seen": 86773730,
      "step": 4031,
      "time_per_iteration": 2.7421014308929443
    },
    {
      "auxiliary_loss_clip": 0.01186945,
      "auxiliary_loss_mlp": 0.01027999,
      "balance_loss_clip": 1.0494349,
      "balance_loss_mlp": 1.01941586,
      "epoch": 0.4848193350568148,
      "flos": 23075945919360.0,
      "grad_norm": 3.455817854709085,
      "language_loss": 0.76269138,
      "learning_rate": 2.1952105919180056e-06,
      "loss": 0.78484082,
      "num_input_tokens_seen": 86792985,
      "step": 4032,
      "time_per_iteration": 2.695338249206543
    },
    {
      "auxiliary_loss_clip": 0.01188043,
      "auxiliary_loss_mlp": 0.01020338,
      "balance_loss_clip": 1.05302954,
      "balance_loss_mlp": 1.01209426,
      "epoch": 0.48493957794745385,
      "flos": 22455481363200.0,
      "grad_norm": 2.6892683787291167,
      "language_loss": 0.68185592,
      "learning_rate": 2.1944353249145456e-06,
      "loss": 0.70393968,
      "num_input_tokens_seen": 86812095,
      "step": 4033,
      "time_per_iteration": 2.6536617279052734
    },
    {
      "auxiliary_loss_clip": 0.01186202,
      "auxiliary_loss_mlp": 0.01026407,
      "balance_loss_clip": 1.05786133,
      "balance_loss_mlp": 1.01843786,
      "epoch": 0.48505982083809296,
      "flos": 25046112948480.0,
      "grad_norm": 2.1045998013949543,
      "language_loss": 0.74579489,
      "learning_rate": 2.193660028415401e-06,
      "loss": 0.76792097,
      "num_input_tokens_seen": 86832875,
      "step": 4034,
      "time_per_iteration": 2.6523890495300293
    },
    {
      "auxiliary_loss_clip": 0.01181898,
      "auxiliary_loss_mlp": 0.01024317,
      "balance_loss_clip": 1.05085564,
      "balance_loss_mlp": 1.01602054,
      "epoch": 0.485180063728732,
      "flos": 26761386090240.0,
      "grad_norm": 1.8100871224162938,
      "language_loss": 0.82000011,
      "learning_rate": 2.1928847025381852e-06,
      "loss": 0.84206223,
      "num_input_tokens_seen": 86853480,
      "step": 4035,
      "time_per_iteration": 2.7232651710510254
    },
    {
      "auxiliary_loss_clip": 0.01185565,
      "auxiliary_loss_mlp": 0.01028621,
      "balance_loss_clip": 1.05150247,
      "balance_loss_mlp": 1.01929879,
      "epoch": 0.4853003066193711,
      "flos": 24059143969920.0,
      "grad_norm": 1.735101118617723,
      "language_loss": 0.83808708,
      "learning_rate": 2.192109347400512e-06,
      "loss": 0.8602289,
      "num_input_tokens_seen": 86873695,
      "step": 4036,
      "time_per_iteration": 2.650473117828369
    },
    {
      "auxiliary_loss_clip": 0.01192082,
      "auxiliary_loss_mlp": 0.01029445,
      "balance_loss_clip": 1.0522244,
      "balance_loss_mlp": 1.02021849,
      "epoch": 0.48542054951001024,
      "flos": 23076376882560.0,
      "grad_norm": 1.7489731714773649,
      "language_loss": 0.79195487,
      "learning_rate": 2.191333963120004e-06,
      "loss": 0.81417012,
      "num_input_tokens_seen": 86892675,
      "step": 4037,
      "time_per_iteration": 2.679033041000366
    },
    {
      "auxiliary_loss_clip": 0.01191835,
      "auxiliary_loss_mlp": 0.01031435,
      "balance_loss_clip": 1.0542177,
      "balance_loss_mlp": 1.02223158,
      "epoch": 0.4855407924006493,
      "flos": 25664889565440.0,
      "grad_norm": 3.388430757880863,
      "language_loss": 0.70129186,
      "learning_rate": 2.190558549814286e-06,
      "loss": 0.72352457,
      "num_input_tokens_seen": 86912835,
      "step": 4038,
      "time_per_iteration": 2.737484931945801
    },
    {
      "auxiliary_loss_clip": 0.01188606,
      "auxiliary_loss_mlp": 0.01027083,
      "balance_loss_clip": 1.05168581,
      "balance_loss_mlp": 1.01879203,
      "epoch": 0.4856610352912884,
      "flos": 23987933256960.0,
      "grad_norm": 1.903238826160772,
      "language_loss": 0.79675007,
      "learning_rate": 2.1897831076009872e-06,
      "loss": 0.81890702,
      "num_input_tokens_seen": 86932475,
      "step": 4039,
      "time_per_iteration": 2.60721492767334
    },
    {
      "auxiliary_loss_clip": 0.01187718,
      "auxiliary_loss_mlp": 0.01025645,
      "balance_loss_clip": 1.0542388,
      "balance_loss_mlp": 1.01730037,
      "epoch": 0.4857812781819275,
      "flos": 24096814358400.0,
      "grad_norm": 1.923365440664652,
      "language_loss": 0.79598451,
      "learning_rate": 2.1890076365977426e-06,
      "loss": 0.8181181,
      "num_input_tokens_seen": 86952300,
      "step": 4040,
      "time_per_iteration": 2.650996446609497
    },
    {
      "auxiliary_loss_clip": 0.0109475,
      "auxiliary_loss_mlp": 0.01001753,
      "balance_loss_clip": 1.02367628,
      "balance_loss_mlp": 1.00097179,
      "epoch": 0.48590152107256657,
      "flos": 56266635185280.0,
      "grad_norm": 0.860411751804818,
      "language_loss": 0.52744633,
      "learning_rate": 2.188232136922189e-06,
      "loss": 0.54841131,
      "num_input_tokens_seen": 87010420,
      "step": 4041,
      "time_per_iteration": 3.1108310222625732
    },
    {
      "auxiliary_loss_clip": 0.01192174,
      "auxiliary_loss_mlp": 0.01026768,
      "balance_loss_clip": 1.04700077,
      "balance_loss_mlp": 1.01785147,
      "epoch": 0.4860217639632057,
      "flos": 20046988667520.0,
      "grad_norm": 2.311854679239663,
      "language_loss": 0.75793165,
      "learning_rate": 2.187456608691971e-06,
      "loss": 0.78012109,
      "num_input_tokens_seen": 87029295,
      "step": 4042,
      "time_per_iteration": 3.7497262954711914
    },
    {
      "auxiliary_loss_clip": 0.01195676,
      "auxiliary_loss_mlp": 0.01027647,
      "balance_loss_clip": 1.05377507,
      "balance_loss_mlp": 1.01892066,
      "epoch": 0.4861420068538448,
      "flos": 17822143232640.0,
      "grad_norm": 2.058293292328662,
      "language_loss": 0.87461275,
      "learning_rate": 2.1866810520247334e-06,
      "loss": 0.896846,
      "num_input_tokens_seen": 87048165,
      "step": 4043,
      "time_per_iteration": 2.6903488636016846
    },
    {
      "auxiliary_loss_clip": 0.01192882,
      "auxiliary_loss_mlp": 0.01025599,
      "balance_loss_clip": 1.05374551,
      "balance_loss_mlp": 1.01647985,
      "epoch": 0.48626224974448384,
      "flos": 26250125857920.0,
      "grad_norm": 1.7866069975197774,
      "language_loss": 0.65215522,
      "learning_rate": 2.185905467038129e-06,
      "loss": 0.67434001,
      "num_input_tokens_seen": 87067070,
      "step": 4044,
      "time_per_iteration": 3.7807412147521973
    },
    {
      "auxiliary_loss_clip": 0.01183085,
      "auxiliary_loss_mlp": 0.01028857,
      "balance_loss_clip": 1.05710196,
      "balance_loss_mlp": 1.02063167,
      "epoch": 0.48638249263512295,
      "flos": 22054502862720.0,
      "grad_norm": 1.8950115677398036,
      "language_loss": 0.77687424,
      "learning_rate": 2.1851298538498127e-06,
      "loss": 0.79899365,
      "num_input_tokens_seen": 87086785,
      "step": 4045,
      "time_per_iteration": 2.636777400970459
    },
    {
      "auxiliary_loss_clip": 0.01195964,
      "auxiliary_loss_mlp": 0.01208914,
      "balance_loss_clip": 1.05746579,
      "balance_loss_mlp": 1.00069988,
      "epoch": 0.48650273552576206,
      "flos": 25119945354240.0,
      "grad_norm": 2.0528140933982764,
      "language_loss": 0.80273807,
      "learning_rate": 2.184354212577446e-06,
      "loss": 0.82678682,
      "num_input_tokens_seen": 87107090,
      "step": 4046,
      "time_per_iteration": 2.661010503768921
    },
    {
      "auxiliary_loss_clip": 0.01188442,
      "auxiliary_loss_mlp": 0.01025817,
      "balance_loss_clip": 1.05663788,
      "balance_loss_mlp": 1.01723385,
      "epoch": 0.4866229784164011,
      "flos": 17456931699840.0,
      "grad_norm": 2.6343048557275393,
      "language_loss": 0.6245122,
      "learning_rate": 2.1835785433386907e-06,
      "loss": 0.64665473,
      "num_input_tokens_seen": 87125905,
      "step": 4047,
      "time_per_iteration": 3.5071451663970947
    },
    {
      "auxiliary_loss_clip": 0.01181495,
      "auxiliary_loss_mlp": 0.01032729,
      "balance_loss_clip": 1.05043328,
      "balance_loss_mlp": 1.02404499,
      "epoch": 0.48674322130704023,
      "flos": 23331127115520.0,
      "grad_norm": 1.9903087083875348,
      "language_loss": 0.65481168,
      "learning_rate": 2.182802846251216e-06,
      "loss": 0.67695391,
      "num_input_tokens_seen": 87146175,
      "step": 4048,
      "time_per_iteration": 2.722832679748535
    },
    {
      "auxiliary_loss_clip": 0.01196779,
      "auxiliary_loss_mlp": 0.01023772,
      "balance_loss_clip": 1.04932761,
      "balance_loss_mlp": 1.01530838,
      "epoch": 0.4868634641976793,
      "flos": 28804344030720.0,
      "grad_norm": 2.3549820754938438,
      "language_loss": 0.72562397,
      "learning_rate": 2.182027121432696e-06,
      "loss": 0.7478295,
      "num_input_tokens_seen": 87166800,
      "step": 4049,
      "time_per_iteration": 2.733386278152466
    },
    {
      "auxiliary_loss_clip": 0.01187495,
      "auxiliary_loss_mlp": 0.01030168,
      "balance_loss_clip": 1.05520689,
      "balance_loss_mlp": 1.0204525,
      "epoch": 0.4869837070883184,
      "flos": 19025976574080.0,
      "grad_norm": 2.0982494863676036,
      "language_loss": 0.82335973,
      "learning_rate": 2.1812513690008054e-06,
      "loss": 0.84553641,
      "num_input_tokens_seen": 87185920,
      "step": 4050,
      "time_per_iteration": 2.673800230026245
    },
    {
      "auxiliary_loss_clip": 0.01194397,
      "auxiliary_loss_mlp": 0.01028199,
      "balance_loss_clip": 1.05571961,
      "balance_loss_mlp": 1.01900172,
      "epoch": 0.4871039499789575,
      "flos": 15121409483520.0,
      "grad_norm": 2.465118416309249,
      "language_loss": 0.79839504,
      "learning_rate": 2.180475589073227e-06,
      "loss": 0.82062101,
      "num_input_tokens_seen": 87203620,
      "step": 4051,
      "time_per_iteration": 2.6523027420043945
    },
    {
      "auxiliary_loss_clip": 0.0117928,
      "auxiliary_loss_mlp": 0.01026785,
      "balance_loss_clip": 1.05303288,
      "balance_loss_mlp": 1.01841068,
      "epoch": 0.48722419286959656,
      "flos": 26174066808960.0,
      "grad_norm": 2.308343876680137,
      "language_loss": 0.73670816,
      "learning_rate": 2.1796997817676456e-06,
      "loss": 0.75876886,
      "num_input_tokens_seen": 87224630,
      "step": 4052,
      "time_per_iteration": 2.7967679500579834
    },
    {
      "auxiliary_loss_clip": 0.01189009,
      "auxiliary_loss_mlp": 0.01208306,
      "balance_loss_clip": 1.05417752,
      "balance_loss_mlp": 1.00076365,
      "epoch": 0.4873444357602357,
      "flos": 24026142349440.0,
      "grad_norm": 1.5231228355833797,
      "language_loss": 0.67121327,
      "learning_rate": 2.1789239472017494e-06,
      "loss": 0.69518638,
      "num_input_tokens_seen": 87246280,
      "step": 4053,
      "time_per_iteration": 3.6016502380371094
    },
    {
      "auxiliary_loss_clip": 0.01191369,
      "auxiliary_loss_mlp": 0.01026262,
      "balance_loss_clip": 1.05101895,
      "balance_loss_mlp": 1.01799464,
      "epoch": 0.4874646786508748,
      "flos": 22820441500800.0,
      "grad_norm": 3.1740499700179297,
      "language_loss": 0.73125279,
      "learning_rate": 2.1781480854932326e-06,
      "loss": 0.75342917,
      "num_input_tokens_seen": 87266045,
      "step": 4054,
      "time_per_iteration": 2.6562294960021973
    },
    {
      "auxiliary_loss_clip": 0.01192505,
      "auxiliary_loss_mlp": 0.01030662,
      "balance_loss_clip": 1.05040812,
      "balance_loss_mlp": 1.02265167,
      "epoch": 0.48758492154151384,
      "flos": 21287594557440.0,
      "grad_norm": 1.7954806308311035,
      "language_loss": 0.79189378,
      "learning_rate": 2.1773721967597933e-06,
      "loss": 0.81412548,
      "num_input_tokens_seen": 87284495,
      "step": 4055,
      "time_per_iteration": 2.722320556640625
    },
    {
      "auxiliary_loss_clip": 0.01088585,
      "auxiliary_loss_mlp": 0.01002435,
      "balance_loss_clip": 1.02092564,
      "balance_loss_mlp": 1.00155246,
      "epoch": 0.48770516443215295,
      "flos": 62244109180800.0,
      "grad_norm": 0.8625950786529273,
      "language_loss": 0.57400137,
      "learning_rate": 2.1765962811191322e-06,
      "loss": 0.59491158,
      "num_input_tokens_seen": 87338960,
      "step": 4056,
      "time_per_iteration": 3.131519079208374
    },
    {
      "auxiliary_loss_clip": 0.01100157,
      "auxiliary_loss_mlp": 0.01002944,
      "balance_loss_clip": 1.01932073,
      "balance_loss_mlp": 1.00201976,
      "epoch": 0.48782540732279206,
      "flos": 66133451882880.0,
      "grad_norm": 0.8495201101483674,
      "language_loss": 0.62008786,
      "learning_rate": 2.1758203386889566e-06,
      "loss": 0.64111888,
      "num_input_tokens_seen": 87401730,
      "step": 4057,
      "time_per_iteration": 3.3307178020477295
    },
    {
      "auxiliary_loss_clip": 0.01192387,
      "auxiliary_loss_mlp": 0.01208836,
      "balance_loss_clip": 1.05102968,
      "balance_loss_mlp": 1.00072265,
      "epoch": 0.4879456502134311,
      "flos": 14607922608000.0,
      "grad_norm": 4.432671172115135,
      "language_loss": 0.84608698,
      "learning_rate": 2.1750443695869746e-06,
      "loss": 0.87009919,
      "num_input_tokens_seen": 87417300,
      "step": 4058,
      "time_per_iteration": 2.746582269668579
    },
    {
      "auxiliary_loss_clip": 0.01190351,
      "auxiliary_loss_mlp": 0.0103144,
      "balance_loss_clip": 1.05503464,
      "balance_loss_mlp": 1.02317834,
      "epoch": 0.4880658931040702,
      "flos": 19500464257920.0,
      "grad_norm": 1.8693586312050716,
      "language_loss": 0.85735542,
      "learning_rate": 2.174268373930901e-06,
      "loss": 0.87957335,
      "num_input_tokens_seen": 87434815,
      "step": 4059,
      "time_per_iteration": 2.6345725059509277
    },
    {
      "auxiliary_loss_clip": 0.0118719,
      "auxiliary_loss_mlp": 0.01209395,
      "balance_loss_clip": 1.05280018,
      "balance_loss_mlp": 1.00072384,
      "epoch": 0.48818613599470934,
      "flos": 16723060928640.0,
      "grad_norm": 2.153859358411318,
      "language_loss": 0.80089879,
      "learning_rate": 2.1734923518384537e-06,
      "loss": 0.82486463,
      "num_input_tokens_seen": 87451420,
      "step": 4060,
      "time_per_iteration": 2.6677026748657227
    },
    {
      "auxiliary_loss_clip": 0.01178277,
      "auxiliary_loss_mlp": 0.01032577,
      "balance_loss_clip": 1.05163693,
      "balance_loss_mlp": 1.02436972,
      "epoch": 0.4883063788853484,
      "flos": 26756932803840.0,
      "grad_norm": 1.9907864768799552,
      "language_loss": 0.82768601,
      "learning_rate": 2.1727163034273547e-06,
      "loss": 0.84979457,
      "num_input_tokens_seen": 87469585,
      "step": 4061,
      "time_per_iteration": 2.6725761890411377
    },
    {
      "auxiliary_loss_clip": 0.01191239,
      "auxiliary_loss_mlp": 0.01030419,
      "balance_loss_clip": 1.05402374,
      "balance_loss_mlp": 1.02147269,
      "epoch": 0.4884266217759875,
      "flos": 16763388923520.0,
      "grad_norm": 2.2356305089182253,
      "language_loss": 0.79252356,
      "learning_rate": 2.17194022881533e-06,
      "loss": 0.81474012,
      "num_input_tokens_seen": 87485675,
      "step": 4062,
      "time_per_iteration": 2.8275794982910156
    },
    {
      "auxiliary_loss_clip": 0.01194796,
      "auxiliary_loss_mlp": 0.01034016,
      "balance_loss_clip": 1.05182385,
      "balance_loss_mlp": 1.02517045,
      "epoch": 0.4885468646666266,
      "flos": 24207132003840.0,
      "grad_norm": 1.9041762368362476,
      "language_loss": 0.67361701,
      "learning_rate": 2.1711641281201092e-06,
      "loss": 0.69590515,
      "num_input_tokens_seen": 87505605,
      "step": 4063,
      "time_per_iteration": 2.6797311305999756
    },
    {
      "auxiliary_loss_clip": 0.01186735,
      "auxiliary_loss_mlp": 0.01027658,
      "balance_loss_clip": 1.05577075,
      "balance_loss_mlp": 1.01934934,
      "epoch": 0.48866710755726567,
      "flos": 14610795696000.0,
      "grad_norm": 2.290254935495438,
      "language_loss": 0.79593897,
      "learning_rate": 2.1703880014594264e-06,
      "loss": 0.81808293,
      "num_input_tokens_seen": 87523195,
      "step": 4064,
      "time_per_iteration": 2.7308764457702637
    },
    {
      "auxiliary_loss_clip": 0.01190124,
      "auxiliary_loss_mlp": 0.01035146,
      "balance_loss_clip": 1.05367112,
      "balance_loss_mlp": 1.02714705,
      "epoch": 0.4887873504479048,
      "flos": 28804451771520.0,
      "grad_norm": 2.22900195705358,
      "language_loss": 0.73489666,
      "learning_rate": 2.1696118489510182e-06,
      "loss": 0.75714934,
      "num_input_tokens_seen": 87544125,
      "step": 4065,
      "time_per_iteration": 2.739065647125244
    },
    {
      "auxiliary_loss_clip": 0.01200809,
      "auxiliary_loss_mlp": 0.01209134,
      "balance_loss_clip": 1.05417812,
      "balance_loss_mlp": 1.00070059,
      "epoch": 0.48890759333854383,
      "flos": 22784387224320.0,
      "grad_norm": 2.073658813198971,
      "language_loss": 0.72593051,
      "learning_rate": 2.1688356707126286e-06,
      "loss": 0.75002992,
      "num_input_tokens_seen": 87563745,
      "step": 4066,
      "time_per_iteration": 2.74369740486145
    },
    {
      "auxiliary_loss_clip": 0.01191223,
      "auxiliary_loss_mlp": 0.01026648,
      "balance_loss_clip": 1.05260849,
      "balance_loss_mlp": 1.01800561,
      "epoch": 0.48902783622918294,
      "flos": 17786088956160.0,
      "grad_norm": 2.485632213977766,
      "language_loss": 0.70148468,
      "learning_rate": 2.168059466862001e-06,
      "loss": 0.72366345,
      "num_input_tokens_seen": 87581895,
      "step": 4067,
      "time_per_iteration": 2.6104025840759277
    },
    {
      "auxiliary_loss_clip": 0.01190973,
      "auxiliary_loss_mlp": 0.01029781,
      "balance_loss_clip": 1.05096745,
      "balance_loss_mlp": 1.0219903,
      "epoch": 0.48914807911982205,
      "flos": 22310294590080.0,
      "grad_norm": 2.0250140278285937,
      "language_loss": 0.82219923,
      "learning_rate": 2.167283237516887e-06,
      "loss": 0.84440684,
      "num_input_tokens_seen": 87600170,
      "step": 4068,
      "time_per_iteration": 2.6818058490753174
    },
    {
      "auxiliary_loss_clip": 0.01193421,
      "auxiliary_loss_mlp": 0.01035396,
      "balance_loss_clip": 1.05221438,
      "balance_loss_mlp": 1.02678275,
      "epoch": 0.4892683220104611,
      "flos": 16363020954240.0,
      "grad_norm": 1.7906457810083027,
      "language_loss": 0.74408138,
      "learning_rate": 2.1665069827950383e-06,
      "loss": 0.76636952,
      "num_input_tokens_seen": 87617455,
      "step": 4069,
      "time_per_iteration": 3.6030170917510986
    },
    {
      "auxiliary_loss_clip": 0.01191302,
      "auxiliary_loss_mlp": 0.01027898,
      "balance_loss_clip": 1.05357921,
      "balance_loss_mlp": 1.01985753,
      "epoch": 0.4893885649011002,
      "flos": 15739144606080.0,
      "grad_norm": 1.832279562200833,
      "language_loss": 0.87138116,
      "learning_rate": 2.1657307028142126e-06,
      "loss": 0.89357316,
      "num_input_tokens_seen": 87634995,
      "step": 4070,
      "time_per_iteration": 3.616490364074707
    },
    {
      "auxiliary_loss_clip": 0.01191238,
      "auxiliary_loss_mlp": 0.01033009,
      "balance_loss_clip": 1.05459523,
      "balance_loss_mlp": 1.02377653,
      "epoch": 0.48950880779173933,
      "flos": 28581984887040.0,
      "grad_norm": 11.216729344673345,
      "language_loss": 0.67183983,
      "learning_rate": 2.164954397692171e-06,
      "loss": 0.69408232,
      "num_input_tokens_seen": 87654420,
      "step": 4071,
      "time_per_iteration": 2.705580949783325
    },
    {
      "auxiliary_loss_clip": 0.01100414,
      "auxiliary_loss_mlp": 0.01001227,
      "balance_loss_clip": 1.02375543,
      "balance_loss_mlp": 1.00029135,
      "epoch": 0.4896290506823784,
      "flos": 66186310746240.0,
      "grad_norm": 1.0789091546458403,
      "language_loss": 0.77282804,
      "learning_rate": 2.164178067546678e-06,
      "loss": 0.79384446,
      "num_input_tokens_seen": 87713585,
      "step": 4072,
      "time_per_iteration": 3.237957239151001
    },
    {
      "auxiliary_loss_clip": 0.01194965,
      "auxiliary_loss_mlp": 0.01029132,
      "balance_loss_clip": 1.05116343,
      "balance_loss_mlp": 1.02033484,
      "epoch": 0.4897492935730175,
      "flos": 12531065207040.0,
      "grad_norm": 2.8124718547133605,
      "language_loss": 0.90985572,
      "learning_rate": 2.163401712495504e-06,
      "loss": 0.93209666,
      "num_input_tokens_seen": 87731280,
      "step": 4073,
      "time_per_iteration": 2.6446049213409424
    },
    {
      "auxiliary_loss_clip": 0.01196654,
      "auxiliary_loss_mlp": 0.01033919,
      "balance_loss_clip": 1.049227,
      "balance_loss_mlp": 1.02508545,
      "epoch": 0.4898695364636566,
      "flos": 23476816679040.0,
      "grad_norm": 1.941050297235517,
      "language_loss": 0.79427779,
      "learning_rate": 2.1626253326564194e-06,
      "loss": 0.81658351,
      "num_input_tokens_seen": 87750230,
      "step": 4074,
      "time_per_iteration": 3.6446807384490967
    },
    {
      "auxiliary_loss_clip": 0.01188099,
      "auxiliary_loss_mlp": 0.01031223,
      "balance_loss_clip": 1.04969907,
      "balance_loss_mlp": 1.02245474,
      "epoch": 0.48998977935429566,
      "flos": 27160209774720.0,
      "grad_norm": 1.8636004904203052,
      "language_loss": 0.77002454,
      "learning_rate": 2.161848928147201e-06,
      "loss": 0.79221773,
      "num_input_tokens_seen": 87770500,
      "step": 4075,
      "time_per_iteration": 2.6730406284332275
    },
    {
      "auxiliary_loss_clip": 0.0118994,
      "auxiliary_loss_mlp": 0.01023558,
      "balance_loss_clip": 1.05633187,
      "balance_loss_mlp": 1.01489782,
      "epoch": 0.4901100222449348,
      "flos": 20339588856960.0,
      "grad_norm": 2.123636227969882,
      "language_loss": 0.80961013,
      "learning_rate": 2.161072499085629e-06,
      "loss": 0.83174515,
      "num_input_tokens_seen": 87789495,
      "step": 4076,
      "time_per_iteration": 2.6811530590057373
    },
    {
      "auxiliary_loss_clip": 0.01199413,
      "auxiliary_loss_mlp": 0.01026099,
      "balance_loss_clip": 1.05352616,
      "balance_loss_mlp": 1.01802826,
      "epoch": 0.4902302651355739,
      "flos": 30446359384320.0,
      "grad_norm": 1.5431959110146518,
      "language_loss": 0.83498228,
      "learning_rate": 2.160296045589487e-06,
      "loss": 0.85723746,
      "num_input_tokens_seen": 87812955,
      "step": 4077,
      "time_per_iteration": 2.7111759185791016
    },
    {
      "auxiliary_loss_clip": 0.01186563,
      "auxiliary_loss_mlp": 0.01023323,
      "balance_loss_clip": 1.05482328,
      "balance_loss_mlp": 1.01462674,
      "epoch": 0.49035050802621294,
      "flos": 19174180089600.0,
      "grad_norm": 1.9181670585924266,
      "language_loss": 0.69850188,
      "learning_rate": 2.159519567776562e-06,
      "loss": 0.72060072,
      "num_input_tokens_seen": 87832605,
      "step": 4078,
      "time_per_iteration": 3.6188557147979736
    },
    {
      "auxiliary_loss_clip": 0.011994,
      "auxiliary_loss_mlp": 0.01027815,
      "balance_loss_clip": 1.04700947,
      "balance_loss_mlp": 1.01921439,
      "epoch": 0.49047075091685205,
      "flos": 22228489365120.0,
      "grad_norm": 3.5800738184130387,
      "language_loss": 0.70942193,
      "learning_rate": 2.1587430657646463e-06,
      "loss": 0.73169404,
      "num_input_tokens_seen": 87846040,
      "step": 4079,
      "time_per_iteration": 2.6656649112701416
    },
    {
      "auxiliary_loss_clip": 0.01189791,
      "auxiliary_loss_mlp": 0.01022349,
      "balance_loss_clip": 1.05327559,
      "balance_loss_mlp": 1.01402235,
      "epoch": 0.4905909938074911,
      "flos": 20156516213760.0,
      "grad_norm": 1.996104420941468,
      "language_loss": 0.78170776,
      "learning_rate": 2.157966539671533e-06,
      "loss": 0.80382919,
      "num_input_tokens_seen": 87865680,
      "step": 4080,
      "time_per_iteration": 2.583996534347534
    },
    {
      "auxiliary_loss_clip": 0.01194183,
      "auxiliary_loss_mlp": 0.01026827,
      "balance_loss_clip": 1.05142176,
      "balance_loss_mlp": 1.01884246,
      "epoch": 0.4907112366981302,
      "flos": 17202217380480.0,
      "grad_norm": 2.3531395927775436,
      "language_loss": 0.67442393,
      "learning_rate": 2.157189989615021e-06,
      "loss": 0.69663405,
      "num_input_tokens_seen": 87884270,
      "step": 4081,
      "time_per_iteration": 2.712409496307373
    },
    {
      "auxiliary_loss_clip": 0.0118874,
      "auxiliary_loss_mlp": 0.01208856,
      "balance_loss_clip": 1.05186319,
      "balance_loss_mlp": 1.00062919,
      "epoch": 0.4908314795887693,
      "flos": 21688968107520.0,
      "grad_norm": 2.0771136056526993,
      "language_loss": 0.74798906,
      "learning_rate": 2.156413415712913e-06,
      "loss": 0.77196497,
      "num_input_tokens_seen": 87906320,
      "step": 4082,
      "time_per_iteration": 2.6805508136749268
    },
    {
      "auxiliary_loss_clip": 0.01197732,
      "auxiliary_loss_mlp": 0.01209304,
      "balance_loss_clip": 1.05509198,
      "balance_loss_mlp": 1.0007484,
      "epoch": 0.4909517224794084,
      "flos": 26213676531840.0,
      "grad_norm": 1.672255760292518,
      "language_loss": 0.78822553,
      "learning_rate": 2.155636818083014e-06,
      "loss": 0.81229585,
      "num_input_tokens_seen": 87927690,
      "step": 4083,
      "time_per_iteration": 2.687678337097168
    },
    {
      "auxiliary_loss_clip": 0.01188362,
      "auxiliary_loss_mlp": 0.01024477,
      "balance_loss_clip": 1.05401063,
      "balance_loss_mlp": 1.01695442,
      "epoch": 0.4910719653700475,
      "flos": 23148377694720.0,
      "grad_norm": 5.602409610084566,
      "language_loss": 0.84173948,
      "learning_rate": 2.154860196843134e-06,
      "loss": 0.86386788,
      "num_input_tokens_seen": 87946885,
      "step": 4084,
      "time_per_iteration": 2.6999549865722656
    },
    {
      "auxiliary_loss_clip": 0.01186795,
      "auxiliary_loss_mlp": 0.01029997,
      "balance_loss_clip": 1.05583286,
      "balance_loss_mlp": 1.02180099,
      "epoch": 0.4911922082606866,
      "flos": 23331845387520.0,
      "grad_norm": 1.7206393151992418,
      "language_loss": 0.76901495,
      "learning_rate": 2.154083552111085e-06,
      "loss": 0.79118288,
      "num_input_tokens_seen": 87966055,
      "step": 4085,
      "time_per_iteration": 2.6246423721313477
    },
    {
      "auxiliary_loss_clip": 0.01187591,
      "auxiliary_loss_mlp": 0.01024757,
      "balance_loss_clip": 1.05462432,
      "balance_loss_mlp": 1.01639485,
      "epoch": 0.49131245115132566,
      "flos": 29203239542400.0,
      "grad_norm": 1.691045057729974,
      "language_loss": 0.82112378,
      "learning_rate": 2.1533068840046834e-06,
      "loss": 0.84324729,
      "num_input_tokens_seen": 87986320,
      "step": 4086,
      "time_per_iteration": 2.661287546157837
    },
    {
      "auxiliary_loss_clip": 0.0118327,
      "auxiliary_loss_mlp": 0.01209033,
      "balance_loss_clip": 1.04989505,
      "balance_loss_mlp": 1.00063848,
      "epoch": 0.49143269404196477,
      "flos": 20147465986560.0,
      "grad_norm": 2.3252548133515765,
      "language_loss": 0.6175431,
      "learning_rate": 2.152530192641749e-06,
      "loss": 0.64146614,
      "num_input_tokens_seen": 88001230,
      "step": 4087,
      "time_per_iteration": 2.677154779434204
    },
    {
      "auxiliary_loss_clip": 0.01191513,
      "auxiliary_loss_mlp": 0.01030325,
      "balance_loss_clip": 1.05425835,
      "balance_loss_mlp": 1.02186131,
      "epoch": 0.4915529369326039,
      "flos": 24389809597440.0,
      "grad_norm": 1.7786045723328596,
      "language_loss": 0.72559649,
      "learning_rate": 2.1517534781401068e-06,
      "loss": 0.74781489,
      "num_input_tokens_seen": 88019110,
      "step": 4088,
      "time_per_iteration": 2.665675401687622
    },
    {
      "auxiliary_loss_clip": 0.0118696,
      "auxiliary_loss_mlp": 0.01022942,
      "balance_loss_clip": 1.05374205,
      "balance_loss_mlp": 1.01441216,
      "epoch": 0.49167317982324293,
      "flos": 10524305197440.0,
      "grad_norm": 5.151392147913704,
      "language_loss": 0.69845867,
      "learning_rate": 2.150976740617581e-06,
      "loss": 0.72055769,
      "num_input_tokens_seen": 88035670,
      "step": 4089,
      "time_per_iteration": 2.5944690704345703
    },
    {
      "auxiliary_loss_clip": 0.01196878,
      "auxiliary_loss_mlp": 0.01029289,
      "balance_loss_clip": 1.05488932,
      "balance_loss_mlp": 1.0207895,
      "epoch": 0.49179342271388204,
      "flos": 25593427457280.0,
      "grad_norm": 1.8341544481501044,
      "language_loss": 0.71464485,
      "learning_rate": 2.150199980192006e-06,
      "loss": 0.73690653,
      "num_input_tokens_seen": 88054790,
      "step": 4090,
      "time_per_iteration": 2.6692118644714355
    },
    {
      "auxiliary_loss_clip": 0.01182379,
      "auxiliary_loss_mlp": 0.01025095,
      "balance_loss_clip": 1.0504421,
      "balance_loss_mlp": 1.01691723,
      "epoch": 0.49191366560452116,
      "flos": 21102043875840.0,
      "grad_norm": 4.228696292248948,
      "language_loss": 0.81071621,
      "learning_rate": 2.1494231969812114e-06,
      "loss": 0.83279091,
      "num_input_tokens_seen": 88073780,
      "step": 4091,
      "time_per_iteration": 2.6339080333709717
    },
    {
      "auxiliary_loss_clip": 0.01199318,
      "auxiliary_loss_mlp": 0.0102654,
      "balance_loss_clip": 1.05468678,
      "balance_loss_mlp": 1.01763558,
      "epoch": 0.4920339084951602,
      "flos": 26067520091520.0,
      "grad_norm": 2.3352948383974357,
      "language_loss": 0.81483006,
      "learning_rate": 2.1486463911030372e-06,
      "loss": 0.8370887,
      "num_input_tokens_seen": 88094430,
      "step": 4092,
      "time_per_iteration": 2.7408392429351807
    },
    {
      "auxiliary_loss_clip": 0.01188482,
      "auxiliary_loss_mlp": 0.01028849,
      "balance_loss_clip": 1.05089021,
      "balance_loss_mlp": 1.02045667,
      "epoch": 0.4921541513857993,
      "flos": 25081269384960.0,
      "grad_norm": 2.307527980503155,
      "language_loss": 0.74672383,
      "learning_rate": 2.147869562675324e-06,
      "loss": 0.76889718,
      "num_input_tokens_seen": 88113400,
      "step": 4093,
      "time_per_iteration": 2.648465633392334
    },
    {
      "auxiliary_loss_clip": 0.01189565,
      "auxiliary_loss_mlp": 0.01026809,
      "balance_loss_clip": 1.05599344,
      "balance_loss_mlp": 1.01832712,
      "epoch": 0.49227439427643843,
      "flos": 24389809597440.0,
      "grad_norm": 1.8779119689876835,
      "language_loss": 0.72520399,
      "learning_rate": 2.147092711815915e-06,
      "loss": 0.74736774,
      "num_input_tokens_seen": 88132750,
      "step": 4094,
      "time_per_iteration": 2.6528117656707764
    },
    {
      "auxiliary_loss_clip": 0.01189763,
      "auxiliary_loss_mlp": 0.0102477,
      "balance_loss_clip": 1.05168211,
      "balance_loss_mlp": 1.01708055,
      "epoch": 0.4923946371670775,
      "flos": 11363753018880.0,
      "grad_norm": 11.275439484768825,
      "language_loss": 0.86672044,
      "learning_rate": 2.1463158386426593e-06,
      "loss": 0.88886571,
      "num_input_tokens_seen": 88150560,
      "step": 4095,
      "time_per_iteration": 2.645688772201538
    },
    {
      "auxiliary_loss_clip": 0.01198133,
      "auxiliary_loss_mlp": 0.01029107,
      "balance_loss_clip": 1.05461669,
      "balance_loss_mlp": 1.02033949,
      "epoch": 0.4925148800577166,
      "flos": 30445964334720.0,
      "grad_norm": 2.0609421230657725,
      "language_loss": 0.77409101,
      "learning_rate": 2.145538943273407e-06,
      "loss": 0.79636347,
      "num_input_tokens_seen": 88170835,
      "step": 4096,
      "time_per_iteration": 3.6663362979888916
    },
    {
      "auxiliary_loss_clip": 0.01186955,
      "auxiliary_loss_mlp": 0.0102877,
      "balance_loss_clip": 1.05734682,
      "balance_loss_mlp": 1.02058041,
      "epoch": 0.49263512294835565,
      "flos": 20850454039680.0,
      "grad_norm": 2.3937053369378423,
      "language_loss": 0.72016376,
      "learning_rate": 2.144762025826013e-06,
      "loss": 0.74232101,
      "num_input_tokens_seen": 88189925,
      "step": 4097,
      "time_per_iteration": 3.6190054416656494
    },
    {
      "auxiliary_loss_clip": 0.01192649,
      "auxiliary_loss_mlp": 0.01031006,
      "balance_loss_clip": 1.05477643,
      "balance_loss_mlp": 1.02258396,
      "epoch": 0.49275536583899476,
      "flos": 23767477534080.0,
      "grad_norm": 2.472116341160935,
      "language_loss": 0.87180865,
      "learning_rate": 2.143985086418334e-06,
      "loss": 0.89404517,
      "num_input_tokens_seen": 88205105,
      "step": 4098,
      "time_per_iteration": 2.673018455505371
    },
    {
      "auxiliary_loss_clip": 0.01192848,
      "auxiliary_loss_mlp": 0.01024217,
      "balance_loss_clip": 1.05349827,
      "balance_loss_mlp": 1.01623023,
      "epoch": 0.4928756087296339,
      "flos": 22273522041600.0,
      "grad_norm": 1.3569656705438609,
      "language_loss": 0.76807535,
      "learning_rate": 2.1432081251682324e-06,
      "loss": 0.79024601,
      "num_input_tokens_seen": 88225475,
      "step": 4099,
      "time_per_iteration": 2.6772983074188232
    },
    {
      "auxiliary_loss_clip": 0.01189475,
      "auxiliary_loss_mlp": 0.01025342,
      "balance_loss_clip": 1.05922878,
      "balance_loss_mlp": 1.01653278,
      "epoch": 0.49299585162027293,
      "flos": 19645471463040.0,
      "grad_norm": 1.9630618475748636,
      "language_loss": 0.8694483,
      "learning_rate": 2.142431142193572e-06,
      "loss": 0.89159644,
      "num_input_tokens_seen": 88243255,
      "step": 4100,
      "time_per_iteration": 3.6031839847564697
    },
    {
      "auxiliary_loss_clip": 0.01185724,
      "auxiliary_loss_mlp": 0.01031292,
      "balance_loss_clip": 1.05764115,
      "balance_loss_mlp": 1.02287555,
      "epoch": 0.49311609451091204,
      "flos": 38837138497920.0,
      "grad_norm": 19.825389305840933,
      "language_loss": 0.71726966,
      "learning_rate": 2.1416541376122207e-06,
      "loss": 0.73943979,
      "num_input_tokens_seen": 88263435,
      "step": 4101,
      "time_per_iteration": 2.7202160358428955
    },
    {
      "auxiliary_loss_clip": 0.01185014,
      "auxiliary_loss_mlp": 0.01027771,
      "balance_loss_clip": 1.05450463,
      "balance_loss_mlp": 1.01850879,
      "epoch": 0.49323633740155115,
      "flos": 28329102161280.0,
      "grad_norm": 1.8015304537379566,
      "language_loss": 0.7332983,
      "learning_rate": 2.1408771115420496e-06,
      "loss": 0.75542617,
      "num_input_tokens_seen": 88283295,
      "step": 4102,
      "time_per_iteration": 2.684858798980713
    },
    {
      "auxiliary_loss_clip": 0.01196451,
      "auxiliary_loss_mlp": 0.01027865,
      "balance_loss_clip": 1.05378687,
      "balance_loss_mlp": 1.02009249,
      "epoch": 0.4933565802921902,
      "flos": 21135584200320.0,
      "grad_norm": 1.7218263043216773,
      "language_loss": 0.64764732,
      "learning_rate": 2.140100064100932e-06,
      "loss": 0.66989052,
      "num_input_tokens_seen": 88299270,
      "step": 4103,
      "time_per_iteration": 2.679729700088501
    },
    {
      "auxiliary_loss_clip": 0.0118621,
      "auxiliary_loss_mlp": 0.01020689,
      "balance_loss_clip": 1.05489862,
      "balance_loss_mlp": 1.01276147,
      "epoch": 0.4934768231828293,
      "flos": 18039007595520.0,
      "grad_norm": 1.9254031724607583,
      "language_loss": 0.76089853,
      "learning_rate": 2.139322995406746e-06,
      "loss": 0.78296751,
      "num_input_tokens_seen": 88316905,
      "step": 4104,
      "time_per_iteration": 2.595397472381592
    },
    {
      "auxiliary_loss_clip": 0.01188136,
      "auxiliary_loss_mlp": 0.01028046,
      "balance_loss_clip": 1.05857456,
      "balance_loss_mlp": 1.01930201,
      "epoch": 0.4935970660734684,
      "flos": 23469957181440.0,
      "grad_norm": 2.289074658043692,
      "language_loss": 0.79808551,
      "learning_rate": 2.1385459055773727e-06,
      "loss": 0.82024735,
      "num_input_tokens_seen": 88335095,
      "step": 4105,
      "time_per_iteration": 3.5528454780578613
    },
    {
      "auxiliary_loss_clip": 0.01179443,
      "auxiliary_loss_mlp": 0.0120852,
      "balance_loss_clip": 1.0452888,
      "balance_loss_mlp": 1.00068986,
      "epoch": 0.4937173089641075,
      "flos": 64479258840960.0,
      "grad_norm": 2.2696204916325082,
      "language_loss": 0.73475885,
      "learning_rate": 2.137768794730696e-06,
      "loss": 0.75863844,
      "num_input_tokens_seen": 88358545,
      "step": 4106,
      "time_per_iteration": 3.1170241832733154
    },
    {
      "auxiliary_loss_clip": 0.01194145,
      "auxiliary_loss_mlp": 0.01031553,
      "balance_loss_clip": 1.05367017,
      "balance_loss_mlp": 1.02296972,
      "epoch": 0.4938375518547466,
      "flos": 22346025644160.0,
      "grad_norm": 2.128634482079776,
      "language_loss": 0.80515122,
      "learning_rate": 2.1369916629846026e-06,
      "loss": 0.82740819,
      "num_input_tokens_seen": 88378295,
      "step": 4107,
      "time_per_iteration": 2.670071601867676
    },
    {
      "auxiliary_loss_clip": 0.01189959,
      "auxiliary_loss_mlp": 0.01022383,
      "balance_loss_clip": 1.05092931,
      "balance_loss_mlp": 1.01419306,
      "epoch": 0.4939577947453857,
      "flos": 17858700299520.0,
      "grad_norm": 2.1927089721719972,
      "language_loss": 0.75309306,
      "learning_rate": 2.136214510456983e-06,
      "loss": 0.77521646,
      "num_input_tokens_seen": 88396750,
      "step": 4108,
      "time_per_iteration": 2.709505081176758
    },
    {
      "auxiliary_loss_clip": 0.01114847,
      "auxiliary_loss_mlp": 0.01199191,
      "balance_loss_clip": 1.02267838,
      "balance_loss_mlp": 0.99989569,
      "epoch": 0.49407803763602476,
      "flos": 70066746875520.0,
      "grad_norm": 0.8898343758877418,
      "language_loss": 0.6307857,
      "learning_rate": 2.1354373372657296e-06,
      "loss": 0.65392601,
      "num_input_tokens_seen": 88455190,
      "step": 4109,
      "time_per_iteration": 3.2641894817352295
    },
    {
      "auxiliary_loss_clip": 0.01185133,
      "auxiliary_loss_mlp": 0.01029505,
      "balance_loss_clip": 1.05727136,
      "balance_loss_mlp": 1.0218693,
      "epoch": 0.49419828052666387,
      "flos": 24317485562880.0,
      "grad_norm": 6.554552674602021,
      "language_loss": 0.70912379,
      "learning_rate": 2.1346601435287404e-06,
      "loss": 0.73127013,
      "num_input_tokens_seen": 88477460,
      "step": 4110,
      "time_per_iteration": 2.6403768062591553
    },
    {
      "auxiliary_loss_clip": 0.01187041,
      "auxiliary_loss_mlp": 0.01024774,
      "balance_loss_clip": 1.04918838,
      "balance_loss_mlp": 1.01673317,
      "epoch": 0.494318523417303,
      "flos": 29386060790400.0,
      "grad_norm": 5.625034189970791,
      "language_loss": 0.80342567,
      "learning_rate": 2.1338829293639144e-06,
      "loss": 0.82554382,
      "num_input_tokens_seen": 88497820,
      "step": 4111,
      "time_per_iteration": 2.711557388305664
    },
    {
      "auxiliary_loss_clip": 0.01193961,
      "auxiliary_loss_mlp": 0.01030076,
      "balance_loss_clip": 1.05068612,
      "balance_loss_mlp": 1.02147555,
      "epoch": 0.49443876630794203,
      "flos": 15268284195840.0,
      "grad_norm": 2.074914548988524,
      "language_loss": 0.82841742,
      "learning_rate": 2.1331056948891547e-06,
      "loss": 0.85065782,
      "num_input_tokens_seen": 88514920,
      "step": 4112,
      "time_per_iteration": 2.711179256439209
    },
    {
      "auxiliary_loss_clip": 0.01183678,
      "auxiliary_loss_mlp": 0.01024304,
      "balance_loss_clip": 1.05026031,
      "balance_loss_mlp": 1.01586962,
      "epoch": 0.49455900919858115,
      "flos": 12347453859840.0,
      "grad_norm": 2.146362057137003,
      "language_loss": 0.76792622,
      "learning_rate": 2.1323284402223666e-06,
      "loss": 0.7900061,
      "num_input_tokens_seen": 88530910,
      "step": 4113,
      "time_per_iteration": 2.6349093914031982
    },
    {
      "auxiliary_loss_clip": 0.0118482,
      "auxiliary_loss_mlp": 0.01207321,
      "balance_loss_clip": 1.0599339,
      "balance_loss_mlp": 1.00056601,
      "epoch": 0.4946792520892202,
      "flos": 22779610715520.0,
      "grad_norm": 1.9505502103752872,
      "language_loss": 0.88506079,
      "learning_rate": 2.1315511654814597e-06,
      "loss": 0.90898216,
      "num_input_tokens_seen": 88549320,
      "step": 4114,
      "time_per_iteration": 2.5976016521453857
    },
    {
      "auxiliary_loss_clip": 0.01185944,
      "auxiliary_loss_mlp": 0.01023279,
      "balance_loss_clip": 1.0560174,
      "balance_loss_mlp": 1.01587296,
      "epoch": 0.4947994949798593,
      "flos": 23148126299520.0,
      "grad_norm": 2.4498595418440137,
      "language_loss": 0.78291255,
      "learning_rate": 2.1307738707843456e-06,
      "loss": 0.80500472,
      "num_input_tokens_seen": 88568985,
      "step": 4115,
      "time_per_iteration": 2.6735527515411377
    },
    {
      "auxiliary_loss_clip": 0.01193985,
      "auxiliary_loss_mlp": 0.01025899,
      "balance_loss_clip": 1.05699396,
      "balance_loss_mlp": 1.0179832,
      "epoch": 0.4949197378704984,
      "flos": 23659997063040.0,
      "grad_norm": 1.8970496322347912,
      "language_loss": 0.68984443,
      "learning_rate": 2.1299965562489385e-06,
      "loss": 0.71204323,
      "num_input_tokens_seen": 88588790,
      "step": 4116,
      "time_per_iteration": 2.6430630683898926
    },
    {
      "auxiliary_loss_clip": 0.01183419,
      "auxiliary_loss_mlp": 0.01027927,
      "balance_loss_clip": 1.0510323,
      "balance_loss_mlp": 1.01947463,
      "epoch": 0.4950399807611375,
      "flos": 26911493026560.0,
      "grad_norm": 1.412440975294247,
      "language_loss": 0.78810924,
      "learning_rate": 2.129219221993158e-06,
      "loss": 0.81022263,
      "num_input_tokens_seen": 88613575,
      "step": 4117,
      "time_per_iteration": 2.7486026287078857
    },
    {
      "auxiliary_loss_clip": 0.01095878,
      "auxiliary_loss_mlp": 0.01003435,
      "balance_loss_clip": 1.0180974,
      "balance_loss_mlp": 1.00260627,
      "epoch": 0.4951602236517766,
      "flos": 67315270187520.0,
      "grad_norm": 0.7873186863496977,
      "language_loss": 0.59918678,
      "learning_rate": 2.128441868134924e-06,
      "loss": 0.62017989,
      "num_input_tokens_seen": 88675510,
      "step": 4118,
      "time_per_iteration": 3.246103525161743
    },
    {
      "auxiliary_loss_clip": 0.01195382,
      "auxiliary_loss_mlp": 0.01025428,
      "balance_loss_clip": 1.0509069,
      "balance_loss_mlp": 1.01761365,
      "epoch": 0.4952804665424157,
      "flos": 19901442758400.0,
      "grad_norm": 2.342626070841878,
      "language_loss": 0.82766724,
      "learning_rate": 2.1276644947921606e-06,
      "loss": 0.84987539,
      "num_input_tokens_seen": 88694425,
      "step": 4119,
      "time_per_iteration": 2.6364212036132812
    },
    {
      "auxiliary_loss_clip": 0.01187599,
      "auxiliary_loss_mlp": 0.01027035,
      "balance_loss_clip": 1.05382705,
      "balance_loss_mlp": 1.01823103,
      "epoch": 0.49540070943305475,
      "flos": 18806813740800.0,
      "grad_norm": 3.0131487747605825,
      "language_loss": 0.82381207,
      "learning_rate": 2.126887102082795e-06,
      "loss": 0.84595841,
      "num_input_tokens_seen": 88714450,
      "step": 4120,
      "time_per_iteration": 2.718700647354126
    },
    {
      "auxiliary_loss_clip": 0.01189757,
      "auxiliary_loss_mlp": 0.01028597,
      "balance_loss_clip": 1.04680085,
      "balance_loss_mlp": 1.02070534,
      "epoch": 0.49552095232369386,
      "flos": 24934179191040.0,
      "grad_norm": 1.7498447168783335,
      "language_loss": 0.70645678,
      "learning_rate": 2.126109690124757e-06,
      "loss": 0.72864032,
      "num_input_tokens_seen": 88735265,
      "step": 4121,
      "time_per_iteration": 2.7146079540252686
    },
    {
      "auxiliary_loss_clip": 0.01197535,
      "auxiliary_loss_mlp": 0.01023186,
      "balance_loss_clip": 1.04932749,
      "balance_loss_mlp": 1.01541328,
      "epoch": 0.495641195214333,
      "flos": 22857249962880.0,
      "grad_norm": 1.8149105482533678,
      "language_loss": 0.71484065,
      "learning_rate": 2.1253322590359786e-06,
      "loss": 0.73704785,
      "num_input_tokens_seen": 88754600,
      "step": 4122,
      "time_per_iteration": 3.6560330390930176
    },
    {
      "auxiliary_loss_clip": 0.01183893,
      "auxiliary_loss_mlp": 0.01026417,
      "balance_loss_clip": 1.05218244,
      "balance_loss_mlp": 1.01863265,
      "epoch": 0.49576143810497203,
      "flos": 25769748343680.0,
      "grad_norm": 1.8762217634788843,
      "language_loss": 0.74358642,
      "learning_rate": 2.124554808934397e-06,
      "loss": 0.76568949,
      "num_input_tokens_seen": 88775180,
      "step": 4123,
      "time_per_iteration": 3.5541701316833496
    },
    {
      "auxiliary_loss_clip": 0.01187322,
      "auxiliary_loss_mlp": 0.01030097,
      "balance_loss_clip": 1.04634285,
      "balance_loss_mlp": 1.02150846,
      "epoch": 0.49588168099561114,
      "flos": 22128838058880.0,
      "grad_norm": 1.9071196630784144,
      "language_loss": 0.73088324,
      "learning_rate": 2.1237773399379496e-06,
      "loss": 0.75305742,
      "num_input_tokens_seen": 88796145,
      "step": 4124,
      "time_per_iteration": 2.716820001602173
    },
    {
      "auxiliary_loss_clip": 0.01192812,
      "auxiliary_loss_mlp": 0.01027095,
      "balance_loss_clip": 1.04722881,
      "balance_loss_mlp": 1.01819038,
      "epoch": 0.49600192388625025,
      "flos": 24387331559040.0,
      "grad_norm": 1.7981951371525937,
      "language_loss": 0.87115371,
      "learning_rate": 2.122999852164578e-06,
      "loss": 0.89335281,
      "num_input_tokens_seen": 88816765,
      "step": 4125,
      "time_per_iteration": 2.8027989864349365
    },
    {
      "auxiliary_loss_clip": 0.01193942,
      "auxiliary_loss_mlp": 0.01022783,
      "balance_loss_clip": 1.05052245,
      "balance_loss_mlp": 1.01461124,
      "epoch": 0.4961221667768893,
      "flos": 22857429530880.0,
      "grad_norm": 2.7960165769406142,
      "language_loss": 0.58416915,
      "learning_rate": 2.122222345732227e-06,
      "loss": 0.60633636,
      "num_input_tokens_seen": 88836680,
      "step": 4126,
      "time_per_iteration": 2.6940696239471436
    },
    {
      "auxiliary_loss_clip": 0.01192864,
      "auxiliary_loss_mlp": 0.01027194,
      "balance_loss_clip": 1.04942393,
      "balance_loss_mlp": 1.01878357,
      "epoch": 0.4962424096675284,
      "flos": 17858089768320.0,
      "grad_norm": 2.6559378918907752,
      "language_loss": 0.83298296,
      "learning_rate": 2.121444820758843e-06,
      "loss": 0.85518354,
      "num_input_tokens_seen": 88855320,
      "step": 4127,
      "time_per_iteration": 3.6460793018341064
    },
    {
      "auxiliary_loss_clip": 0.01190634,
      "auxiliary_loss_mlp": 0.01035677,
      "balance_loss_clip": 1.04891026,
      "balance_loss_mlp": 1.0267179,
      "epoch": 0.49636265255816747,
      "flos": 21793611404160.0,
      "grad_norm": 2.347460672119671,
      "language_loss": 0.78867304,
      "learning_rate": 2.120667277362376e-06,
      "loss": 0.81093615,
      "num_input_tokens_seen": 88874035,
      "step": 4128,
      "time_per_iteration": 2.738762378692627
    },
    {
      "auxiliary_loss_clip": 0.01188813,
      "auxiliary_loss_mlp": 0.01035745,
      "balance_loss_clip": 1.0585084,
      "balance_loss_mlp": 1.02707291,
      "epoch": 0.4964828954488066,
      "flos": 16358603581440.0,
      "grad_norm": 3.6863902737050354,
      "language_loss": 0.84767294,
      "learning_rate": 2.1198897156607796e-06,
      "loss": 0.86991853,
      "num_input_tokens_seen": 88891390,
      "step": 4129,
      "time_per_iteration": 2.5283052921295166
    },
    {
      "auxiliary_loss_clip": 0.0119308,
      "auxiliary_loss_mlp": 0.01027507,
      "balance_loss_clip": 1.05460942,
      "balance_loss_mlp": 1.01912034,
      "epoch": 0.4966031383394457,
      "flos": 24711101775360.0,
      "grad_norm": 10.459946944778293,
      "language_loss": 0.74212098,
      "learning_rate": 2.1191121357720085e-06,
      "loss": 0.76432681,
      "num_input_tokens_seen": 88909450,
      "step": 4130,
      "time_per_iteration": 2.6374382972717285
    },
    {
      "auxiliary_loss_clip": 0.01186353,
      "auxiliary_loss_mlp": 0.01026157,
      "balance_loss_clip": 1.04853511,
      "balance_loss_mlp": 1.01754379,
      "epoch": 0.49672338123008475,
      "flos": 22930615491840.0,
      "grad_norm": 2.0121389571472146,
      "language_loss": 0.74585909,
      "learning_rate": 2.1183345378140206e-06,
      "loss": 0.76798415,
      "num_input_tokens_seen": 88929195,
      "step": 4131,
      "time_per_iteration": 2.6800732612609863
    },
    {
      "auxiliary_loss_clip": 0.01093268,
      "auxiliary_loss_mlp": 0.01006382,
      "balance_loss_clip": 1.02259398,
      "balance_loss_mlp": 1.00548232,
      "epoch": 0.49684362412072386,
      "flos": 65976736844160.0,
      "grad_norm": 0.8745428683695422,
      "language_loss": 0.61982846,
      "learning_rate": 2.1175569219047783e-06,
      "loss": 0.64082491,
      "num_input_tokens_seen": 88990635,
      "step": 4132,
      "time_per_iteration": 4.157306671142578
    },
    {
      "auxiliary_loss_clip": 0.01184797,
      "auxiliary_loss_mlp": 0.01029511,
      "balance_loss_clip": 1.0557878,
      "balance_loss_mlp": 1.0218581,
      "epoch": 0.49696386701136297,
      "flos": 19971288754560.0,
      "grad_norm": 1.9788164752724806,
      "language_loss": 0.73296905,
      "learning_rate": 2.1167792881622437e-06,
      "loss": 0.75511205,
      "num_input_tokens_seen": 89009655,
      "step": 4133,
      "time_per_iteration": 2.641397714614868
    },
    {
      "auxiliary_loss_clip": 0.01184242,
      "auxiliary_loss_mlp": 0.01028738,
      "balance_loss_clip": 1.05211496,
      "balance_loss_mlp": 1.02124524,
      "epoch": 0.497084109902002,
      "flos": 24750819239040.0,
      "grad_norm": 1.6865308342399903,
      "language_loss": 0.80839014,
      "learning_rate": 2.116001636704384e-06,
      "loss": 0.83051991,
      "num_input_tokens_seen": 89030040,
      "step": 4134,
      "time_per_iteration": 2.7280609607696533
    },
    {
      "auxiliary_loss_clip": 0.01205396,
      "auxiliary_loss_mlp": 0.01028584,
      "balance_loss_clip": 1.05082893,
      "balance_loss_mlp": 1.02030742,
      "epoch": 0.49720435279264114,
      "flos": 21871825269120.0,
      "grad_norm": 2.9769715008671085,
      "language_loss": 0.80789787,
      "learning_rate": 2.1152239676491685e-06,
      "loss": 0.83023769,
      "num_input_tokens_seen": 89048145,
      "step": 4135,
      "time_per_iteration": 2.7559220790863037
    },
    {
      "auxiliary_loss_clip": 0.01194056,
      "auxiliary_loss_mlp": 0.01026662,
      "balance_loss_clip": 1.05014241,
      "balance_loss_mlp": 1.01857984,
      "epoch": 0.49732459568328025,
      "flos": 23805794367360.0,
      "grad_norm": 1.7314587364058138,
      "language_loss": 0.73286235,
      "learning_rate": 2.114446281114569e-06,
      "loss": 0.75506955,
      "num_input_tokens_seen": 89067165,
      "step": 4136,
      "time_per_iteration": 2.6679022312164307
    },
    {
      "auxiliary_loss_clip": 0.01181312,
      "auxiliary_loss_mlp": 0.01026662,
      "balance_loss_clip": 1.05147958,
      "balance_loss_mlp": 1.01843023,
      "epoch": 0.4974448385739193,
      "flos": 20047742853120.0,
      "grad_norm": 2.1337103558972887,
      "language_loss": 0.76359731,
      "learning_rate": 2.1136685772185587e-06,
      "loss": 0.78567702,
      "num_input_tokens_seen": 89086190,
      "step": 4137,
      "time_per_iteration": 2.643718957901001
    },
    {
      "auxiliary_loss_clip": 0.01188028,
      "auxiliary_loss_mlp": 0.01208922,
      "balance_loss_clip": 1.04675961,
      "balance_loss_mlp": 1.00055218,
      "epoch": 0.4975650814645584,
      "flos": 24821347593600.0,
      "grad_norm": 1.8122959867854773,
      "language_loss": 0.77947307,
      "learning_rate": 2.1128908560791163e-06,
      "loss": 0.8034426,
      "num_input_tokens_seen": 89106020,
      "step": 4138,
      "time_per_iteration": 2.7163193225860596
    },
    {
      "auxiliary_loss_clip": 0.01185221,
      "auxiliary_loss_mlp": 0.01025836,
      "balance_loss_clip": 1.05661058,
      "balance_loss_mlp": 1.0174911,
      "epoch": 0.4976853243551975,
      "flos": 19829477859840.0,
      "grad_norm": 2.3124854456307196,
      "language_loss": 0.78419143,
      "learning_rate": 2.1121131178142203e-06,
      "loss": 0.80630201,
      "num_input_tokens_seen": 89125385,
      "step": 4139,
      "time_per_iteration": 2.701552152633667
    },
    {
      "auxiliary_loss_clip": 0.01191643,
      "auxiliary_loss_mlp": 0.01021396,
      "balance_loss_clip": 1.05244994,
      "balance_loss_mlp": 1.01354015,
      "epoch": 0.4978055672458366,
      "flos": 23142990654720.0,
      "grad_norm": 1.619945487377555,
      "language_loss": 0.82599729,
      "learning_rate": 2.1113353625418544e-06,
      "loss": 0.84812766,
      "num_input_tokens_seen": 89143935,
      "step": 4140,
      "time_per_iteration": 2.6465022563934326
    },
    {
      "auxiliary_loss_clip": 0.01183058,
      "auxiliary_loss_mlp": 0.01028818,
      "balance_loss_clip": 1.05739594,
      "balance_loss_mlp": 1.02144504,
      "epoch": 0.4979258101364757,
      "flos": 15559914718080.0,
      "grad_norm": 1.8413039442477528,
      "language_loss": 0.79261971,
      "learning_rate": 2.1105575903800017e-06,
      "loss": 0.81473845,
      "num_input_tokens_seen": 89162655,
      "step": 4141,
      "time_per_iteration": 2.6194095611572266
    },
    {
      "auxiliary_loss_clip": 0.01192106,
      "auxiliary_loss_mlp": 0.01024015,
      "balance_loss_clip": 1.05332637,
      "balance_loss_mlp": 1.01559901,
      "epoch": 0.4980460530271148,
      "flos": 26356169784960.0,
      "grad_norm": 2.4981338144704215,
      "language_loss": 0.85511661,
      "learning_rate": 2.1097798014466502e-06,
      "loss": 0.87727785,
      "num_input_tokens_seen": 89182255,
      "step": 4142,
      "time_per_iteration": 2.652803659439087
    },
    {
      "auxiliary_loss_clip": 0.01192863,
      "auxiliary_loss_mlp": 0.01030027,
      "balance_loss_clip": 1.05646002,
      "balance_loss_mlp": 1.02115178,
      "epoch": 0.49816629591775385,
      "flos": 17274541415040.0,
      "grad_norm": 4.3650695540497075,
      "language_loss": 0.59430653,
      "learning_rate": 2.109001995859791e-06,
      "loss": 0.61653543,
      "num_input_tokens_seen": 89201155,
      "step": 4143,
      "time_per_iteration": 2.7489840984344482
    },
    {
      "auxiliary_loss_clip": 0.01095098,
      "auxiliary_loss_mlp": 0.01001496,
      "balance_loss_clip": 1.02227211,
      "balance_loss_mlp": 1.00059628,
      "epoch": 0.49828653880839296,
      "flos": 64930947344640.0,
      "grad_norm": 0.7872024038798299,
      "language_loss": 0.60003376,
      "learning_rate": 2.108224173737415e-06,
      "loss": 0.62099969,
      "num_input_tokens_seen": 89264455,
      "step": 4144,
      "time_per_iteration": 3.17710542678833
    },
    {
      "auxiliary_loss_clip": 0.01184972,
      "auxiliary_loss_mlp": 0.01027154,
      "balance_loss_clip": 1.04906571,
      "balance_loss_mlp": 1.0178076,
      "epoch": 0.498406781699032,
      "flos": 27484806003840.0,
      "grad_norm": 1.7642551411821237,
      "language_loss": 0.76184249,
      "learning_rate": 2.1074463351975183e-06,
      "loss": 0.78396374,
      "num_input_tokens_seen": 89283340,
      "step": 4145,
      "time_per_iteration": 2.6218581199645996
    },
    {
      "auxiliary_loss_clip": 0.01197161,
      "auxiliary_loss_mlp": 0.010234,
      "balance_loss_clip": 1.05139124,
      "balance_loss_mlp": 1.01540065,
      "epoch": 0.49852702458967113,
      "flos": 31499870307840.0,
      "grad_norm": 1.8516497035933217,
      "language_loss": 0.71944022,
      "learning_rate": 2.106668480358098e-06,
      "loss": 0.74164581,
      "num_input_tokens_seen": 89303565,
      "step": 4146,
      "time_per_iteration": 2.751462936401367
    },
    {
      "auxiliary_loss_clip": 0.01201828,
      "auxiliary_loss_mlp": 0.01026867,
      "balance_loss_clip": 1.04808903,
      "balance_loss_mlp": 1.01817632,
      "epoch": 0.49864726748031024,
      "flos": 22852868503680.0,
      "grad_norm": 2.0632586839555564,
      "language_loss": 0.71644413,
      "learning_rate": 2.105890609337154e-06,
      "loss": 0.73873103,
      "num_input_tokens_seen": 89322080,
      "step": 4147,
      "time_per_iteration": 2.7072932720184326
    },
    {
      "auxiliary_loss_clip": 0.01082655,
      "auxiliary_loss_mlp": 0.01000108,
      "balance_loss_clip": 1.02062464,
      "balance_loss_mlp": 0.99920845,
      "epoch": 0.4987675103709493,
      "flos": 70405708544640.0,
      "grad_norm": 0.6911568496568391,
      "language_loss": 0.63804579,
      "learning_rate": 2.1051127222526883e-06,
      "loss": 0.65887332,
      "num_input_tokens_seen": 89394195,
      "step": 4148,
      "time_per_iteration": 4.199272394180298
    },
    {
      "auxiliary_loss_clip": 0.01186423,
      "auxiliary_loss_mlp": 0.01021949,
      "balance_loss_clip": 1.05681086,
      "balance_loss_mlp": 1.01377702,
      "epoch": 0.4988877532615884,
      "flos": 28767571482240.0,
      "grad_norm": 1.6447208951703862,
      "language_loss": 0.80981809,
      "learning_rate": 2.1043348192227067e-06,
      "loss": 0.83190185,
      "num_input_tokens_seen": 89414565,
      "step": 4149,
      "time_per_iteration": 2.677027463912964
    },
    {
      "auxiliary_loss_clip": 0.01180684,
      "auxiliary_loss_mlp": 0.01029655,
      "balance_loss_clip": 1.04998434,
      "balance_loss_mlp": 1.0216502,
      "epoch": 0.4990079961522275,
      "flos": 16872700988160.0,
      "grad_norm": 2.0629352478220375,
      "language_loss": 0.62515646,
      "learning_rate": 2.1035569003652156e-06,
      "loss": 0.64725983,
      "num_input_tokens_seen": 89433195,
      "step": 4150,
      "time_per_iteration": 3.5539510250091553
    },
    {
      "auxiliary_loss_clip": 0.01193475,
      "auxiliary_loss_mlp": 0.01035081,
      "balance_loss_clip": 1.04945982,
      "balance_loss_mlp": 1.025365,
      "epoch": 0.4991282390428666,
      "flos": 13291042187520.0,
      "grad_norm": 2.5695435084195126,
      "language_loss": 0.8161934,
      "learning_rate": 2.1027789657982255e-06,
      "loss": 0.83847892,
      "num_input_tokens_seen": 89447410,
      "step": 4151,
      "time_per_iteration": 2.638745069503784
    },
    {
      "auxiliary_loss_clip": 0.0119346,
      "auxiliary_loss_mlp": 0.01034646,
      "balance_loss_clip": 1.04840875,
      "balance_loss_mlp": 1.02631903,
      "epoch": 0.4992484819335057,
      "flos": 21537496454400.0,
      "grad_norm": 2.46672852786511,
      "language_loss": 0.77282989,
      "learning_rate": 2.1020010156397482e-06,
      "loss": 0.79511094,
      "num_input_tokens_seen": 89464630,
      "step": 4152,
      "time_per_iteration": 2.6645543575286865
    },
    {
      "auxiliary_loss_clip": 0.01187468,
      "auxiliary_loss_mlp": 0.01027992,
      "balance_loss_clip": 1.05309081,
      "balance_loss_mlp": 1.01984358,
      "epoch": 0.4993687248241448,
      "flos": 24860095390080.0,
      "grad_norm": 1.7518812513150548,
      "language_loss": 0.77465689,
      "learning_rate": 2.101223050007797e-06,
      "loss": 0.79681152,
      "num_input_tokens_seen": 89483180,
      "step": 4153,
      "time_per_iteration": 3.594761371612549
    },
    {
      "auxiliary_loss_clip": 0.01081166,
      "auxiliary_loss_mlp": 0.01000261,
      "balance_loss_clip": 1.01888561,
      "balance_loss_mlp": 0.99942106,
      "epoch": 0.49948896771478385,
      "flos": 62941602453120.0,
      "grad_norm": 0.8203247911781525,
      "language_loss": 0.53774083,
      "learning_rate": 2.1004450690203904e-06,
      "loss": 0.55855507,
      "num_input_tokens_seen": 89539260,
      "step": 4154,
      "time_per_iteration": 3.14102840423584
    },
    {
      "auxiliary_loss_clip": 0.01080943,
      "auxiliary_loss_mlp": 0.01000694,
      "balance_loss_clip": 1.01903796,
      "balance_loss_mlp": 0.99987108,
      "epoch": 0.49960921060542296,
      "flos": 68284213516800.0,
      "grad_norm": 0.8515289744614988,
      "language_loss": 0.63285625,
      "learning_rate": 2.099667072795546e-06,
      "loss": 0.65367258,
      "num_input_tokens_seen": 89601380,
      "step": 4155,
      "time_per_iteration": 3.1378839015960693
    },
    {
      "auxiliary_loss_clip": 0.01185024,
      "auxiliary_loss_mlp": 0.01029696,
      "balance_loss_clip": 1.05180645,
      "balance_loss_mlp": 1.02104723,
      "epoch": 0.49972945349606207,
      "flos": 23659350618240.0,
      "grad_norm": 2.7643118355012946,
      "language_loss": 0.79788566,
      "learning_rate": 2.0988890614512864e-06,
      "loss": 0.82003289,
      "num_input_tokens_seen": 89621270,
      "step": 4156,
      "time_per_iteration": 2.6564576625823975
    },
    {
      "auxiliary_loss_clip": 0.01194325,
      "auxiliary_loss_mlp": 0.01029903,
      "balance_loss_clip": 1.05541158,
      "balance_loss_mlp": 1.02173758,
      "epoch": 0.4998496963867011,
      "flos": 19755825022080.0,
      "grad_norm": 4.289922225703048,
      "language_loss": 0.84407961,
      "learning_rate": 2.098111035105635e-06,
      "loss": 0.86632192,
      "num_input_tokens_seen": 89639695,
      "step": 4157,
      "time_per_iteration": 2.602576971054077
    },
    {
      "auxiliary_loss_clip": 0.01193258,
      "auxiliary_loss_mlp": 0.01029391,
      "balance_loss_clip": 1.05150592,
      "balance_loss_mlp": 1.02119577,
      "epoch": 0.49996993927734024,
      "flos": 22265728790400.0,
      "grad_norm": 1.6876476655585415,
      "language_loss": 0.72982812,
      "learning_rate": 2.0973329938766176e-06,
      "loss": 0.75205463,
      "num_input_tokens_seen": 89657125,
      "step": 4158,
      "time_per_iteration": 2.6753532886505127
    },
    {
      "auxiliary_loss_clip": 0.0119197,
      "auxiliary_loss_mlp": 0.01031118,
      "balance_loss_clip": 1.05436027,
      "balance_loss_mlp": 1.02226651,
      "epoch": 0.5000901821679793,
      "flos": 23327212533120.0,
      "grad_norm": 5.052896714084705,
      "language_loss": 0.78877485,
      "learning_rate": 2.0965549378822618e-06,
      "loss": 0.81100571,
      "num_input_tokens_seen": 89678415,
      "step": 4159,
      "time_per_iteration": 3.8355207443237305
    },
    {
      "auxiliary_loss_clip": 0.01194384,
      "auxiliary_loss_mlp": 0.0102846,
      "balance_loss_clip": 1.04115915,
      "balance_loss_mlp": 1.02003741,
      "epoch": 0.5002104250586185,
      "flos": 20339014239360.0,
      "grad_norm": 1.9452265581120203,
      "language_loss": 0.84229851,
      "learning_rate": 2.095776867240599e-06,
      "loss": 0.86452687,
      "num_input_tokens_seen": 89695405,
      "step": 4160,
      "time_per_iteration": 2.731735944747925
    },
    {
      "auxiliary_loss_clip": 0.0118529,
      "auxiliary_loss_mlp": 0.01028949,
      "balance_loss_clip": 1.04870868,
      "balance_loss_mlp": 1.02117062,
      "epoch": 0.5003306679492575,
      "flos": 13991372634240.0,
      "grad_norm": 2.1397398409947193,
      "language_loss": 0.82360148,
      "learning_rate": 2.094998782069661e-06,
      "loss": 0.84574383,
      "num_input_tokens_seen": 89713110,
      "step": 4161,
      "time_per_iteration": 2.723897695541382
    },
    {
      "auxiliary_loss_clip": 0.01185795,
      "auxiliary_loss_mlp": 0.01028782,
      "balance_loss_clip": 1.0569663,
      "balance_loss_mlp": 1.02033615,
      "epoch": 0.5004509108398966,
      "flos": 27672762896640.0,
      "grad_norm": 1.785400063249577,
      "language_loss": 0.75702214,
      "learning_rate": 2.0942206824874845e-06,
      "loss": 0.77916789,
      "num_input_tokens_seen": 89735885,
      "step": 4162,
      "time_per_iteration": 2.6551198959350586
    },
    {
      "auxiliary_loss_clip": 0.01182561,
      "auxiliary_loss_mlp": 0.01026237,
      "balance_loss_clip": 1.05351198,
      "balance_loss_mlp": 1.01802349,
      "epoch": 0.5005711537305357,
      "flos": 14976186796800.0,
      "grad_norm": 2.310867007201922,
      "language_loss": 0.7924993,
      "learning_rate": 2.093442568612105e-06,
      "loss": 0.8145873,
      "num_input_tokens_seen": 89753690,
      "step": 4163,
      "time_per_iteration": 2.617410182952881
    },
    {
      "auxiliary_loss_clip": 0.01182611,
      "auxiliary_loss_mlp": 0.01023585,
      "balance_loss_clip": 1.05312681,
      "balance_loss_mlp": 1.01516294,
      "epoch": 0.5006913966211748,
      "flos": 26503259978880.0,
      "grad_norm": 2.3913992366877412,
      "language_loss": 0.84782684,
      "learning_rate": 2.0926644405615613e-06,
      "loss": 0.86988884,
      "num_input_tokens_seen": 89774590,
      "step": 4164,
      "time_per_iteration": 2.63340163230896
    },
    {
      "auxiliary_loss_clip": 0.01184395,
      "auxiliary_loss_mlp": 0.01028581,
      "balance_loss_clip": 1.04971278,
      "balance_loss_mlp": 1.02062356,
      "epoch": 0.5008116395118138,
      "flos": 20449295971200.0,
      "grad_norm": 2.027693044000367,
      "language_loss": 0.81398988,
      "learning_rate": 2.091886298453897e-06,
      "loss": 0.83611965,
      "num_input_tokens_seen": 89792775,
      "step": 4165,
      "time_per_iteration": 2.6887876987457275
    },
    {
      "auxiliary_loss_clip": 0.01184489,
      "auxiliary_loss_mlp": 0.01021851,
      "balance_loss_clip": 1.05272412,
      "balance_loss_mlp": 1.01426005,
      "epoch": 0.500931882402453,
      "flos": 21579871524480.0,
      "grad_norm": 6.416495944564292,
      "language_loss": 0.72877419,
      "learning_rate": 2.091108142407153e-06,
      "loss": 0.75083756,
      "num_input_tokens_seen": 89811515,
      "step": 4166,
      "time_per_iteration": 2.6403708457946777
    },
    {
      "auxiliary_loss_clip": 0.01094952,
      "auxiliary_loss_mlp": 0.01005431,
      "balance_loss_clip": 1.02686501,
      "balance_loss_mlp": 1.00456071,
      "epoch": 0.5010521252930921,
      "flos": 57785011925760.0,
      "grad_norm": 0.8440468419447005,
      "language_loss": 0.62405193,
      "learning_rate": 2.090329972539377e-06,
      "loss": 0.64505565,
      "num_input_tokens_seen": 89870080,
      "step": 4167,
      "time_per_iteration": 3.2171995639801025
    },
    {
      "auxiliary_loss_clip": 0.01189596,
      "auxiliary_loss_mlp": 0.01027618,
      "balance_loss_clip": 1.04221439,
      "balance_loss_mlp": 1.01973808,
      "epoch": 0.5011723681837311,
      "flos": 18625500864000.0,
      "grad_norm": 2.200996374131747,
      "language_loss": 0.68597573,
      "learning_rate": 2.089551788968616e-06,
      "loss": 0.70814788,
      "num_input_tokens_seen": 89888045,
      "step": 4168,
      "time_per_iteration": 2.743647575378418
    },
    {
      "auxiliary_loss_clip": 0.01079377,
      "auxiliary_loss_mlp": 0.01001388,
      "balance_loss_clip": 1.01814747,
      "balance_loss_mlp": 1.00058985,
      "epoch": 0.5012926110743702,
      "flos": 55883146608000.0,
      "grad_norm": 0.8362663128629643,
      "language_loss": 0.60699719,
      "learning_rate": 2.08877359181292e-06,
      "loss": 0.62780482,
      "num_input_tokens_seen": 89944610,
      "step": 4169,
      "time_per_iteration": 3.227325201034546
    },
    {
      "auxiliary_loss_clip": 0.01193156,
      "auxiliary_loss_mlp": 0.01025946,
      "balance_loss_clip": 1.04668665,
      "balance_loss_mlp": 1.01810789,
      "epoch": 0.5014128539650093,
      "flos": 24238266117120.0,
      "grad_norm": 2.5463522404096666,
      "language_loss": 0.85676903,
      "learning_rate": 2.0879953811903396e-06,
      "loss": 0.87896001,
      "num_input_tokens_seen": 89959495,
      "step": 4170,
      "time_per_iteration": 2.707547664642334
    },
    {
      "auxiliary_loss_clip": 0.01187129,
      "auxiliary_loss_mlp": 0.01029039,
      "balance_loss_clip": 1.05591261,
      "balance_loss_mlp": 1.0205512,
      "epoch": 0.5015330968556484,
      "flos": 27527468382720.0,
      "grad_norm": 1.8289346645533202,
      "language_loss": 0.78774214,
      "learning_rate": 2.08721715721893e-06,
      "loss": 0.80990386,
      "num_input_tokens_seen": 89978820,
      "step": 4171,
      "time_per_iteration": 2.6911556720733643
    },
    {
      "auxiliary_loss_clip": 0.0118572,
      "auxiliary_loss_mlp": 0.01026036,
      "balance_loss_clip": 1.05440462,
      "balance_loss_mlp": 1.01792407,
      "epoch": 0.5016533397462875,
      "flos": 23800802376960.0,
      "grad_norm": 1.8667302457959731,
      "language_loss": 0.772403,
      "learning_rate": 2.0864389200167477e-06,
      "loss": 0.79452056,
      "num_input_tokens_seen": 89997075,
      "step": 4172,
      "time_per_iteration": 2.730039358139038
    },
    {
      "auxiliary_loss_clip": 0.01186636,
      "auxiliary_loss_mlp": 0.01208268,
      "balance_loss_clip": 1.05257881,
      "balance_loss_mlp": 1.00058556,
      "epoch": 0.5017735826369266,
      "flos": 25295009264640.0,
      "grad_norm": 2.2573914953250154,
      "language_loss": 0.7940737,
      "learning_rate": 2.0856606697018504e-06,
      "loss": 0.81802273,
      "num_input_tokens_seen": 90015085,
      "step": 4173,
      "time_per_iteration": 2.6391615867614746
    },
    {
      "auxiliary_loss_clip": 0.01182457,
      "auxiliary_loss_mlp": 0.01029725,
      "balance_loss_clip": 1.0498482,
      "balance_loss_mlp": 1.02129054,
      "epoch": 0.5018938255275657,
      "flos": 16873203778560.0,
      "grad_norm": 3.0705621975889508,
      "language_loss": 0.73571932,
      "learning_rate": 2.084882406392297e-06,
      "loss": 0.75784117,
      "num_input_tokens_seen": 90033045,
      "step": 4174,
      "time_per_iteration": 2.6052439212799072
    },
    {
      "auxiliary_loss_clip": 0.01190774,
      "auxiliary_loss_mlp": 0.01027584,
      "balance_loss_clip": 1.05694568,
      "balance_loss_mlp": 1.01947212,
      "epoch": 0.5020140684182047,
      "flos": 25515429073920.0,
      "grad_norm": 2.660469323374579,
      "language_loss": 0.71167707,
      "learning_rate": 2.0841041302061496e-06,
      "loss": 0.73386061,
      "num_input_tokens_seen": 90052505,
      "step": 4175,
      "time_per_iteration": 3.5719048976898193
    },
    {
      "auxiliary_loss_clip": 0.0117683,
      "auxiliary_loss_mlp": 0.01029373,
      "balance_loss_clip": 1.04755521,
      "balance_loss_mlp": 1.02129018,
      "epoch": 0.5021343113088439,
      "flos": 23659278791040.0,
      "grad_norm": 1.9997466387592995,
      "language_loss": 0.75975907,
      "learning_rate": 2.083325841261473e-06,
      "loss": 0.78182107,
      "num_input_tokens_seen": 90071565,
      "step": 4176,
      "time_per_iteration": 2.6468727588653564
    },
    {
      "auxiliary_loss_clip": 0.01181333,
      "auxiliary_loss_mlp": 0.01028509,
      "balance_loss_clip": 1.04862583,
      "balance_loss_mlp": 1.02059913,
      "epoch": 0.502254554199483,
      "flos": 24534673148160.0,
      "grad_norm": 3.8075727044830905,
      "language_loss": 0.66488016,
      "learning_rate": 2.0825475396763322e-06,
      "loss": 0.68697858,
      "num_input_tokens_seen": 90092215,
      "step": 4177,
      "time_per_iteration": 3.6040632724761963
    },
    {
      "auxiliary_loss_clip": 0.01189707,
      "auxiliary_loss_mlp": 0.0102683,
      "balance_loss_clip": 1.04232061,
      "balance_loss_mlp": 1.01827025,
      "epoch": 0.502374797090122,
      "flos": 34240285607040.0,
      "grad_norm": 1.3929457055643049,
      "language_loss": 0.65818828,
      "learning_rate": 2.081769225568796e-06,
      "loss": 0.68035364,
      "num_input_tokens_seen": 90114665,
      "step": 4178,
      "time_per_iteration": 2.8560450077056885
    },
    {
      "auxiliary_loss_clip": 0.01185997,
      "auxiliary_loss_mlp": 0.01029964,
      "balance_loss_clip": 1.05067217,
      "balance_loss_mlp": 1.0211668,
      "epoch": 0.5024950399807612,
      "flos": 26031106679040.0,
      "grad_norm": 1.6761192964973848,
      "language_loss": 0.75968462,
      "learning_rate": 2.0809908990569327e-06,
      "loss": 0.78184426,
      "num_input_tokens_seen": 90136445,
      "step": 4179,
      "time_per_iteration": 3.615945339202881
    },
    {
      "auxiliary_loss_clip": 0.01188491,
      "auxiliary_loss_mlp": 0.01029104,
      "balance_loss_clip": 1.05263543,
      "balance_loss_mlp": 1.02048492,
      "epoch": 0.5026152828714002,
      "flos": 21252438120960.0,
      "grad_norm": 1.8342653911448303,
      "language_loss": 0.79266834,
      "learning_rate": 2.0802125602588146e-06,
      "loss": 0.81484437,
      "num_input_tokens_seen": 90155710,
      "step": 4180,
      "time_per_iteration": 2.676126718521118
    },
    {
      "auxiliary_loss_clip": 0.01185699,
      "auxiliary_loss_mlp": 0.01033794,
      "balance_loss_clip": 1.05756509,
      "balance_loss_mlp": 1.02536607,
      "epoch": 0.5027355257620393,
      "flos": 30956111245440.0,
      "grad_norm": 1.8774672928566374,
      "language_loss": 0.66444743,
      "learning_rate": 2.0794342092925146e-06,
      "loss": 0.68664241,
      "num_input_tokens_seen": 90176845,
      "step": 4181,
      "time_per_iteration": 2.6582255363464355
    },
    {
      "auxiliary_loss_clip": 0.01190421,
      "auxiliary_loss_mlp": 0.010317,
      "balance_loss_clip": 1.05700517,
      "balance_loss_mlp": 1.02348018,
      "epoch": 0.5028557686526784,
      "flos": 24791147233920.0,
      "grad_norm": 2.046541891878454,
      "language_loss": 0.67700666,
      "learning_rate": 2.078655846276108e-06,
      "loss": 0.69922793,
      "num_input_tokens_seen": 90197175,
      "step": 4182,
      "time_per_iteration": 2.6326942443847656
    },
    {
      "auxiliary_loss_clip": 0.01183527,
      "auxiliary_loss_mlp": 0.01025382,
      "balance_loss_clip": 1.05177283,
      "balance_loss_mlp": 1.01715624,
      "epoch": 0.5029760115433175,
      "flos": 22966992990720.0,
      "grad_norm": 2.0377444357684533,
      "language_loss": 0.68877715,
      "learning_rate": 2.0778774713276727e-06,
      "loss": 0.71086621,
      "num_input_tokens_seen": 90216650,
      "step": 4183,
      "time_per_iteration": 2.7720141410827637
    },
    {
      "auxiliary_loss_clip": 0.01181194,
      "auxiliary_loss_mlp": 0.0102703,
      "balance_loss_clip": 1.05021191,
      "balance_loss_mlp": 1.0184474,
      "epoch": 0.5030962544339566,
      "flos": 15305164485120.0,
      "grad_norm": 7.079798939133971,
      "language_loss": 0.68019903,
      "learning_rate": 2.077099084565287e-06,
      "loss": 0.70228124,
      "num_input_tokens_seen": 90234055,
      "step": 4184,
      "time_per_iteration": 2.66790771484375
    },
    {
      "auxiliary_loss_clip": 0.01181905,
      "auxiliary_loss_mlp": 0.0102637,
      "balance_loss_clip": 1.04856956,
      "balance_loss_mlp": 1.0183239,
      "epoch": 0.5032164973245957,
      "flos": 24494847943680.0,
      "grad_norm": 2.3158625869080542,
      "language_loss": 0.65478128,
      "learning_rate": 2.0763206861070313e-06,
      "loss": 0.67686403,
      "num_input_tokens_seen": 90253115,
      "step": 4185,
      "time_per_iteration": 3.598243474960327
    },
    {
      "auxiliary_loss_clip": 0.01185413,
      "auxiliary_loss_mlp": 0.01032029,
      "balance_loss_clip": 1.0556097,
      "balance_loss_mlp": 1.02359462,
      "epoch": 0.5033367402152348,
      "flos": 16213452721920.0,
      "grad_norm": 2.477807144245697,
      "language_loss": 0.75446069,
      "learning_rate": 2.0755422760709876e-06,
      "loss": 0.77663511,
      "num_input_tokens_seen": 90270515,
      "step": 4186,
      "time_per_iteration": 2.5707764625549316
    },
    {
      "auxiliary_loss_clip": 0.01183446,
      "auxiliary_loss_mlp": 0.0103204,
      "balance_loss_clip": 1.04490161,
      "balance_loss_mlp": 1.02365971,
      "epoch": 0.5034569831058738,
      "flos": 21391375927680.0,
      "grad_norm": 1.915086022279511,
      "language_loss": 0.77095747,
      "learning_rate": 2.0747638545752417e-06,
      "loss": 0.79311234,
      "num_input_tokens_seen": 90289075,
      "step": 4187,
      "time_per_iteration": 2.6886885166168213
    },
    {
      "auxiliary_loss_clip": 0.01189191,
      "auxiliary_loss_mlp": 0.01025168,
      "balance_loss_clip": 1.05560672,
      "balance_loss_mlp": 1.01718092,
      "epoch": 0.503577225996513,
      "flos": 20558751690240.0,
      "grad_norm": 1.9543313637689725,
      "language_loss": 0.83492875,
      "learning_rate": 2.073985421737878e-06,
      "loss": 0.85707235,
      "num_input_tokens_seen": 90306385,
      "step": 4188,
      "time_per_iteration": 2.6787495613098145
    },
    {
      "auxiliary_loss_clip": 0.01188849,
      "auxiliary_loss_mlp": 0.01025165,
      "balance_loss_clip": 1.05488944,
      "balance_loss_mlp": 1.01686192,
      "epoch": 0.5036974688871521,
      "flos": 27229157930880.0,
      "grad_norm": 3.5106094953536586,
      "language_loss": 0.73861802,
      "learning_rate": 2.0732069776769844e-06,
      "loss": 0.76075816,
      "num_input_tokens_seen": 90323795,
      "step": 4189,
      "time_per_iteration": 2.726956844329834
    },
    {
      "auxiliary_loss_clip": 0.01183786,
      "auxiliary_loss_mlp": 0.01029302,
      "balance_loss_clip": 1.05607677,
      "balance_loss_mlp": 1.0203675,
      "epoch": 0.5038177117777911,
      "flos": 20412164286720.0,
      "grad_norm": 2.270616252568371,
      "language_loss": 0.73067212,
      "learning_rate": 2.072428522510651e-06,
      "loss": 0.75280303,
      "num_input_tokens_seen": 90340360,
      "step": 4190,
      "time_per_iteration": 2.6055548191070557
    },
    {
      "auxiliary_loss_clip": 0.01180266,
      "auxiliary_loss_mlp": 0.01028358,
      "balance_loss_clip": 1.04821777,
      "balance_loss_mlp": 1.02051365,
      "epoch": 0.5039379546684303,
      "flos": 21907987286400.0,
      "grad_norm": 2.4178731709428254,
      "language_loss": 0.76494473,
      "learning_rate": 2.071650056356968e-06,
      "loss": 0.78703099,
      "num_input_tokens_seen": 90357900,
      "step": 4191,
      "time_per_iteration": 2.677124261856079
    },
    {
      "auxiliary_loss_clip": 0.01181483,
      "auxiliary_loss_mlp": 0.01029739,
      "balance_loss_clip": 1.05450749,
      "balance_loss_mlp": 1.02183557,
      "epoch": 0.5040581975590693,
      "flos": 20010718909440.0,
      "grad_norm": 2.026318695002732,
      "language_loss": 0.80098379,
      "learning_rate": 2.070871579334028e-06,
      "loss": 0.82309598,
      "num_input_tokens_seen": 90377010,
      "step": 4192,
      "time_per_iteration": 2.604602336883545
    },
    {
      "auxiliary_loss_clip": 0.01181693,
      "auxiliary_loss_mlp": 0.01025228,
      "balance_loss_clip": 1.05452824,
      "balance_loss_mlp": 1.01688969,
      "epoch": 0.5041784404497084,
      "flos": 20959837931520.0,
      "grad_norm": 2.243633427179727,
      "language_loss": 0.71786273,
      "learning_rate": 2.0700930915599264e-06,
      "loss": 0.73993194,
      "num_input_tokens_seen": 90396740,
      "step": 4193,
      "time_per_iteration": 2.666048526763916
    },
    {
      "auxiliary_loss_clip": 0.01181536,
      "auxiliary_loss_mlp": 0.01026328,
      "balance_loss_clip": 1.05434299,
      "balance_loss_mlp": 1.01856184,
      "epoch": 0.5042986833403476,
      "flos": 12495082757760.0,
      "grad_norm": 2.1133327150661887,
      "language_loss": 0.78404331,
      "learning_rate": 2.0693145931527583e-06,
      "loss": 0.80612195,
      "num_input_tokens_seen": 90413220,
      "step": 4194,
      "time_per_iteration": 2.6943624019622803
    },
    {
      "auxiliary_loss_clip": 0.01181377,
      "auxiliary_loss_mlp": 0.0102798,
      "balance_loss_clip": 1.05053949,
      "balance_loss_mlp": 1.01971245,
      "epoch": 0.5044189262309866,
      "flos": 29202305788800.0,
      "grad_norm": 1.5670602545573549,
      "language_loss": 0.77701831,
      "learning_rate": 2.068536084230622e-06,
      "loss": 0.7991119,
      "num_input_tokens_seen": 90435085,
      "step": 4195,
      "time_per_iteration": 2.780693292617798
    },
    {
      "auxiliary_loss_clip": 0.01187052,
      "auxiliary_loss_mlp": 0.01031925,
      "balance_loss_clip": 1.05506635,
      "balance_loss_mlp": 1.02235866,
      "epoch": 0.5045391691216257,
      "flos": 23873198238720.0,
      "grad_norm": 2.329557464478533,
      "language_loss": 0.8861174,
      "learning_rate": 2.067757564911616e-06,
      "loss": 0.90830719,
      "num_input_tokens_seen": 90453660,
      "step": 4196,
      "time_per_iteration": 2.612914800643921
    },
    {
      "auxiliary_loss_clip": 0.0119621,
      "auxiliary_loss_mlp": 0.01208478,
      "balance_loss_clip": 1.05323672,
      "balance_loss_mlp": 1.00065947,
      "epoch": 0.5046594120122648,
      "flos": 24644990793600.0,
      "grad_norm": 3.832081571586154,
      "language_loss": 0.92597765,
      "learning_rate": 2.0669790353138407e-06,
      "loss": 0.9500246,
      "num_input_tokens_seen": 90472625,
      "step": 4197,
      "time_per_iteration": 2.7068960666656494
    },
    {
      "auxiliary_loss_clip": 0.01183584,
      "auxiliary_loss_mlp": 0.0120862,
      "balance_loss_clip": 1.04934692,
      "balance_loss_mlp": 1.00080466,
      "epoch": 0.5047796549029039,
      "flos": 23362835846400.0,
      "grad_norm": 2.3885187191160595,
      "language_loss": 0.73075444,
      "learning_rate": 2.0662004955553995e-06,
      "loss": 0.75467646,
      "num_input_tokens_seen": 90492325,
      "step": 4198,
      "time_per_iteration": 2.6500344276428223
    },
    {
      "auxiliary_loss_clip": 0.01181539,
      "auxiliary_loss_mlp": 0.0102199,
      "balance_loss_clip": 1.0487802,
      "balance_loss_mlp": 1.01405704,
      "epoch": 0.5048998977935429,
      "flos": 17304095329920.0,
      "grad_norm": 2.159695884650177,
      "language_loss": 0.77065301,
      "learning_rate": 2.065421945754395e-06,
      "loss": 0.79268825,
      "num_input_tokens_seen": 90510055,
      "step": 4199,
      "time_per_iteration": 2.6792819499969482
    },
    {
      "auxiliary_loss_clip": 0.01191609,
      "auxiliary_loss_mlp": 0.01027472,
      "balance_loss_clip": 1.04931378,
      "balance_loss_mlp": 1.01988471,
      "epoch": 0.505020140684182,
      "flos": 34856979235200.0,
      "grad_norm": 3.4779192810839867,
      "language_loss": 0.78225297,
      "learning_rate": 2.0646433860289344e-06,
      "loss": 0.80444378,
      "num_input_tokens_seen": 90528980,
      "step": 4200,
      "time_per_iteration": 2.7970643043518066
    },
    {
      "auxiliary_loss_clip": 0.01189706,
      "auxiliary_loss_mlp": 0.01209119,
      "balance_loss_clip": 1.05337656,
      "balance_loss_mlp": 1.00075197,
      "epoch": 0.5051403835748212,
      "flos": 24863974058880.0,
      "grad_norm": 2.3190849733786454,
      "language_loss": 0.82571614,
      "learning_rate": 2.0638648164971233e-06,
      "loss": 0.84970438,
      "num_input_tokens_seen": 90547445,
      "step": 4201,
      "time_per_iteration": 2.6786446571350098
    },
    {
      "auxiliary_loss_clip": 0.01185752,
      "auxiliary_loss_mlp": 0.01024256,
      "balance_loss_clip": 1.05267978,
      "balance_loss_mlp": 1.01700199,
      "epoch": 0.5052606264654602,
      "flos": 20959694277120.0,
      "grad_norm": 2.1804311464756387,
      "language_loss": 0.88304579,
      "learning_rate": 2.06308623727707e-06,
      "loss": 0.90514588,
      "num_input_tokens_seen": 90567545,
      "step": 4202,
      "time_per_iteration": 3.5623648166656494
    },
    {
      "auxiliary_loss_clip": 0.01180721,
      "auxiliary_loss_mlp": 0.01025519,
      "balance_loss_clip": 1.05260539,
      "balance_loss_mlp": 1.0168047,
      "epoch": 0.5053808693560993,
      "flos": 19642382893440.0,
      "grad_norm": 3.24315000443397,
      "language_loss": 0.76434249,
      "learning_rate": 2.0623076484868846e-06,
      "loss": 0.78640491,
      "num_input_tokens_seen": 90585000,
      "step": 4203,
      "time_per_iteration": 2.548647880554199
    },
    {
      "auxiliary_loss_clip": 0.0108751,
      "auxiliary_loss_mlp": 0.01000446,
      "balance_loss_clip": 1.01895893,
      "balance_loss_mlp": 0.99962312,
      "epoch": 0.5055011122467384,
      "flos": 67504915019520.0,
      "grad_norm": 0.8327350150707393,
      "language_loss": 0.60682189,
      "learning_rate": 2.061529050244679e-06,
      "loss": 0.62770152,
      "num_input_tokens_seen": 90644745,
      "step": 4204,
      "time_per_iteration": 4.0922465324401855
    },
    {
      "auxiliary_loss_clip": 0.01195086,
      "auxiliary_loss_mlp": 0.01020821,
      "balance_loss_clip": 1.049842,
      "balance_loss_mlp": 1.01230955,
      "epoch": 0.5056213551373775,
      "flos": 16872952383360.0,
      "grad_norm": 2.8388898414587485,
      "language_loss": 0.73932046,
      "learning_rate": 2.060750442668565e-06,
      "loss": 0.7614795,
      "num_input_tokens_seen": 90662500,
      "step": 4205,
      "time_per_iteration": 2.643085479736328
    },
    {
      "auxiliary_loss_clip": 0.01186453,
      "auxiliary_loss_mlp": 0.01026645,
      "balance_loss_clip": 1.05611181,
      "balance_loss_mlp": 1.01821661,
      "epoch": 0.5057415980280165,
      "flos": 15334179696000.0,
      "grad_norm": 2.488769475648384,
      "language_loss": 0.64214325,
      "learning_rate": 2.059971825876657e-06,
      "loss": 0.66427428,
      "num_input_tokens_seen": 90677010,
      "step": 4206,
      "time_per_iteration": 3.5586419105529785
    },
    {
      "auxiliary_loss_clip": 0.0118658,
      "auxiliary_loss_mlp": 0.01023321,
      "balance_loss_clip": 1.05385792,
      "balance_loss_mlp": 1.01536727,
      "epoch": 0.5058618409186557,
      "flos": 19025976574080.0,
      "grad_norm": 1.84612313087402,
      "language_loss": 0.76561081,
      "learning_rate": 2.0591931999870713e-06,
      "loss": 0.78770983,
      "num_input_tokens_seen": 90695935,
      "step": 4207,
      "time_per_iteration": 2.62520170211792
    },
    {
      "auxiliary_loss_clip": 0.01079584,
      "auxiliary_loss_mlp": 0.00999817,
      "balance_loss_clip": 1.01842594,
      "balance_loss_mlp": 0.99904263,
      "epoch": 0.5059820838092948,
      "flos": 63453114080640.0,
      "grad_norm": 0.839257722739257,
      "language_loss": 0.57540536,
      "learning_rate": 2.0584145651179234e-06,
      "loss": 0.59619939,
      "num_input_tokens_seen": 90751645,
      "step": 4208,
      "time_per_iteration": 3.176307201385498
    },
    {
      "auxiliary_loss_clip": 0.01186984,
      "auxiliary_loss_mlp": 0.01208061,
      "balance_loss_clip": 1.05259979,
      "balance_loss_mlp": 1.00070071,
      "epoch": 0.5061023266999338,
      "flos": 15441803821440.0,
      "grad_norm": 2.5558238237644098,
      "language_loss": 0.79473495,
      "learning_rate": 2.0576359213873327e-06,
      "loss": 0.81868541,
      "num_input_tokens_seen": 90766795,
      "step": 4209,
      "time_per_iteration": 2.579697370529175
    },
    {
      "auxiliary_loss_clip": 0.01193689,
      "auxiliary_loss_mlp": 0.01027915,
      "balance_loss_clip": 1.04906583,
      "balance_loss_mlp": 1.01962399,
      "epoch": 0.506222569590573,
      "flos": 22451063990400.0,
      "grad_norm": 2.5502285170554764,
      "language_loss": 0.70502567,
      "learning_rate": 2.056857268913419e-06,
      "loss": 0.72724164,
      "num_input_tokens_seen": 90786845,
      "step": 4210,
      "time_per_iteration": 2.7011120319366455
    },
    {
      "auxiliary_loss_clip": 0.01185531,
      "auxiliary_loss_mlp": 0.01025981,
      "balance_loss_clip": 1.05550468,
      "balance_loss_mlp": 1.01814258,
      "epoch": 0.506342812481212,
      "flos": 17558665994880.0,
      "grad_norm": 3.0814334786998607,
      "language_loss": 0.84145784,
      "learning_rate": 2.056078607814303e-06,
      "loss": 0.86357296,
      "num_input_tokens_seen": 90802630,
      "step": 4211,
      "time_per_iteration": 2.5936238765716553
    },
    {
      "auxiliary_loss_clip": 0.01185053,
      "auxiliary_loss_mlp": 0.01023094,
      "balance_loss_clip": 1.05331326,
      "balance_loss_mlp": 1.01460063,
      "epoch": 0.5064630553718511,
      "flos": 23402050519680.0,
      "grad_norm": 7.2128194671034604,
      "language_loss": 0.78169495,
      "learning_rate": 2.055299938208106e-06,
      "loss": 0.80377638,
      "num_input_tokens_seen": 90823620,
      "step": 4212,
      "time_per_iteration": 3.526092290878296
    },
    {
      "auxiliary_loss_clip": 0.01189977,
      "auxiliary_loss_mlp": 0.0102906,
      "balance_loss_clip": 1.0557878,
      "balance_loss_mlp": 1.02057278,
      "epoch": 0.5065832982624903,
      "flos": 23987035416960.0,
      "grad_norm": 2.838366281081007,
      "language_loss": 0.86154592,
      "learning_rate": 2.0545212602129526e-06,
      "loss": 0.88373631,
      "num_input_tokens_seen": 90843475,
      "step": 4213,
      "time_per_iteration": 2.6509599685668945
    },
    {
      "auxiliary_loss_clip": 0.01177811,
      "auxiliary_loss_mlp": 0.01029453,
      "balance_loss_clip": 1.04727006,
      "balance_loss_mlp": 1.02040541,
      "epoch": 0.5067035411531293,
      "flos": 21503058289920.0,
      "grad_norm": 3.4420952593781635,
      "language_loss": 0.66591978,
      "learning_rate": 2.0537425739469673e-06,
      "loss": 0.68799239,
      "num_input_tokens_seen": 90862410,
      "step": 4214,
      "time_per_iteration": 2.638489246368408
    },
    {
      "auxiliary_loss_clip": 0.01085863,
      "auxiliary_loss_mlp": 0.01003237,
      "balance_loss_clip": 1.0177213,
      "balance_loss_mlp": 1.0023787,
      "epoch": 0.5068237840437684,
      "flos": 65934397687680.0,
      "grad_norm": 0.8392365993149677,
      "language_loss": 0.59412342,
      "learning_rate": 2.052963879528276e-06,
      "loss": 0.61501437,
      "num_input_tokens_seen": 90922280,
      "step": 4215,
      "time_per_iteration": 3.1353132724761963
    },
    {
      "auxiliary_loss_clip": 0.0118742,
      "auxiliary_loss_mlp": 0.01025341,
      "balance_loss_clip": 1.05543494,
      "balance_loss_mlp": 1.01710916,
      "epoch": 0.5069440269344075,
      "flos": 27264206626560.0,
      "grad_norm": 2.289248189632405,
      "language_loss": 0.76495248,
      "learning_rate": 2.052185177075007e-06,
      "loss": 0.78708005,
      "num_input_tokens_seen": 90941850,
      "step": 4216,
      "time_per_iteration": 2.653268575668335
    },
    {
      "auxiliary_loss_clip": 0.01187128,
      "auxiliary_loss_mlp": 0.01032286,
      "balance_loss_clip": 1.05345869,
      "balance_loss_mlp": 1.02435899,
      "epoch": 0.5070642698250466,
      "flos": 23366319465600.0,
      "grad_norm": 2.0974440765806577,
      "language_loss": 0.82881141,
      "learning_rate": 2.051406466705288e-06,
      "loss": 0.85100555,
      "num_input_tokens_seen": 90961390,
      "step": 4217,
      "time_per_iteration": 2.717123508453369
    },
    {
      "auxiliary_loss_clip": 0.01181769,
      "auxiliary_loss_mlp": 0.01026854,
      "balance_loss_clip": 1.05377531,
      "balance_loss_mlp": 1.01917648,
      "epoch": 0.5071845127156857,
      "flos": 20340127560960.0,
      "grad_norm": 3.8504697134400057,
      "language_loss": 0.81127584,
      "learning_rate": 2.0506277485372486e-06,
      "loss": 0.8333621,
      "num_input_tokens_seen": 90980215,
      "step": 4218,
      "time_per_iteration": 2.5991621017456055
    },
    {
      "auxiliary_loss_clip": 0.01182925,
      "auxiliary_loss_mlp": 0.01027738,
      "balance_loss_clip": 1.05533195,
      "balance_loss_mlp": 1.0197686,
      "epoch": 0.5073047556063248,
      "flos": 12092955022080.0,
      "grad_norm": 1.8795622676213488,
      "language_loss": 0.67037475,
      "learning_rate": 2.04984902268902e-06,
      "loss": 0.6924814,
      "num_input_tokens_seen": 90997415,
      "step": 4219,
      "time_per_iteration": 2.597229480743408
    },
    {
      "auxiliary_loss_clip": 0.01189329,
      "auxiliary_loss_mlp": 0.01028869,
      "balance_loss_clip": 1.05075431,
      "balance_loss_mlp": 1.01963663,
      "epoch": 0.5074249984969639,
      "flos": 19682854542720.0,
      "grad_norm": 7.376170819856794,
      "language_loss": 0.75812745,
      "learning_rate": 2.0490702892787345e-06,
      "loss": 0.78030944,
      "num_input_tokens_seen": 91016475,
      "step": 4220,
      "time_per_iteration": 2.6208109855651855
    },
    {
      "auxiliary_loss_clip": 0.01176949,
      "auxiliary_loss_mlp": 0.01024794,
      "balance_loss_clip": 1.05006385,
      "balance_loss_mlp": 1.01674414,
      "epoch": 0.5075452413876029,
      "flos": 28765703975040.0,
      "grad_norm": 1.815183710695805,
      "language_loss": 0.6242075,
      "learning_rate": 2.0482915484245246e-06,
      "loss": 0.64622486,
      "num_input_tokens_seen": 91038095,
      "step": 4221,
      "time_per_iteration": 2.7218754291534424
    },
    {
      "auxiliary_loss_clip": 0.01184635,
      "auxiliary_loss_mlp": 0.01029757,
      "balance_loss_clip": 1.0479933,
      "balance_loss_mlp": 1.02100134,
      "epoch": 0.5076654842782421,
      "flos": 20339445202560.0,
      "grad_norm": 3.091089981114504,
      "language_loss": 0.84300637,
      "learning_rate": 2.047512800244526e-06,
      "loss": 0.86515027,
      "num_input_tokens_seen": 91053360,
      "step": 4222,
      "time_per_iteration": 2.6629910469055176
    },
    {
      "auxiliary_loss_clip": 0.01184792,
      "auxiliary_loss_mlp": 0.01025636,
      "balance_loss_clip": 1.05357909,
      "balance_loss_mlp": 1.01735663,
      "epoch": 0.5077857271688812,
      "flos": 26359653404160.0,
      "grad_norm": 3.576320736384139,
      "language_loss": 0.78847027,
      "learning_rate": 2.046734044856873e-06,
      "loss": 0.81057453,
      "num_input_tokens_seen": 91072770,
      "step": 4223,
      "time_per_iteration": 2.677544116973877
    },
    {
      "auxiliary_loss_clip": 0.01183133,
      "auxiliary_loss_mlp": 0.01025437,
      "balance_loss_clip": 1.05233371,
      "balance_loss_mlp": 1.01746821,
      "epoch": 0.5079059700595202,
      "flos": 21798962530560.0,
      "grad_norm": 2.187065759457972,
      "language_loss": 0.8129577,
      "learning_rate": 2.045955282379702e-06,
      "loss": 0.83504343,
      "num_input_tokens_seen": 91091430,
      "step": 4224,
      "time_per_iteration": 2.6056313514709473
    },
    {
      "auxiliary_loss_clip": 0.01182324,
      "auxiliary_loss_mlp": 0.01026327,
      "balance_loss_clip": 1.05049264,
      "balance_loss_mlp": 1.01788068,
      "epoch": 0.5080262129501594,
      "flos": 13187943175680.0,
      "grad_norm": 3.497542367160815,
      "language_loss": 0.76020753,
      "learning_rate": 2.045176512931152e-06,
      "loss": 0.78229403,
      "num_input_tokens_seen": 91106060,
      "step": 4225,
      "time_per_iteration": 2.624175548553467
    },
    {
      "auxiliary_loss_clip": 0.01191732,
      "auxiliary_loss_mlp": 0.01025205,
      "balance_loss_clip": 1.04929209,
      "balance_loss_mlp": 1.01719451,
      "epoch": 0.5081464558407984,
      "flos": 25301473712640.0,
      "grad_norm": 2.0018980688612835,
      "language_loss": 0.76046813,
      "learning_rate": 2.0443977366293604e-06,
      "loss": 0.78263748,
      "num_input_tokens_seen": 91124100,
      "step": 4226,
      "time_per_iteration": 2.7133772373199463
    },
    {
      "auxiliary_loss_clip": 0.01193051,
      "auxiliary_loss_mlp": 0.0103361,
      "balance_loss_clip": 1.04465985,
      "balance_loss_mlp": 1.02433598,
      "epoch": 0.5082666987314375,
      "flos": 30951226995840.0,
      "grad_norm": 2.470465889625455,
      "language_loss": 0.77049303,
      "learning_rate": 2.043618953592468e-06,
      "loss": 0.79275966,
      "num_input_tokens_seen": 91146555,
      "step": 4227,
      "time_per_iteration": 2.7654433250427246
    },
    {
      "auxiliary_loss_clip": 0.0118823,
      "auxiliary_loss_mlp": 0.01025959,
      "balance_loss_clip": 1.0526799,
      "balance_loss_mlp": 1.01693439,
      "epoch": 0.5083869416220766,
      "flos": 19682495406720.0,
      "grad_norm": 1.6968056392271358,
      "language_loss": 0.81220257,
      "learning_rate": 2.0428401639386144e-06,
      "loss": 0.83434445,
      "num_input_tokens_seen": 91167120,
      "step": 4228,
      "time_per_iteration": 3.5828258991241455
    },
    {
      "auxiliary_loss_clip": 0.01086158,
      "auxiliary_loss_mlp": 0.01004419,
      "balance_loss_clip": 1.01799417,
      "balance_loss_mlp": 1.00360227,
      "epoch": 0.5085071845127157,
      "flos": 71817535589760.0,
      "grad_norm": 0.8266907817854594,
      "language_loss": 0.58040905,
      "learning_rate": 2.042061367785943e-06,
      "loss": 0.6013149,
      "num_input_tokens_seen": 91220260,
      "step": 4229,
      "time_per_iteration": 3.15201735496521
    },
    {
      "auxiliary_loss_clip": 0.01193016,
      "auxiliary_loss_mlp": 0.01026567,
      "balance_loss_clip": 1.0475843,
      "balance_loss_mlp": 1.01800728,
      "epoch": 0.5086274274033548,
      "flos": 35951608252800.0,
      "grad_norm": 2.220016371819707,
      "language_loss": 0.75201666,
      "learning_rate": 2.041282565252594e-06,
      "loss": 0.77421242,
      "num_input_tokens_seen": 91240425,
      "step": 4230,
      "time_per_iteration": 2.7547364234924316
    },
    {
      "auxiliary_loss_clip": 0.01189957,
      "auxiliary_loss_mlp": 0.01025861,
      "balance_loss_clip": 1.04874098,
      "balance_loss_mlp": 1.01759362,
      "epoch": 0.5087476702939938,
      "flos": 23513732881920.0,
      "grad_norm": 2.6497868833440017,
      "language_loss": 0.77232075,
      "learning_rate": 2.040503756456714e-06,
      "loss": 0.79447889,
      "num_input_tokens_seen": 91259635,
      "step": 4231,
      "time_per_iteration": 3.6051371097564697
    },
    {
      "auxiliary_loss_clip": 0.0117849,
      "auxiliary_loss_mlp": 0.01027034,
      "balance_loss_clip": 1.05002975,
      "balance_loss_mlp": 1.0182724,
      "epoch": 0.508867913184633,
      "flos": 15122091841920.0,
      "grad_norm": 1.8694947253688126,
      "language_loss": 0.78973126,
      "learning_rate": 2.0397249415164456e-06,
      "loss": 0.81178653,
      "num_input_tokens_seen": 91276990,
      "step": 4232,
      "time_per_iteration": 3.4581165313720703
    },
    {
      "auxiliary_loss_clip": 0.0118085,
      "auxiliary_loss_mlp": 0.01026964,
      "balance_loss_clip": 1.04787457,
      "balance_loss_mlp": 1.01854181,
      "epoch": 0.508988156075272,
      "flos": 25885309374720.0,
      "grad_norm": 1.8162729206454513,
      "language_loss": 0.8066057,
      "learning_rate": 2.0389461205499354e-06,
      "loss": 0.82868385,
      "num_input_tokens_seen": 91296125,
      "step": 4233,
      "time_per_iteration": 2.689711570739746
    },
    {
      "auxiliary_loss_clip": 0.01192107,
      "auxiliary_loss_mlp": 0.01027447,
      "balance_loss_clip": 1.04947293,
      "balance_loss_mlp": 1.0191319,
      "epoch": 0.5091083989659111,
      "flos": 13844857057920.0,
      "grad_norm": 2.035124999041599,
      "language_loss": 0.73773444,
      "learning_rate": 2.03816729367533e-06,
      "loss": 0.7599299,
      "num_input_tokens_seen": 91314280,
      "step": 4234,
      "time_per_iteration": 2.779888391494751
    },
    {
      "auxiliary_loss_clip": 0.01188398,
      "auxiliary_loss_mlp": 0.01031751,
      "balance_loss_clip": 1.0520184,
      "balance_loss_mlp": 1.02324581,
      "epoch": 0.5092286418565503,
      "flos": 21104881050240.0,
      "grad_norm": 3.0222969560330584,
      "language_loss": 0.7126593,
      "learning_rate": 2.0373884610107765e-06,
      "loss": 0.73486078,
      "num_input_tokens_seen": 91334595,
      "step": 4235,
      "time_per_iteration": 2.6771416664123535
    },
    {
      "auxiliary_loss_clip": 0.01186516,
      "auxiliary_loss_mlp": 0.01026168,
      "balance_loss_clip": 1.04977036,
      "balance_loss_mlp": 1.0178709,
      "epoch": 0.5093488847471893,
      "flos": 18621298972800.0,
      "grad_norm": 2.4435124406552386,
      "language_loss": 0.697833,
      "learning_rate": 2.0366096226744225e-06,
      "loss": 0.7199598,
      "num_input_tokens_seen": 91349790,
      "step": 4236,
      "time_per_iteration": 2.5551929473876953
    },
    {
      "auxiliary_loss_clip": 0.01176104,
      "auxiliary_loss_mlp": 0.01036055,
      "balance_loss_clip": 1.05086052,
      "balance_loss_mlp": 1.02734089,
      "epoch": 0.5094691276378284,
      "flos": 23803783205760.0,
      "grad_norm": 1.7847107391128967,
      "language_loss": 0.76765645,
      "learning_rate": 2.035830778784418e-06,
      "loss": 0.78977799,
      "num_input_tokens_seen": 91370465,
      "step": 4237,
      "time_per_iteration": 2.6519083976745605
    },
    {
      "auxiliary_loss_clip": 0.01190197,
      "auxiliary_loss_mlp": 0.01026377,
      "balance_loss_clip": 1.05364275,
      "balance_loss_mlp": 1.01742423,
      "epoch": 0.5095893705284675,
      "flos": 17420410546560.0,
      "grad_norm": 3.1106983295518176,
      "language_loss": 0.80141324,
      "learning_rate": 2.0350519294589134e-06,
      "loss": 0.82357895,
      "num_input_tokens_seen": 91388505,
      "step": 4238,
      "time_per_iteration": 3.595844030380249
    },
    {
      "auxiliary_loss_clip": 0.01185413,
      "auxiliary_loss_mlp": 0.01022595,
      "balance_loss_clip": 1.04497337,
      "balance_loss_mlp": 1.01364219,
      "epoch": 0.5097096134191066,
      "flos": 25849362839040.0,
      "grad_norm": 3.196633448965855,
      "language_loss": 0.82791811,
      "learning_rate": 2.0342730748160588e-06,
      "loss": 0.84999818,
      "num_input_tokens_seen": 91408970,
      "step": 4239,
      "time_per_iteration": 2.739531993865967
    },
    {
      "auxiliary_loss_clip": 0.01183657,
      "auxiliary_loss_mlp": 0.01027137,
      "balance_loss_clip": 1.04712296,
      "balance_loss_mlp": 1.01892948,
      "epoch": 0.5098298563097456,
      "flos": 27745122844800.0,
      "grad_norm": 3.21063335469203,
      "language_loss": 0.70336902,
      "learning_rate": 2.033494214974006e-06,
      "loss": 0.72547698,
      "num_input_tokens_seen": 91430115,
      "step": 4240,
      "time_per_iteration": 2.709540605545044
    },
    {
      "auxiliary_loss_clip": 0.01177882,
      "auxiliary_loss_mlp": 0.0102643,
      "balance_loss_clip": 1.0507772,
      "balance_loss_mlp": 1.01840103,
      "epoch": 0.5099500992003848,
      "flos": 21358913011200.0,
      "grad_norm": 1.7160039988477498,
      "language_loss": 0.83888769,
      "learning_rate": 2.0327153500509067e-06,
      "loss": 0.8609308,
      "num_input_tokens_seen": 91449140,
      "step": 4241,
      "time_per_iteration": 2.632646322250366
    },
    {
      "auxiliary_loss_clip": 0.011893,
      "auxiliary_loss_mlp": 0.01026008,
      "balance_loss_clip": 1.05241895,
      "balance_loss_mlp": 1.01778841,
      "epoch": 0.5100703420910239,
      "flos": 19865999013120.0,
      "grad_norm": 1.9506380832800263,
      "language_loss": 0.84491825,
      "learning_rate": 2.031936480164916e-06,
      "loss": 0.86707139,
      "num_input_tokens_seen": 91466880,
      "step": 4242,
      "time_per_iteration": 2.6468935012817383
    },
    {
      "auxiliary_loss_clip": 0.01183263,
      "auxiliary_loss_mlp": 0.01028057,
      "balance_loss_clip": 1.05269623,
      "balance_loss_mlp": 1.01943266,
      "epoch": 0.5101905849816629,
      "flos": 24648797635200.0,
      "grad_norm": 1.8312642907077084,
      "language_loss": 0.80334449,
      "learning_rate": 2.0311576054341857e-06,
      "loss": 0.82545769,
      "num_input_tokens_seen": 91487495,
      "step": 4243,
      "time_per_iteration": 2.662973403930664
    },
    {
      "auxiliary_loss_clip": 0.01183715,
      "auxiliary_loss_mlp": 0.01026769,
      "balance_loss_clip": 1.05547678,
      "balance_loss_mlp": 1.01803708,
      "epoch": 0.5103108278723021,
      "flos": 22930076787840.0,
      "grad_norm": 1.5394406974622101,
      "language_loss": 0.62744153,
      "learning_rate": 2.0303787259768715e-06,
      "loss": 0.64954633,
      "num_input_tokens_seen": 91508395,
      "step": 4244,
      "time_per_iteration": 2.656352996826172
    },
    {
      "auxiliary_loss_clip": 0.01190761,
      "auxiliary_loss_mlp": 0.0102964,
      "balance_loss_clip": 1.0542872,
      "balance_loss_mlp": 1.02124715,
      "epoch": 0.5104310707629411,
      "flos": 21506613736320.0,
      "grad_norm": 2.506120977641466,
      "language_loss": 0.68826592,
      "learning_rate": 2.0295998419111294e-06,
      "loss": 0.71046996,
      "num_input_tokens_seen": 91525685,
      "step": 4245,
      "time_per_iteration": 2.6296589374542236
    },
    {
      "auxiliary_loss_clip": 0.01196184,
      "auxiliary_loss_mlp": 0.01030573,
      "balance_loss_clip": 1.04482996,
      "balance_loss_mlp": 1.0217154,
      "epoch": 0.5105513136535802,
      "flos": 14903180403840.0,
      "grad_norm": 2.80688754247511,
      "language_loss": 0.73882133,
      "learning_rate": 2.028820953355115e-06,
      "loss": 0.76108885,
      "num_input_tokens_seen": 91543785,
      "step": 4246,
      "time_per_iteration": 2.7528278827667236
    },
    {
      "auxiliary_loss_clip": 0.01193441,
      "auxiliary_loss_mlp": 0.01026399,
      "balance_loss_clip": 1.04940653,
      "balance_loss_mlp": 1.01747024,
      "epoch": 0.5106715565442194,
      "flos": 22602212421120.0,
      "grad_norm": 1.7988161235201197,
      "language_loss": 0.79259688,
      "learning_rate": 2.0280420604269834e-06,
      "loss": 0.81479526,
      "num_input_tokens_seen": 91563325,
      "step": 4247,
      "time_per_iteration": 2.6621100902557373
    },
    {
      "auxiliary_loss_clip": 0.01081188,
      "auxiliary_loss_mlp": 0.01000821,
      "balance_loss_clip": 1.0157361,
      "balance_loss_mlp": 0.99999207,
      "epoch": 0.5107917994348584,
      "flos": 71027645558400.0,
      "grad_norm": 0.7044779700046397,
      "language_loss": 0.58937716,
      "learning_rate": 2.027263163244895e-06,
      "loss": 0.61019725,
      "num_input_tokens_seen": 91632450,
      "step": 4248,
      "time_per_iteration": 3.320021629333496
    },
    {
      "auxiliary_loss_clip": 0.01183588,
      "auxiliary_loss_mlp": 0.01034164,
      "balance_loss_clip": 1.05315781,
      "balance_loss_mlp": 1.02608132,
      "epoch": 0.5109120423254975,
      "flos": 24827416992000.0,
      "grad_norm": 2.484591540697633,
      "language_loss": 0.74737442,
      "learning_rate": 2.026484261927005e-06,
      "loss": 0.76955187,
      "num_input_tokens_seen": 91651945,
      "step": 4249,
      "time_per_iteration": 2.701749801635742
    },
    {
      "auxiliary_loss_clip": 0.01194379,
      "auxiliary_loss_mlp": 0.0102679,
      "balance_loss_clip": 1.05730343,
      "balance_loss_mlp": 1.01746225,
      "epoch": 0.5110322852161366,
      "flos": 21247661612160.0,
      "grad_norm": 2.4338251727800264,
      "language_loss": 0.73971486,
      "learning_rate": 2.025705356591475e-06,
      "loss": 0.76192653,
      "num_input_tokens_seen": 91669635,
      "step": 4250,
      "time_per_iteration": 2.664006471633911
    },
    {
      "auxiliary_loss_clip": 0.01092014,
      "auxiliary_loss_mlp": 0.01199415,
      "balance_loss_clip": 1.01626182,
      "balance_loss_mlp": 1.00018394,
      "epoch": 0.5111525281067757,
      "flos": 66457114358400.0,
      "grad_norm": 0.7583828442545729,
      "language_loss": 0.57929367,
      "learning_rate": 2.024926447356462e-06,
      "loss": 0.60220802,
      "num_input_tokens_seen": 91731920,
      "step": 4251,
      "time_per_iteration": 3.1423258781433105
    },
    {
      "auxiliary_loss_clip": 0.01183851,
      "auxiliary_loss_mlp": 0.01031434,
      "balance_loss_clip": 1.05254853,
      "balance_loss_mlp": 1.02179623,
      "epoch": 0.5112727709974147,
      "flos": 14866731077760.0,
      "grad_norm": 2.18716002644584,
      "language_loss": 0.78986287,
      "learning_rate": 2.024147534340127e-06,
      "loss": 0.81201577,
      "num_input_tokens_seen": 91749780,
      "step": 4252,
      "time_per_iteration": 2.6275110244750977
    },
    {
      "auxiliary_loss_clip": 0.01182456,
      "auxiliary_loss_mlp": 0.01025709,
      "balance_loss_clip": 1.04647255,
      "balance_loss_mlp": 1.01738191,
      "epoch": 0.5113930138880539,
      "flos": 21177600134400.0,
      "grad_norm": 1.9069345735557255,
      "language_loss": 0.79929501,
      "learning_rate": 2.02336861766063e-06,
      "loss": 0.82137668,
      "num_input_tokens_seen": 91768840,
      "step": 4253,
      "time_per_iteration": 2.6829330921173096
    },
    {
      "auxiliary_loss_clip": 0.01193263,
      "auxiliary_loss_mlp": 0.01028621,
      "balance_loss_clip": 1.05453539,
      "balance_loss_mlp": 1.01988316,
      "epoch": 0.511513256778693,
      "flos": 20409111630720.0,
      "grad_norm": 21.712744832271348,
      "language_loss": 0.78886104,
      "learning_rate": 2.0225896974361327e-06,
      "loss": 0.81107998,
      "num_input_tokens_seen": 91788945,
      "step": 4254,
      "time_per_iteration": 2.714338541030884
    },
    {
      "auxiliary_loss_clip": 0.01094931,
      "auxiliary_loss_mlp": 0.0100195,
      "balance_loss_clip": 1.01507556,
      "balance_loss_mlp": 1.00108016,
      "epoch": 0.511633499669332,
      "flos": 69879975131520.0,
      "grad_norm": 0.852208898018359,
      "language_loss": 0.59962118,
      "learning_rate": 2.0218107737847962e-06,
      "loss": 0.62058997,
      "num_input_tokens_seen": 91850990,
      "step": 4255,
      "time_per_iteration": 4.097308397293091
    },
    {
      "auxiliary_loss_clip": 0.01184777,
      "auxiliary_loss_mlp": 0.01031364,
      "balance_loss_clip": 1.05639505,
      "balance_loss_mlp": 1.02266192,
      "epoch": 0.5117537425599712,
      "flos": 24097855852800.0,
      "grad_norm": 2.6662329160864826,
      "language_loss": 0.74952972,
      "learning_rate": 2.0210318468247826e-06,
      "loss": 0.77169114,
      "num_input_tokens_seen": 91869960,
      "step": 4256,
      "time_per_iteration": 2.632931709289551
    },
    {
      "auxiliary_loss_clip": 0.01186249,
      "auxiliary_loss_mlp": 0.01023734,
      "balance_loss_clip": 1.04809642,
      "balance_loss_mlp": 1.01580667,
      "epoch": 0.5118739854506102,
      "flos": 20959550622720.0,
      "grad_norm": 3.6389076612541422,
      "language_loss": 0.81867242,
      "learning_rate": 2.020252916674255e-06,
      "loss": 0.84077221,
      "num_input_tokens_seen": 91889075,
      "step": 4257,
      "time_per_iteration": 3.569154739379883
    },
    {
      "auxiliary_loss_clip": 0.01185501,
      "auxiliary_loss_mlp": 0.01027319,
      "balance_loss_clip": 1.04982638,
      "balance_loss_mlp": 1.01822376,
      "epoch": 0.5119942283412493,
      "flos": 17457326749440.0,
      "grad_norm": 2.442737803567726,
      "language_loss": 0.8118974,
      "learning_rate": 2.019473983451375e-06,
      "loss": 0.83402562,
      "num_input_tokens_seen": 91907495,
      "step": 4258,
      "time_per_iteration": 2.6321516036987305
    },
    {
      "auxiliary_loss_clip": 0.01195442,
      "auxiliary_loss_mlp": 0.01028221,
      "balance_loss_clip": 1.04842734,
      "balance_loss_mlp": 1.01964974,
      "epoch": 0.5121144712318885,
      "flos": 21066743784960.0,
      "grad_norm": 1.8135218081474742,
      "language_loss": 0.71574914,
      "learning_rate": 2.0186950472743076e-06,
      "loss": 0.73798573,
      "num_input_tokens_seen": 91927400,
      "step": 4259,
      "time_per_iteration": 3.7017507553100586
    },
    {
      "auxiliary_loss_clip": 0.01183609,
      "auxiliary_loss_mlp": 0.01027535,
      "balance_loss_clip": 1.05419111,
      "balance_loss_mlp": 1.01902938,
      "epoch": 0.5122347141225275,
      "flos": 19860791541120.0,
      "grad_norm": 1.8377437703198374,
      "language_loss": 0.74197751,
      "learning_rate": 2.0179161082612162e-06,
      "loss": 0.76408899,
      "num_input_tokens_seen": 91946790,
      "step": 4260,
      "time_per_iteration": 2.5680091381073
    },
    {
      "auxiliary_loss_clip": 0.01181519,
      "auxiliary_loss_mlp": 0.01026201,
      "balance_loss_clip": 1.04730976,
      "balance_loss_mlp": 1.01792192,
      "epoch": 0.5123549570131666,
      "flos": 22528487756160.0,
      "grad_norm": 2.4162143759856507,
      "language_loss": 0.72917771,
      "learning_rate": 2.017137166530266e-06,
      "loss": 0.75125492,
      "num_input_tokens_seen": 91966325,
      "step": 4261,
      "time_per_iteration": 2.662541151046753
    },
    {
      "auxiliary_loss_clip": 0.01189554,
      "auxiliary_loss_mlp": 0.01025476,
      "balance_loss_clip": 1.04944658,
      "balance_loss_mlp": 1.01755214,
      "epoch": 0.5124751999038056,
      "flos": 20333375804160.0,
      "grad_norm": 2.04429305193311,
      "language_loss": 0.80213594,
      "learning_rate": 2.0163582221996213e-06,
      "loss": 0.82428634,
      "num_input_tokens_seen": 91984700,
      "step": 4262,
      "time_per_iteration": 2.6543450355529785
    },
    {
      "auxiliary_loss_clip": 0.01188174,
      "auxiliary_loss_mlp": 0.01028136,
      "balance_loss_clip": 1.05128431,
      "balance_loss_mlp": 1.01961279,
      "epoch": 0.5125954427944448,
      "flos": 39785970211200.0,
      "grad_norm": 3.083705893861122,
      "language_loss": 0.68192649,
      "learning_rate": 2.015579275387446e-06,
      "loss": 0.70408964,
      "num_input_tokens_seen": 92010020,
      "step": 4263,
      "time_per_iteration": 2.739715814590454
    },
    {
      "auxiliary_loss_clip": 0.01179023,
      "auxiliary_loss_mlp": 0.01029858,
      "balance_loss_clip": 1.05012155,
      "balance_loss_mlp": 1.02140021,
      "epoch": 0.5127156856850839,
      "flos": 29205394358400.0,
      "grad_norm": 2.113609306849704,
      "language_loss": 0.68294376,
      "learning_rate": 2.0148003262119085e-06,
      "loss": 0.70503259,
      "num_input_tokens_seen": 92030990,
      "step": 4264,
      "time_per_iteration": 2.6551449298858643
    },
    {
      "auxiliary_loss_clip": 0.01187392,
      "auxiliary_loss_mlp": 0.0102743,
      "balance_loss_clip": 1.04796791,
      "balance_loss_mlp": 1.01864481,
      "epoch": 0.5128359285757229,
      "flos": 13553693412480.0,
      "grad_norm": 2.074722879966189,
      "language_loss": 0.76513988,
      "learning_rate": 2.0140213747911728e-06,
      "loss": 0.78728813,
      "num_input_tokens_seen": 92049525,
      "step": 4265,
      "time_per_iteration": 3.571049690246582
    },
    {
      "auxiliary_loss_clip": 0.01190281,
      "auxiliary_loss_mlp": 0.01034564,
      "balance_loss_clip": 1.05246687,
      "balance_loss_mlp": 1.02543819,
      "epoch": 0.5129561714663621,
      "flos": 25192089820800.0,
      "grad_norm": 1.8767679418774008,
      "language_loss": 0.80698144,
      "learning_rate": 2.013242421243406e-06,
      "loss": 0.82922995,
      "num_input_tokens_seen": 92068430,
      "step": 4266,
      "time_per_iteration": 2.718743324279785
    },
    {
      "auxiliary_loss_clip": 0.01192588,
      "auxiliary_loss_mlp": 0.01022895,
      "balance_loss_clip": 1.04936838,
      "balance_loss_mlp": 1.01439512,
      "epoch": 0.5130764143570011,
      "flos": 18150223080960.0,
      "grad_norm": 1.6007823701919792,
      "language_loss": 0.79384619,
      "learning_rate": 2.012463465686774e-06,
      "loss": 0.81600106,
      "num_input_tokens_seen": 92088180,
      "step": 4267,
      "time_per_iteration": 2.6452784538269043
    },
    {
      "auxiliary_loss_clip": 0.01101829,
      "auxiliary_loss_mlp": 0.01004444,
      "balance_loss_clip": 1.01969814,
      "balance_loss_mlp": 1.00364494,
      "epoch": 0.5131966572476402,
      "flos": 59794896418560.0,
      "grad_norm": 0.7758686886508049,
      "language_loss": 0.54974139,
      "learning_rate": 2.0116845082394446e-06,
      "loss": 0.57080406,
      "num_input_tokens_seen": 92153015,
      "step": 4268,
      "time_per_iteration": 3.247450351715088
    },
    {
      "auxiliary_loss_clip": 0.01190154,
      "auxiliary_loss_mlp": 0.01026896,
      "balance_loss_clip": 1.05308533,
      "balance_loss_mlp": 1.01794338,
      "epoch": 0.5133169001382794,
      "flos": 18515219132160.0,
      "grad_norm": 2.3893841244686818,
      "language_loss": 0.78658617,
      "learning_rate": 2.0109055490195836e-06,
      "loss": 0.80875659,
      "num_input_tokens_seen": 92171470,
      "step": 4269,
      "time_per_iteration": 2.6113779544830322
    },
    {
      "auxiliary_loss_clip": 0.01198413,
      "auxiliary_loss_mlp": 0.01027513,
      "balance_loss_clip": 1.04144681,
      "balance_loss_mlp": 1.01924002,
      "epoch": 0.5134371430289184,
      "flos": 15523537219200.0,
      "grad_norm": 2.0057138535168204,
      "language_loss": 0.6474793,
      "learning_rate": 2.0101265881453605e-06,
      "loss": 0.66973853,
      "num_input_tokens_seen": 92189945,
      "step": 4270,
      "time_per_iteration": 2.6972339153289795
    },
    {
      "auxiliary_loss_clip": 0.01182948,
      "auxiliary_loss_mlp": 0.01033913,
      "balance_loss_clip": 1.05272245,
      "balance_loss_mlp": 1.02581239,
      "epoch": 0.5135573859195575,
      "flos": 21433786911360.0,
      "grad_norm": 1.9316220066804777,
      "language_loss": 0.78387541,
      "learning_rate": 2.009347625734941e-06,
      "loss": 0.8060441,
      "num_input_tokens_seen": 92209855,
      "step": 4271,
      "time_per_iteration": 2.6329424381256104
    },
    {
      "auxiliary_loss_clip": 0.01187095,
      "auxiliary_loss_mlp": 0.01027692,
      "balance_loss_clip": 1.05745101,
      "balance_loss_mlp": 1.01930022,
      "epoch": 0.5136776288101966,
      "flos": 17712651600000.0,
      "grad_norm": 3.4111722984759396,
      "language_loss": 0.75288284,
      "learning_rate": 2.0085686619064954e-06,
      "loss": 0.77503073,
      "num_input_tokens_seen": 92226295,
      "step": 4272,
      "time_per_iteration": 2.55787992477417
    },
    {
      "auxiliary_loss_clip": 0.0118916,
      "auxiliary_loss_mlp": 0.01037422,
      "balance_loss_clip": 1.054986,
      "balance_loss_mlp": 1.02891016,
      "epoch": 0.5137978717008357,
      "flos": 16581680997120.0,
      "grad_norm": 2.0339184731930877,
      "language_loss": 0.82648289,
      "learning_rate": 2.00778969677819e-06,
      "loss": 0.84874868,
      "num_input_tokens_seen": 92243330,
      "step": 4273,
      "time_per_iteration": 2.7351882457733154
    },
    {
      "auxiliary_loss_clip": 0.01183284,
      "auxiliary_loss_mlp": 0.01025333,
      "balance_loss_clip": 1.04838789,
      "balance_loss_mlp": 1.01706553,
      "epoch": 0.5139181145914747,
      "flos": 20668243322880.0,
      "grad_norm": 1.953148602713094,
      "language_loss": 0.64123499,
      "learning_rate": 2.0070107304681934e-06,
      "loss": 0.6633212,
      "num_input_tokens_seen": 92262285,
      "step": 4274,
      "time_per_iteration": 2.707871913909912
    },
    {
      "auxiliary_loss_clip": 0.01188278,
      "auxiliary_loss_mlp": 0.01025355,
      "balance_loss_clip": 1.05152988,
      "balance_loss_mlp": 1.01699817,
      "epoch": 0.5140383574821139,
      "flos": 32926996546560.0,
      "grad_norm": 2.3005309269703287,
      "language_loss": 0.78253919,
      "learning_rate": 2.006231763094675e-06,
      "loss": 0.80467552,
      "num_input_tokens_seen": 92283305,
      "step": 4275,
      "time_per_iteration": 2.7484776973724365
    },
    {
      "auxiliary_loss_clip": 0.01180159,
      "auxiliary_loss_mlp": 0.01025469,
      "balance_loss_clip": 1.05203962,
      "balance_loss_mlp": 1.01740479,
      "epoch": 0.514158600372753,
      "flos": 19537093152000.0,
      "grad_norm": 2.1921449330284224,
      "language_loss": 0.87624466,
      "learning_rate": 2.0054527947758027e-06,
      "loss": 0.89830095,
      "num_input_tokens_seen": 92302105,
      "step": 4276,
      "time_per_iteration": 2.6508612632751465
    },
    {
      "auxiliary_loss_clip": 0.01082809,
      "auxiliary_loss_mlp": 0.01000763,
      "balance_loss_clip": 1.01725388,
      "balance_loss_mlp": 1.00008953,
      "epoch": 0.514278843263392,
      "flos": 62523855279360.0,
      "grad_norm": 0.7212764245483181,
      "language_loss": 0.5583359,
      "learning_rate": 2.004673825629746e-06,
      "loss": 0.57917166,
      "num_input_tokens_seen": 92362885,
      "step": 4277,
      "time_per_iteration": 3.179435968399048
    },
    {
      "auxiliary_loss_clip": 0.01181387,
      "auxiliary_loss_mlp": 0.01028703,
      "balance_loss_clip": 1.04929185,
      "balance_loss_mlp": 1.02070355,
      "epoch": 0.5143990861540312,
      "flos": 25882328545920.0,
      "grad_norm": 1.846087083354751,
      "language_loss": 0.72601652,
      "learning_rate": 2.0038948557746744e-06,
      "loss": 0.74811745,
      "num_input_tokens_seen": 92384740,
      "step": 4278,
      "time_per_iteration": 2.713042974472046
    },
    {
      "auxiliary_loss_clip": 0.01181333,
      "auxiliary_loss_mlp": 0.01028222,
      "balance_loss_clip": 1.05273604,
      "balance_loss_mlp": 1.02015746,
      "epoch": 0.5145193290446702,
      "flos": 23330660238720.0,
      "grad_norm": 1.7878400502128011,
      "language_loss": 0.75146925,
      "learning_rate": 2.0031158853287558e-06,
      "loss": 0.77356488,
      "num_input_tokens_seen": 92405175,
      "step": 4279,
      "time_per_iteration": 2.6952078342437744
    },
    {
      "auxiliary_loss_clip": 0.01187558,
      "auxiliary_loss_mlp": 0.01028685,
      "balance_loss_clip": 1.05410266,
      "balance_loss_mlp": 1.02048945,
      "epoch": 0.5146395719353093,
      "flos": 22856603518080.0,
      "grad_norm": 1.9791250807388308,
      "language_loss": 0.7022723,
      "learning_rate": 2.0023369144101593e-06,
      "loss": 0.72443473,
      "num_input_tokens_seen": 92423345,
      "step": 4280,
      "time_per_iteration": 2.7158665657043457
    },
    {
      "auxiliary_loss_clip": 0.01178721,
      "auxiliary_loss_mlp": 0.01033509,
      "balance_loss_clip": 1.04902816,
      "balance_loss_mlp": 1.0251708,
      "epoch": 0.5147598148259485,
      "flos": 26391577616640.0,
      "grad_norm": 1.8311276360799982,
      "language_loss": 0.77029932,
      "learning_rate": 2.0015579431370555e-06,
      "loss": 0.79242158,
      "num_input_tokens_seen": 92445025,
      "step": 4281,
      "time_per_iteration": 3.680527448654175
    },
    {
      "auxiliary_loss_clip": 0.01181913,
      "auxiliary_loss_mlp": 0.01027446,
      "balance_loss_clip": 1.05300367,
      "balance_loss_mlp": 1.01893425,
      "epoch": 0.5148800577165875,
      "flos": 29965694561280.0,
      "grad_norm": 2.2924288767277177,
      "language_loss": 0.69993055,
      "learning_rate": 2.000778971627612e-06,
      "loss": 0.72202414,
      "num_input_tokens_seen": 92464490,
      "step": 4282,
      "time_per_iteration": 2.6905221939086914
    },
    {
      "auxiliary_loss_clip": 0.01179232,
      "auxiliary_loss_mlp": 0.01033632,
      "balance_loss_clip": 1.04855251,
      "balance_loss_mlp": 1.02532935,
      "epoch": 0.5150003006072266,
      "flos": 17931383470080.0,
      "grad_norm": 1.8782876196916864,
      "language_loss": 0.89879078,
      "learning_rate": 2e-06,
      "loss": 0.92091942,
      "num_input_tokens_seen": 92482085,
      "step": 4283,
      "time_per_iteration": 2.565143585205078
    },
    {
      "auxiliary_loss_clip": 0.01180558,
      "auxiliary_loss_mlp": 0.01029706,
      "balance_loss_clip": 1.05463028,
      "balance_loss_mlp": 1.02171254,
      "epoch": 0.5151205434978657,
      "flos": 18478733892480.0,
      "grad_norm": 3.969744864280259,
      "language_loss": 0.85630536,
      "learning_rate": 1.9992210283723878e-06,
      "loss": 0.87840801,
      "num_input_tokens_seen": 92499325,
      "step": 4284,
      "time_per_iteration": 3.5729238986968994
    },
    {
      "auxiliary_loss_clip": 0.01180775,
      "auxiliary_loss_mlp": 0.0103066,
      "balance_loss_clip": 1.05559754,
      "balance_loss_mlp": 1.02269709,
      "epoch": 0.5152407863885048,
      "flos": 25341263003520.0,
      "grad_norm": 1.696212759506626,
      "language_loss": 0.79161173,
      "learning_rate": 1.9984420568629448e-06,
      "loss": 0.81372607,
      "num_input_tokens_seen": 92522090,
      "step": 4285,
      "time_per_iteration": 2.603364944458008
    },
    {
      "auxiliary_loss_clip": 0.01184947,
      "auxiliary_loss_mlp": 0.01026366,
      "balance_loss_clip": 1.05271327,
      "balance_loss_mlp": 1.01877236,
      "epoch": 0.5153610292791438,
      "flos": 18329740277760.0,
      "grad_norm": 2.031370402232659,
      "language_loss": 0.78712964,
      "learning_rate": 1.9976630855898405e-06,
      "loss": 0.80924273,
      "num_input_tokens_seen": 92539845,
      "step": 4286,
      "time_per_iteration": 3.508326292037964
    },
    {
      "auxiliary_loss_clip": 0.0118031,
      "auxiliary_loss_mlp": 0.01025036,
      "balance_loss_clip": 1.0460465,
      "balance_loss_mlp": 1.01729977,
      "epoch": 0.515481272169783,
      "flos": 30409945971840.0,
      "grad_norm": 2.026072699882802,
      "language_loss": 0.74783421,
      "learning_rate": 1.9968841146712445e-06,
      "loss": 0.76988769,
      "num_input_tokens_seen": 92559460,
      "step": 4287,
      "time_per_iteration": 2.757277011871338
    },
    {
      "auxiliary_loss_clip": 0.01189418,
      "auxiliary_loss_mlp": 0.01207761,
      "balance_loss_clip": 1.04336095,
      "balance_loss_mlp": 1.00066853,
      "epoch": 0.5156015150604221,
      "flos": 23037305863680.0,
      "grad_norm": 1.6230671409488127,
      "language_loss": 0.71714896,
      "learning_rate": 1.996105144225326e-06,
      "loss": 0.74112076,
      "num_input_tokens_seen": 92579695,
      "step": 4288,
      "time_per_iteration": 2.6820502281188965
    },
    {
      "auxiliary_loss_clip": 0.01184678,
      "auxiliary_loss_mlp": 0.01028755,
      "balance_loss_clip": 1.05425286,
      "balance_loss_mlp": 1.02111959,
      "epoch": 0.5157217579510611,
      "flos": 17858556645120.0,
      "grad_norm": 2.3617888803989233,
      "language_loss": 0.78528029,
      "learning_rate": 1.995326174370254e-06,
      "loss": 0.80741459,
      "num_input_tokens_seen": 92598795,
      "step": 4289,
      "time_per_iteration": 2.593388080596924
    },
    {
      "auxiliary_loss_clip": 0.01180693,
      "auxiliary_loss_mlp": 0.01208091,
      "balance_loss_clip": 1.05192804,
      "balance_loss_mlp": 1.00062764,
      "epoch": 0.5158420008417003,
      "flos": 19171486569600.0,
      "grad_norm": 1.6403721790712709,
      "language_loss": 0.73230171,
      "learning_rate": 1.994547205224197e-06,
      "loss": 0.75618958,
      "num_input_tokens_seen": 92617700,
      "step": 4290,
      "time_per_iteration": 2.580857276916504
    },
    {
      "auxiliary_loss_clip": 0.01178814,
      "auxiliary_loss_mlp": 0.01023729,
      "balance_loss_clip": 1.05000639,
      "balance_loss_mlp": 1.01512194,
      "epoch": 0.5159622437323393,
      "flos": 22419534827520.0,
      "grad_norm": 2.27109718706052,
      "language_loss": 0.67779177,
      "learning_rate": 1.993768236905325e-06,
      "loss": 0.69981718,
      "num_input_tokens_seen": 92638370,
      "step": 4291,
      "time_per_iteration": 3.584606409072876
    },
    {
      "auxiliary_loss_clip": 0.01180879,
      "auxiliary_loss_mlp": 0.01024062,
      "balance_loss_clip": 1.04867423,
      "balance_loss_mlp": 1.01536608,
      "epoch": 0.5160824866229784,
      "flos": 24603010773120.0,
      "grad_norm": 2.5934695871475704,
      "language_loss": 0.66036284,
      "learning_rate": 1.992989269531807e-06,
      "loss": 0.68241227,
      "num_input_tokens_seen": 92657180,
      "step": 4292,
      "time_per_iteration": 2.620120048522949
    },
    {
      "auxiliary_loss_clip": 0.0118881,
      "auxiliary_loss_mlp": 0.01025116,
      "balance_loss_clip": 1.05160296,
      "balance_loss_mlp": 1.01697969,
      "epoch": 0.5162027295136175,
      "flos": 18002737837440.0,
      "grad_norm": 3.0128608536223593,
      "language_loss": 0.68475711,
      "learning_rate": 1.99221030322181e-06,
      "loss": 0.70689642,
      "num_input_tokens_seen": 92673985,
      "step": 4293,
      "time_per_iteration": 2.601839542388916
    },
    {
      "auxiliary_loss_clip": 0.01189887,
      "auxiliary_loss_mlp": 0.01029239,
      "balance_loss_clip": 1.05146253,
      "balance_loss_mlp": 1.02141309,
      "epoch": 0.5163229724042566,
      "flos": 27344611221120.0,
      "grad_norm": 2.1944290921005023,
      "language_loss": 0.81099409,
      "learning_rate": 1.991431338093505e-06,
      "loss": 0.83318532,
      "num_input_tokens_seen": 92696340,
      "step": 4294,
      "time_per_iteration": 2.7281064987182617
    },
    {
      "auxiliary_loss_clip": 0.01186334,
      "auxiliary_loss_mlp": 0.01031818,
      "balance_loss_clip": 1.0540874,
      "balance_loss_mlp": 1.0242331,
      "epoch": 0.5164432152948957,
      "flos": 21762764599680.0,
      "grad_norm": 2.2962218563626497,
      "language_loss": 0.79525685,
      "learning_rate": 1.9906523742650587e-06,
      "loss": 0.81743836,
      "num_input_tokens_seen": 92715200,
      "step": 4295,
      "time_per_iteration": 2.655818462371826
    },
    {
      "auxiliary_loss_clip": 0.01181413,
      "auxiliary_loss_mlp": 0.01031488,
      "balance_loss_clip": 1.05200648,
      "balance_loss_mlp": 1.02246952,
      "epoch": 0.5165634581855347,
      "flos": 25550334115200.0,
      "grad_norm": 2.1579171810703404,
      "language_loss": 0.7719503,
      "learning_rate": 1.9898734118546397e-06,
      "loss": 0.7940793,
      "num_input_tokens_seen": 92735150,
      "step": 4296,
      "time_per_iteration": 2.5997700691223145
    },
    {
      "auxiliary_loss_clip": 0.01204141,
      "auxiliary_loss_mlp": 0.01027004,
      "balance_loss_clip": 1.04882205,
      "balance_loss_mlp": 1.01792002,
      "epoch": 0.5166837010761739,
      "flos": 19901191363200.0,
      "grad_norm": 1.7848432051588456,
      "language_loss": 0.80505896,
      "learning_rate": 1.989094450980416e-06,
      "loss": 0.82737041,
      "num_input_tokens_seen": 92755250,
      "step": 4297,
      "time_per_iteration": 2.699573278427124
    },
    {
      "auxiliary_loss_clip": 0.01180932,
      "auxiliary_loss_mlp": 0.01021242,
      "balance_loss_clip": 1.05078053,
      "balance_loss_mlp": 1.01330256,
      "epoch": 0.516803943966813,
      "flos": 26646076454400.0,
      "grad_norm": 2.714390015116294,
      "language_loss": 0.77092171,
      "learning_rate": 1.9883154917605556e-06,
      "loss": 0.79294342,
      "num_input_tokens_seen": 92774460,
      "step": 4298,
      "time_per_iteration": 2.647944927215576
    },
    {
      "auxiliary_loss_clip": 0.01181542,
      "auxiliary_loss_mlp": 0.01023405,
      "balance_loss_clip": 1.05496311,
      "balance_loss_mlp": 1.01579905,
      "epoch": 0.516924186857452,
      "flos": 19682854542720.0,
      "grad_norm": 2.924979439206456,
      "language_loss": 0.83575439,
      "learning_rate": 1.9875365343132262e-06,
      "loss": 0.85780382,
      "num_input_tokens_seen": 92791580,
      "step": 4299,
      "time_per_iteration": 2.536745071411133
    },
    {
      "auxiliary_loss_clip": 0.01182819,
      "auxiliary_loss_mlp": 0.01208144,
      "balance_loss_clip": 1.05340195,
      "balance_loss_mlp": 1.00059664,
      "epoch": 0.5170444297480912,
      "flos": 15956583586560.0,
      "grad_norm": 2.5740504806748743,
      "language_loss": 0.84852833,
      "learning_rate": 1.9867575787565946e-06,
      "loss": 0.87243795,
      "num_input_tokens_seen": 92806240,
      "step": 4300,
      "time_per_iteration": 2.5494940280914307
    },
    {
      "auxiliary_loss_clip": 0.01186632,
      "auxiliary_loss_mlp": 0.01025859,
      "balance_loss_clip": 1.05413556,
      "balance_loss_mlp": 1.0169301,
      "epoch": 0.5171646726387302,
      "flos": 14174157968640.0,
      "grad_norm": 2.293369152304904,
      "language_loss": 0.85870904,
      "learning_rate": 1.9859786252088275e-06,
      "loss": 0.88083398,
      "num_input_tokens_seen": 92823420,
      "step": 4301,
      "time_per_iteration": 2.5557405948638916
    },
    {
      "auxiliary_loss_clip": 0.01193048,
      "auxiliary_loss_mlp": 0.01030139,
      "balance_loss_clip": 1.04997551,
      "balance_loss_mlp": 1.02116239,
      "epoch": 0.5172849155293693,
      "flos": 23578550974080.0,
      "grad_norm": 8.850874866170964,
      "language_loss": 0.66519034,
      "learning_rate": 1.9851996737880914e-06,
      "loss": 0.68742222,
      "num_input_tokens_seen": 92838605,
      "step": 4302,
      "time_per_iteration": 2.6436381340026855
    },
    {
      "auxiliary_loss_clip": 0.01191142,
      "auxiliary_loss_mlp": 0.01031367,
      "balance_loss_clip": 1.05517054,
      "balance_loss_mlp": 1.02237296,
      "epoch": 0.5174051584200084,
      "flos": 14283541860480.0,
      "grad_norm": 2.48441088676687,
      "language_loss": 0.74593163,
      "learning_rate": 1.9844207246125537e-06,
      "loss": 0.76815677,
      "num_input_tokens_seen": 92855185,
      "step": 4303,
      "time_per_iteration": 2.5420069694519043
    },
    {
      "auxiliary_loss_clip": 0.01182463,
      "auxiliary_loss_mlp": 0.01024485,
      "balance_loss_clip": 1.04937792,
      "balance_loss_mlp": 1.01684332,
      "epoch": 0.5175254013106475,
      "flos": 37889384192640.0,
      "grad_norm": 2.505988595894184,
      "language_loss": 0.68358678,
      "learning_rate": 1.983641777800379e-06,
      "loss": 0.70565629,
      "num_input_tokens_seen": 92877830,
      "step": 4304,
      "time_per_iteration": 2.7857439517974854
    },
    {
      "auxiliary_loss_clip": 0.01092672,
      "auxiliary_loss_mlp": 0.01006029,
      "balance_loss_clip": 1.01708627,
      "balance_loss_mlp": 1.00528991,
      "epoch": 0.5176456442012866,
      "flos": 68549737829760.0,
      "grad_norm": 0.7473846891864199,
      "language_loss": 0.58772886,
      "learning_rate": 1.9828628334697343e-06,
      "loss": 0.60871589,
      "num_input_tokens_seen": 92945040,
      "step": 4305,
      "time_per_iteration": 3.3131473064422607
    },
    {
      "auxiliary_loss_clip": 0.01093343,
      "auxiliary_loss_mlp": 0.01003374,
      "balance_loss_clip": 1.01844776,
      "balance_loss_mlp": 1.00271833,
      "epoch": 0.5177658870919257,
      "flos": 64084137235200.0,
      "grad_norm": 0.7595775994734999,
      "language_loss": 0.54697371,
      "learning_rate": 1.982083891738784e-06,
      "loss": 0.56794095,
      "num_input_tokens_seen": 93005910,
      "step": 4306,
      "time_per_iteration": 3.2208776473999023
    },
    {
      "auxiliary_loss_clip": 0.01178442,
      "auxiliary_loss_mlp": 0.0102654,
      "balance_loss_clip": 1.0512116,
      "balance_loss_mlp": 1.01867235,
      "epoch": 0.5178861299825648,
      "flos": 26651248012800.0,
      "grad_norm": 1.5333843104331573,
      "language_loss": 0.82934701,
      "learning_rate": 1.9813049527256923e-06,
      "loss": 0.8513968,
      "num_input_tokens_seen": 93026305,
      "step": 4307,
      "time_per_iteration": 2.7017228603363037
    },
    {
      "auxiliary_loss_clip": 0.01187429,
      "auxiliary_loss_mlp": 0.01026997,
      "balance_loss_clip": 1.04700756,
      "balance_loss_mlp": 1.01889086,
      "epoch": 0.5180063728732038,
      "flos": 17931886260480.0,
      "grad_norm": 2.5518433162732115,
      "language_loss": 0.82259691,
      "learning_rate": 1.9805260165486252e-06,
      "loss": 0.84474111,
      "num_input_tokens_seen": 93045675,
      "step": 4308,
      "time_per_iteration": 3.554947853088379
    },
    {
      "auxiliary_loss_clip": 0.01183868,
      "auxiliary_loss_mlp": 0.01022869,
      "balance_loss_clip": 1.05436063,
      "balance_loss_mlp": 1.01461411,
      "epoch": 0.518126615763843,
      "flos": 19500895221120.0,
      "grad_norm": 2.36782374305566,
      "language_loss": 0.86349571,
      "learning_rate": 1.9797470833257457e-06,
      "loss": 0.88556302,
      "num_input_tokens_seen": 93065375,
      "step": 4309,
      "time_per_iteration": 2.6238865852355957
    },
    {
      "auxiliary_loss_clip": 0.01186099,
      "auxiliary_loss_mlp": 0.01030874,
      "balance_loss_clip": 1.05478847,
      "balance_loss_mlp": 1.02228487,
      "epoch": 0.5182468586544821,
      "flos": 20704082117760.0,
      "grad_norm": 1.9532342768196551,
      "language_loss": 0.77717388,
      "learning_rate": 1.9789681531752177e-06,
      "loss": 0.79934359,
      "num_input_tokens_seen": 93085595,
      "step": 4310,
      "time_per_iteration": 3.523582935333252
    },
    {
      "auxiliary_loss_clip": 0.01187779,
      "auxiliary_loss_mlp": 0.01024788,
      "balance_loss_clip": 1.04843819,
      "balance_loss_mlp": 1.01731968,
      "epoch": 0.5183671015451211,
      "flos": 23112107936640.0,
      "grad_norm": 1.591989472651271,
      "language_loss": 0.7258482,
      "learning_rate": 1.978189226215204e-06,
      "loss": 0.74797386,
      "num_input_tokens_seen": 93106140,
      "step": 4311,
      "time_per_iteration": 2.655491828918457
    },
    {
      "auxiliary_loss_clip": 0.01182703,
      "auxiliary_loss_mlp": 0.0102666,
      "balance_loss_clip": 1.05504525,
      "balance_loss_mlp": 1.01802301,
      "epoch": 0.5184873444357603,
      "flos": 17597090568960.0,
      "grad_norm": 3.100249547164379,
      "language_loss": 0.76733387,
      "learning_rate": 1.9774103025638675e-06,
      "loss": 0.7894274,
      "num_input_tokens_seen": 93124265,
      "step": 4312,
      "time_per_iteration": 2.564422369003296
    },
    {
      "auxiliary_loss_clip": 0.011953,
      "auxiliary_loss_mlp": 0.0102412,
      "balance_loss_clip": 1.05400765,
      "balance_loss_mlp": 1.01647294,
      "epoch": 0.5186075873263993,
      "flos": 24936800883840.0,
      "grad_norm": 1.5242274254838686,
      "language_loss": 0.76161611,
      "learning_rate": 1.9766313823393696e-06,
      "loss": 0.78381032,
      "num_input_tokens_seen": 93145130,
      "step": 4313,
      "time_per_iteration": 3.5893678665161133
    },
    {
      "auxiliary_loss_clip": 0.01184317,
      "auxiliary_loss_mlp": 0.01027089,
      "balance_loss_clip": 1.04302084,
      "balance_loss_mlp": 1.01842284,
      "epoch": 0.5187278302170384,
      "flos": 15190106244480.0,
      "grad_norm": 2.6294230326868915,
      "language_loss": 0.69410527,
      "learning_rate": 1.975852465659873e-06,
      "loss": 0.71621931,
      "num_input_tokens_seen": 93161110,
      "step": 4314,
      "time_per_iteration": 2.7031333446502686
    },
    {
      "auxiliary_loss_clip": 0.01184946,
      "auxiliary_loss_mlp": 0.010312,
      "balance_loss_clip": 1.05312753,
      "balance_loss_mlp": 1.02322495,
      "epoch": 0.5188480731076776,
      "flos": 25009412227200.0,
      "grad_norm": 2.1108801349343027,
      "language_loss": 0.7008056,
      "learning_rate": 1.9750735526435377e-06,
      "loss": 0.72296715,
      "num_input_tokens_seen": 93178055,
      "step": 4315,
      "time_per_iteration": 2.596543550491333
    },
    {
      "auxiliary_loss_clip": 0.01183474,
      "auxiliary_loss_mlp": 0.01023981,
      "balance_loss_clip": 1.05143023,
      "balance_loss_mlp": 1.01577628,
      "epoch": 0.5189683159983166,
      "flos": 24790141653120.0,
      "grad_norm": 4.327044422952634,
      "language_loss": 0.79363579,
      "learning_rate": 1.974294643408525e-06,
      "loss": 0.81571031,
      "num_input_tokens_seen": 93195850,
      "step": 4316,
      "time_per_iteration": 2.6560637950897217
    },
    {
      "auxiliary_loss_clip": 0.01186878,
      "auxiliary_loss_mlp": 0.01027272,
      "balance_loss_clip": 1.05133069,
      "balance_loss_mlp": 1.01937449,
      "epoch": 0.5190885588889557,
      "flos": 24754266944640.0,
      "grad_norm": 2.0019171988508564,
      "language_loss": 0.67281348,
      "learning_rate": 1.9735157380729947e-06,
      "loss": 0.69495499,
      "num_input_tokens_seen": 93216260,
      "step": 4317,
      "time_per_iteration": 2.704587459564209
    },
    {
      "auxiliary_loss_clip": 0.01187115,
      "auxiliary_loss_mlp": 0.01024974,
      "balance_loss_clip": 1.05048501,
      "balance_loss_mlp": 1.0174042,
      "epoch": 0.5192088017795948,
      "flos": 24712646060160.0,
      "grad_norm": 1.7526265813807933,
      "language_loss": 0.84212708,
      "learning_rate": 1.9727368367551053e-06,
      "loss": 0.86424804,
      "num_input_tokens_seen": 93234810,
      "step": 4318,
      "time_per_iteration": 3.660064697265625
    },
    {
      "auxiliary_loss_clip": 0.01172852,
      "auxiliary_loss_mlp": 0.0102833,
      "balance_loss_clip": 1.04813743,
      "balance_loss_mlp": 1.02018213,
      "epoch": 0.5193290446702339,
      "flos": 27229588894080.0,
      "grad_norm": 2.024161942781723,
      "language_loss": 0.68510938,
      "learning_rate": 1.9719579395730164e-06,
      "loss": 0.70712119,
      "num_input_tokens_seen": 93254185,
      "step": 4319,
      "time_per_iteration": 2.7373006343841553
    },
    {
      "auxiliary_loss_clip": 0.01184517,
      "auxiliary_loss_mlp": 0.01024074,
      "balance_loss_clip": 1.05741692,
      "balance_loss_mlp": 1.01617622,
      "epoch": 0.5194492875608729,
      "flos": 11473352392320.0,
      "grad_norm": 2.919588550589444,
      "language_loss": 0.93865073,
      "learning_rate": 1.9711790466448854e-06,
      "loss": 0.96073657,
      "num_input_tokens_seen": 93268205,
      "step": 4320,
      "time_per_iteration": 2.600757598876953
    },
    {
      "auxiliary_loss_clip": 0.01197895,
      "auxiliary_loss_mlp": 0.01034204,
      "balance_loss_clip": 1.05006886,
      "balance_loss_mlp": 1.02550185,
      "epoch": 0.5195695304515121,
      "flos": 20338906498560.0,
      "grad_norm": 2.2191738513765658,
      "language_loss": 0.71101534,
      "learning_rate": 1.9704001580888704e-06,
      "loss": 0.73333633,
      "num_input_tokens_seen": 93286945,
      "step": 4321,
      "time_per_iteration": 2.7109947204589844
    },
    {
      "auxiliary_loss_clip": 0.01177754,
      "auxiliary_loss_mlp": 0.01208241,
      "balance_loss_clip": 1.04574025,
      "balance_loss_mlp": 1.00068021,
      "epoch": 0.5196897733421512,
      "flos": 20048317470720.0,
      "grad_norm": 1.8382715062615094,
      "language_loss": 0.86840469,
      "learning_rate": 1.9696212740231283e-06,
      "loss": 0.89226472,
      "num_input_tokens_seen": 93305595,
      "step": 4322,
      "time_per_iteration": 2.5739996433258057
    },
    {
      "auxiliary_loss_clip": 0.01190352,
      "auxiliary_loss_mlp": 0.01026211,
      "balance_loss_clip": 1.05174923,
      "balance_loss_mlp": 1.01759243,
      "epoch": 0.5198100162327902,
      "flos": 23805507058560.0,
      "grad_norm": 2.3501980244189853,
      "language_loss": 0.82430559,
      "learning_rate": 1.9688423945658146e-06,
      "loss": 0.84647119,
      "num_input_tokens_seen": 93326460,
      "step": 4323,
      "time_per_iteration": 2.5883994102478027
    },
    {
      "auxiliary_loss_clip": 0.01180995,
      "auxiliary_loss_mlp": 0.01026604,
      "balance_loss_clip": 1.041605,
      "balance_loss_mlp": 1.01780009,
      "epoch": 0.5199302591234293,
      "flos": 24023951619840.0,
      "grad_norm": 2.454444977812376,
      "language_loss": 0.72036707,
      "learning_rate": 1.9680635198350845e-06,
      "loss": 0.74244308,
      "num_input_tokens_seen": 93346170,
      "step": 4324,
      "time_per_iteration": 2.7186105251312256
    },
    {
      "auxiliary_loss_clip": 0.01183423,
      "auxiliary_loss_mlp": 0.01036274,
      "balance_loss_clip": 1.0502286,
      "balance_loss_mlp": 1.02741718,
      "epoch": 0.5200505020140684,
      "flos": 26359366095360.0,
      "grad_norm": 3.5159999071546713,
      "language_loss": 0.73000139,
      "learning_rate": 1.967284649949093e-06,
      "loss": 0.7521984,
      "num_input_tokens_seen": 93365380,
      "step": 4325,
      "time_per_iteration": 2.6670711040496826
    },
    {
      "auxiliary_loss_clip": 0.01181967,
      "auxiliary_loss_mlp": 0.01030196,
      "balance_loss_clip": 1.04469824,
      "balance_loss_mlp": 1.02174437,
      "epoch": 0.5201707449047075,
      "flos": 39604262284800.0,
      "grad_norm": 2.0430788437720673,
      "language_loss": 0.7242164,
      "learning_rate": 1.966505785025994e-06,
      "loss": 0.74633801,
      "num_input_tokens_seen": 93387285,
      "step": 4326,
      "time_per_iteration": 2.8520333766937256
    },
    {
      "auxiliary_loss_clip": 0.01185979,
      "auxiliary_loss_mlp": 0.01025743,
      "balance_loss_clip": 1.04898131,
      "balance_loss_mlp": 1.01729155,
      "epoch": 0.5202909877953465,
      "flos": 53682788292480.0,
      "grad_norm": 1.8135197062765565,
      "language_loss": 0.7629621,
      "learning_rate": 1.965726925183941e-06,
      "loss": 0.78507936,
      "num_input_tokens_seen": 93410390,
      "step": 4327,
      "time_per_iteration": 3.0005943775177
    },
    {
      "auxiliary_loss_clip": 0.01180756,
      "auxiliary_loss_mlp": 0.01023901,
      "balance_loss_clip": 1.05462337,
      "balance_loss_mlp": 1.01617622,
      "epoch": 0.5204112306859857,
      "flos": 19537021324800.0,
      "grad_norm": 1.7975599042306365,
      "language_loss": 0.85228181,
      "learning_rate": 1.964948070541087e-06,
      "loss": 0.87432843,
      "num_input_tokens_seen": 93429050,
      "step": 4328,
      "time_per_iteration": 2.619485378265381
    },
    {
      "auxiliary_loss_clip": 0.01172261,
      "auxiliary_loss_mlp": 0.01032517,
      "balance_loss_clip": 1.04826498,
      "balance_loss_mlp": 1.02429736,
      "epoch": 0.5205314735766248,
      "flos": 15304697608320.0,
      "grad_norm": 10.142746099164304,
      "language_loss": 0.70115554,
      "learning_rate": 1.9641692212155816e-06,
      "loss": 0.7232033,
      "num_input_tokens_seen": 93446815,
      "step": 4329,
      "time_per_iteration": 2.603271961212158
    },
    {
      "auxiliary_loss_clip": 0.01189608,
      "auxiliary_loss_mlp": 0.01031646,
      "balance_loss_clip": 1.05075622,
      "balance_loss_mlp": 1.02302754,
      "epoch": 0.5206517164672638,
      "flos": 59263701160320.0,
      "grad_norm": 2.0544419917677836,
      "language_loss": 0.72892433,
      "learning_rate": 1.9633903773255777e-06,
      "loss": 0.75113684,
      "num_input_tokens_seen": 93469130,
      "step": 4330,
      "time_per_iteration": 2.985344886779785
    },
    {
      "auxiliary_loss_clip": 0.01178351,
      "auxiliary_loss_mlp": 0.01024668,
      "balance_loss_clip": 1.05140603,
      "balance_loss_mlp": 1.01657343,
      "epoch": 0.520771959357903,
      "flos": 26871129118080.0,
      "grad_norm": 1.7505268431476446,
      "language_loss": 0.75149,
      "learning_rate": 1.9626115389892237e-06,
      "loss": 0.77352023,
      "num_input_tokens_seen": 93489920,
      "step": 4331,
      "time_per_iteration": 2.6564698219299316
    },
    {
      "auxiliary_loss_clip": 0.01193006,
      "auxiliary_loss_mlp": 0.010256,
      "balance_loss_clip": 1.04917002,
      "balance_loss_mlp": 1.01746416,
      "epoch": 0.520892202248542,
      "flos": 26907075653760.0,
      "grad_norm": 2.1912011457369163,
      "language_loss": 0.8510921,
      "learning_rate": 1.96183270632467e-06,
      "loss": 0.87327826,
      "num_input_tokens_seen": 93509770,
      "step": 4332,
      "time_per_iteration": 2.679248809814453
    },
    {
      "auxiliary_loss_clip": 0.01181013,
      "auxiliary_loss_mlp": 0.01208781,
      "balance_loss_clip": 1.04654098,
      "balance_loss_mlp": 1.00077295,
      "epoch": 0.5210124451391811,
      "flos": 25849434666240.0,
      "grad_norm": 2.4126222616410065,
      "language_loss": 0.79414535,
      "learning_rate": 1.9610538794500644e-06,
      "loss": 0.81804329,
      "num_input_tokens_seen": 93529320,
      "step": 4333,
      "time_per_iteration": 2.6365203857421875
    },
    {
      "auxiliary_loss_clip": 0.01099275,
      "auxiliary_loss_mlp": 0.01005181,
      "balance_loss_clip": 1.01927781,
      "balance_loss_mlp": 1.00440645,
      "epoch": 0.5211326880298203,
      "flos": 70553804319360.0,
      "grad_norm": 0.7767430357883297,
      "language_loss": 0.59388345,
      "learning_rate": 1.9602750584835542e-06,
      "loss": 0.61492795,
      "num_input_tokens_seen": 93595255,
      "step": 4334,
      "time_per_iteration": 3.2998244762420654
    },
    {
      "auxiliary_loss_clip": 0.01182185,
      "auxiliary_loss_mlp": 0.01023344,
      "balance_loss_clip": 1.04918003,
      "balance_loss_mlp": 1.01537454,
      "epoch": 0.5212529309204593,
      "flos": 15628898787840.0,
      "grad_norm": 2.289591669046954,
      "language_loss": 0.82403815,
      "learning_rate": 1.959496243543286e-06,
      "loss": 0.84609342,
      "num_input_tokens_seen": 93613135,
      "step": 4335,
      "time_per_iteration": 3.539951801300049
    },
    {
      "auxiliary_loss_clip": 0.01188102,
      "auxiliary_loss_mlp": 0.01038952,
      "balance_loss_clip": 1.05710626,
      "balance_loss_mlp": 1.03046465,
      "epoch": 0.5213731738110984,
      "flos": 26242655829120.0,
      "grad_norm": 2.2533544335606477,
      "language_loss": 0.79523218,
      "learning_rate": 1.9587174347474057e-06,
      "loss": 0.81750268,
      "num_input_tokens_seen": 93629645,
      "step": 4336,
      "time_per_iteration": 2.6176693439483643
    },
    {
      "auxiliary_loss_clip": 0.0117802,
      "auxiliary_loss_mlp": 0.01029481,
      "balance_loss_clip": 1.04536748,
      "balance_loss_mlp": 1.02119613,
      "epoch": 0.5214934167017375,
      "flos": 19418407637760.0,
      "grad_norm": 2.5247409783308696,
      "language_loss": 0.81758451,
      "learning_rate": 1.9579386322140574e-06,
      "loss": 0.83965957,
      "num_input_tokens_seen": 93645325,
      "step": 4337,
      "time_per_iteration": 3.582711935043335
    },
    {
      "auxiliary_loss_clip": 0.01183991,
      "auxiliary_loss_mlp": 0.01208574,
      "balance_loss_clip": 1.05484378,
      "balance_loss_mlp": 1.00086331,
      "epoch": 0.5216136595923766,
      "flos": 30955788023040.0,
      "grad_norm": 4.207448335352718,
      "language_loss": 0.80876064,
      "learning_rate": 1.9571598360613854e-06,
      "loss": 0.83268631,
      "num_input_tokens_seen": 93668200,
      "step": 4338,
      "time_per_iteration": 2.7278904914855957
    },
    {
      "auxiliary_loss_clip": 0.01170709,
      "auxiliary_loss_mlp": 0.01023967,
      "balance_loss_clip": 1.04601693,
      "balance_loss_mlp": 1.01583087,
      "epoch": 0.5217339024830157,
      "flos": 21945047143680.0,
      "grad_norm": 2.146156223354211,
      "language_loss": 0.69459826,
      "learning_rate": 1.956381046407532e-06,
      "loss": 0.71654499,
      "num_input_tokens_seen": 93688495,
      "step": 4339,
      "time_per_iteration": 3.551236867904663
    },
    {
      "auxiliary_loss_clip": 0.01184038,
      "auxiliary_loss_mlp": 0.0103157,
      "balance_loss_clip": 1.04827893,
      "balance_loss_mlp": 1.0231595,
      "epoch": 0.5218541453736548,
      "flos": 20923209037440.0,
      "grad_norm": 2.3471473503315994,
      "language_loss": 0.86180902,
      "learning_rate": 1.9556022633706394e-06,
      "loss": 0.88396502,
      "num_input_tokens_seen": 93707285,
      "step": 4340,
      "time_per_iteration": 2.6670022010803223
    },
    {
      "auxiliary_loss_clip": 0.01181395,
      "auxiliary_loss_mlp": 0.01028813,
      "balance_loss_clip": 1.05185544,
      "balance_loss_mlp": 1.02058172,
      "epoch": 0.5219743882642939,
      "flos": 23951663498880.0,
      "grad_norm": 6.297774271821073,
      "language_loss": 0.80180699,
      "learning_rate": 1.954823487068848e-06,
      "loss": 0.82390904,
      "num_input_tokens_seen": 93727495,
      "step": 4341,
      "time_per_iteration": 2.642583131790161
    },
    {
      "auxiliary_loss_clip": 0.01184584,
      "auxiliary_loss_mlp": 0.01030655,
      "balance_loss_clip": 1.05525053,
      "balance_loss_mlp": 1.02259016,
      "epoch": 0.5220946311549329,
      "flos": 28799280213120.0,
      "grad_norm": 1.661171087013023,
      "language_loss": 0.80959308,
      "learning_rate": 1.9540447176202976e-06,
      "loss": 0.83174545,
      "num_input_tokens_seen": 93748740,
      "step": 4342,
      "time_per_iteration": 2.6832480430603027
    },
    {
      "auxiliary_loss_clip": 0.01085283,
      "auxiliary_loss_mlp": 0.01001086,
      "balance_loss_clip": 1.02075171,
      "balance_loss_mlp": 1.00037694,
      "epoch": 0.5222148740455721,
      "flos": 67189369017600.0,
      "grad_norm": 0.8680924568598332,
      "language_loss": 0.60679287,
      "learning_rate": 1.9532659551431272e-06,
      "loss": 0.62765652,
      "num_input_tokens_seen": 93815770,
      "step": 4343,
      "time_per_iteration": 3.319855213165283
    },
    {
      "auxiliary_loss_clip": 0.01183899,
      "auxiliary_loss_mlp": 0.01023751,
      "balance_loss_clip": 1.05226576,
      "balance_loss_mlp": 1.01603174,
      "epoch": 0.5223351169362112,
      "flos": 61856164339200.0,
      "grad_norm": 1.6568331724754053,
      "language_loss": 0.67716932,
      "learning_rate": 1.9524871997554744e-06,
      "loss": 0.69924581,
      "num_input_tokens_seen": 93843530,
      "step": 4344,
      "time_per_iteration": 3.0176525115966797
    },
    {
      "auxiliary_loss_clip": 0.01186563,
      "auxiliary_loss_mlp": 0.01027506,
      "balance_loss_clip": 1.05515409,
      "balance_loss_mlp": 1.01929235,
      "epoch": 0.5224553598268502,
      "flos": 14647388676480.0,
      "grad_norm": 2.118290038017476,
      "language_loss": 0.81075644,
      "learning_rate": 1.951708451575475e-06,
      "loss": 0.83289707,
      "num_input_tokens_seen": 93860595,
      "step": 4345,
      "time_per_iteration": 3.42755126953125
    },
    {
      "auxiliary_loss_clip": 0.01197092,
      "auxiliary_loss_mlp": 0.01028519,
      "balance_loss_clip": 1.04889035,
      "balance_loss_mlp": 1.02059817,
      "epoch": 0.5225756027174894,
      "flos": 14826043946880.0,
      "grad_norm": 4.233817233517706,
      "language_loss": 0.82586068,
      "learning_rate": 1.9509297107212657e-06,
      "loss": 0.84811676,
      "num_input_tokens_seen": 93877365,
      "step": 4346,
      "time_per_iteration": 2.6751983165740967
    },
    {
      "auxiliary_loss_clip": 0.0117936,
      "auxiliary_loss_mlp": 0.01028231,
      "balance_loss_clip": 1.05389166,
      "balance_loss_mlp": 1.02038062,
      "epoch": 0.5226958456081284,
      "flos": 23512009029120.0,
      "grad_norm": 1.6012825069901206,
      "language_loss": 0.78946137,
      "learning_rate": 1.95015097731098e-06,
      "loss": 0.81153727,
      "num_input_tokens_seen": 93896855,
      "step": 4347,
      "time_per_iteration": 2.6172995567321777
    },
    {
      "auxiliary_loss_clip": 0.01179766,
      "auxiliary_loss_mlp": 0.01023351,
      "balance_loss_clip": 1.05345917,
      "balance_loss_mlp": 1.01558709,
      "epoch": 0.5228160884987675,
      "flos": 19062928690560.0,
      "grad_norm": 2.2318921868961854,
      "language_loss": 0.81691349,
      "learning_rate": 1.949372251462751e-06,
      "loss": 0.83894473,
      "num_input_tokens_seen": 93914270,
      "step": 4348,
      "time_per_iteration": 2.586688995361328
    },
    {
      "auxiliary_loss_clip": 0.01189546,
      "auxiliary_loss_mlp": 0.0120798,
      "balance_loss_clip": 1.05165792,
      "balance_loss_mlp": 1.00076795,
      "epoch": 0.5229363313894067,
      "flos": 21063224252160.0,
      "grad_norm": 1.9749617961150254,
      "language_loss": 0.83027732,
      "learning_rate": 1.9485935332947124e-06,
      "loss": 0.85425258,
      "num_input_tokens_seen": 93932180,
      "step": 4349,
      "time_per_iteration": 2.621142625808716
    },
    {
      "auxiliary_loss_clip": 0.01180342,
      "auxiliary_loss_mlp": 0.01021981,
      "balance_loss_clip": 1.0521071,
      "balance_loss_mlp": 1.01418447,
      "epoch": 0.5230565742800457,
      "flos": 14830389492480.0,
      "grad_norm": 2.3471920946894866,
      "language_loss": 0.83352172,
      "learning_rate": 1.947814822924993e-06,
      "loss": 0.85554492,
      "num_input_tokens_seen": 93949690,
      "step": 4350,
      "time_per_iteration": 2.691922903060913
    },
    {
      "auxiliary_loss_clip": 0.01180057,
      "auxiliary_loss_mlp": 0.01029716,
      "balance_loss_clip": 1.055758,
      "balance_loss_mlp": 1.02223539,
      "epoch": 0.5231768171706848,
      "flos": 25813021253760.0,
      "grad_norm": 2.110500783895415,
      "language_loss": 0.82698125,
      "learning_rate": 1.9470361204717236e-06,
      "loss": 0.84907901,
      "num_input_tokens_seen": 93968830,
      "step": 4351,
      "time_per_iteration": 2.6927545070648193
    },
    {
      "auxiliary_loss_clip": 0.01193277,
      "auxiliary_loss_mlp": 0.01208428,
      "balance_loss_clip": 1.05232859,
      "balance_loss_mlp": 1.00089288,
      "epoch": 0.5232970600613239,
      "flos": 22743807834240.0,
      "grad_norm": 1.941107113352125,
      "language_loss": 0.8057934,
      "learning_rate": 1.9462574260530326e-06,
      "loss": 0.8298105,
      "num_input_tokens_seen": 93989110,
      "step": 4352,
      "time_per_iteration": 2.6596131324768066
    },
    {
      "auxiliary_loss_clip": 0.01173505,
      "auxiliary_loss_mlp": 0.01025839,
      "balance_loss_clip": 1.05036485,
      "balance_loss_mlp": 1.01744676,
      "epoch": 0.523417302951963,
      "flos": 17310703432320.0,
      "grad_norm": 1.9055105912765313,
      "language_loss": 0.81046748,
      "learning_rate": 1.9454787397870472e-06,
      "loss": 0.832461,
      "num_input_tokens_seen": 94006430,
      "step": 4353,
      "time_per_iteration": 2.6648337841033936
    },
    {
      "auxiliary_loss_clip": 0.01184948,
      "auxiliary_loss_mlp": 0.01027403,
      "balance_loss_clip": 1.04667735,
      "balance_loss_mlp": 1.01934409,
      "epoch": 0.523537545842602,
      "flos": 18551740285440.0,
      "grad_norm": 2.1413340795378604,
      "language_loss": 0.71791828,
      "learning_rate": 1.944700061791894e-06,
      "loss": 0.74004173,
      "num_input_tokens_seen": 94024825,
      "step": 4354,
      "time_per_iteration": 2.6700854301452637
    },
    {
      "auxiliary_loss_clip": 0.01181441,
      "auxiliary_loss_mlp": 0.01029618,
      "balance_loss_clip": 1.05365276,
      "balance_loss_mlp": 1.02151227,
      "epoch": 0.5236577887332411,
      "flos": 19719267955200.0,
      "grad_norm": 4.2378161403420425,
      "language_loss": 0.65665048,
      "learning_rate": 1.943921392185698e-06,
      "loss": 0.67876107,
      "num_input_tokens_seen": 94043450,
      "step": 4355,
      "time_per_iteration": 2.532393217086792
    },
    {
      "auxiliary_loss_clip": 0.01189276,
      "auxiliary_loss_mlp": 0.01026914,
      "balance_loss_clip": 1.0513941,
      "balance_loss_mlp": 1.01859915,
      "epoch": 0.5237780316238803,
      "flos": 23550218121600.0,
      "grad_norm": 2.8428630514352284,
      "language_loss": 0.76962823,
      "learning_rate": 1.9431427310865814e-06,
      "loss": 0.79179013,
      "num_input_tokens_seen": 94063055,
      "step": 4356,
      "time_per_iteration": 2.6593873500823975
    },
    {
      "auxiliary_loss_clip": 0.01172052,
      "auxiliary_loss_mlp": 0.01029841,
      "balance_loss_clip": 1.0481925,
      "balance_loss_mlp": 1.02162194,
      "epoch": 0.5238982745145193,
      "flos": 22491894775680.0,
      "grad_norm": 1.699702787257498,
      "language_loss": 0.78619206,
      "learning_rate": 1.942364078612667e-06,
      "loss": 0.80821097,
      "num_input_tokens_seen": 94081785,
      "step": 4357,
      "time_per_iteration": 2.6396396160125732
    },
    {
      "auxiliary_loss_clip": 0.01194962,
      "auxiliary_loss_mlp": 0.01025793,
      "balance_loss_clip": 1.04930604,
      "balance_loss_mlp": 1.0179013,
      "epoch": 0.5240185174051584,
      "flos": 27088927234560.0,
      "grad_norm": 1.8753587098069309,
      "language_loss": 0.75520331,
      "learning_rate": 1.9415854348820765e-06,
      "loss": 0.77741086,
      "num_input_tokens_seen": 94101635,
      "step": 4358,
      "time_per_iteration": 2.726637840270996
    },
    {
      "auxiliary_loss_clip": 0.01189225,
      "auxiliary_loss_mlp": 0.01028959,
      "balance_loss_clip": 1.0530777,
      "balance_loss_mlp": 1.01991129,
      "epoch": 0.5241387602957975,
      "flos": 22674680110080.0,
      "grad_norm": 3.249349369664098,
      "language_loss": 0.68683636,
      "learning_rate": 1.940806800012929e-06,
      "loss": 0.70901817,
      "num_input_tokens_seen": 94121705,
      "step": 4359,
      "time_per_iteration": 2.641714572906494
    },
    {
      "auxiliary_loss_clip": 0.01185462,
      "auxiliary_loss_mlp": 0.01208529,
      "balance_loss_clip": 1.04690611,
      "balance_loss_mlp": 1.00088787,
      "epoch": 0.5242590031864366,
      "flos": 40553453134080.0,
      "grad_norm": 1.4540055771052651,
      "language_loss": 0.63907284,
      "learning_rate": 1.9400281741233432e-06,
      "loss": 0.66301274,
      "num_input_tokens_seen": 94146595,
      "step": 4360,
      "time_per_iteration": 2.8471052646636963
    },
    {
      "auxiliary_loss_clip": 0.01091529,
      "auxiliary_loss_mlp": 0.01004559,
      "balance_loss_clip": 1.01995564,
      "balance_loss_mlp": 1.00365341,
      "epoch": 0.5243792460770756,
      "flos": 66676313105280.0,
      "grad_norm": 0.660943439170088,
      "language_loss": 0.52559727,
      "learning_rate": 1.939249557331435e-06,
      "loss": 0.54655814,
      "num_input_tokens_seen": 94212410,
      "step": 4361,
      "time_per_iteration": 4.134632110595703
    },
    {
      "auxiliary_loss_clip": 0.01196307,
      "auxiliary_loss_mlp": 0.01027552,
      "balance_loss_clip": 1.05055785,
      "balance_loss_mlp": 1.0201757,
      "epoch": 0.5244994889677148,
      "flos": 28183663992960.0,
      "grad_norm": 2.8100389724603807,
      "language_loss": 0.72855258,
      "learning_rate": 1.938470949755321e-06,
      "loss": 0.75079107,
      "num_input_tokens_seen": 94232290,
      "step": 4362,
      "time_per_iteration": 2.659452199935913
    },
    {
      "auxiliary_loss_clip": 0.01096846,
      "auxiliary_loss_mlp": 0.01007161,
      "balance_loss_clip": 1.01843286,
      "balance_loss_mlp": 1.00643396,
      "epoch": 0.5246197318583539,
      "flos": 65950379239680.0,
      "grad_norm": 0.8090642566917867,
      "language_loss": 0.55677724,
      "learning_rate": 1.937692351513115e-06,
      "loss": 0.57781732,
      "num_input_tokens_seen": 94291285,
      "step": 4363,
      "time_per_iteration": 3.1479604244232178
    },
    {
      "auxiliary_loss_clip": 0.01188138,
      "auxiliary_loss_mlp": 0.01024224,
      "balance_loss_clip": 1.05281329,
      "balance_loss_mlp": 1.01646018,
      "epoch": 0.5247399747489929,
      "flos": 21033490769280.0,
      "grad_norm": 1.7786970809262312,
      "language_loss": 0.81102514,
      "learning_rate": 1.9369137627229297e-06,
      "loss": 0.83314884,
      "num_input_tokens_seen": 94309685,
      "step": 4364,
      "time_per_iteration": 2.6364588737487793
    },
    {
      "auxiliary_loss_clip": 0.01182886,
      "auxiliary_loss_mlp": 0.01030052,
      "balance_loss_clip": 1.053267,
      "balance_loss_mlp": 1.02186179,
      "epoch": 0.5248602176396321,
      "flos": 19025940660480.0,
      "grad_norm": 1.9349219999909286,
      "language_loss": 0.88456601,
      "learning_rate": 1.936135183502877e-06,
      "loss": 0.90669537,
      "num_input_tokens_seen": 94326985,
      "step": 4365,
      "time_per_iteration": 3.5244898796081543
    },
    {
      "auxiliary_loss_clip": 0.01192307,
      "auxiliary_loss_mlp": 0.01024198,
      "balance_loss_clip": 1.05000067,
      "balance_loss_mlp": 1.01609147,
      "epoch": 0.5249804605302711,
      "flos": 22200084685440.0,
      "grad_norm": 2.2039425603309817,
      "language_loss": 0.8034271,
      "learning_rate": 1.935356613971066e-06,
      "loss": 0.82559216,
      "num_input_tokens_seen": 94347645,
      "step": 4366,
      "time_per_iteration": 3.6386330127716064
    },
    {
      "auxiliary_loss_clip": 0.01183894,
      "auxiliary_loss_mlp": 0.01208511,
      "balance_loss_clip": 1.05136395,
      "balance_loss_mlp": 1.00088692,
      "epoch": 0.5251007034209102,
      "flos": 23805686626560.0,
      "grad_norm": 1.8418767535670952,
      "language_loss": 0.76821899,
      "learning_rate": 1.9345780542456047e-06,
      "loss": 0.79214299,
      "num_input_tokens_seen": 94367020,
      "step": 4367,
      "time_per_iteration": 2.6372487545013428
    },
    {
      "auxiliary_loss_clip": 0.01172661,
      "auxiliary_loss_mlp": 0.01025079,
      "balance_loss_clip": 1.05080414,
      "balance_loss_mlp": 1.01705635,
      "epoch": 0.5252209463115494,
      "flos": 23294605962240.0,
      "grad_norm": 2.4398872822867026,
      "language_loss": 0.71686071,
      "learning_rate": 1.9337995044446007e-06,
      "loss": 0.73883808,
      "num_input_tokens_seen": 94385860,
      "step": 4368,
      "time_per_iteration": 2.650952100753784
    },
    {
      "auxiliary_loss_clip": 0.01187297,
      "auxiliary_loss_mlp": 0.01025556,
      "balance_loss_clip": 1.05281782,
      "balance_loss_mlp": 1.01728332,
      "epoch": 0.5253411892021884,
      "flos": 19828687760640.0,
      "grad_norm": 2.101605406849696,
      "language_loss": 0.79975343,
      "learning_rate": 1.9330209646861596e-06,
      "loss": 0.82188201,
      "num_input_tokens_seen": 94405010,
      "step": 4369,
      "time_per_iteration": 2.6198904514312744
    },
    {
      "auxiliary_loss_clip": 0.01181789,
      "auxiliary_loss_mlp": 0.01028969,
      "balance_loss_clip": 1.05071926,
      "balance_loss_mlp": 1.0213697,
      "epoch": 0.5254614320928275,
      "flos": 24133730561280.0,
      "grad_norm": 3.068889580368602,
      "language_loss": 0.77638918,
      "learning_rate": 1.9322424350883843e-06,
      "loss": 0.79849678,
      "num_input_tokens_seen": 94426845,
      "step": 4370,
      "time_per_iteration": 2.6793997287750244
    },
    {
      "auxiliary_loss_clip": 0.01185321,
      "auxiliary_loss_mlp": 0.01027346,
      "balance_loss_clip": 1.05011487,
      "balance_loss_mlp": 1.01937139,
      "epoch": 0.5255816749834666,
      "flos": 24644954880000.0,
      "grad_norm": 1.802412266270945,
      "language_loss": 0.78789663,
      "learning_rate": 1.931463915769379e-06,
      "loss": 0.81002331,
      "num_input_tokens_seen": 94446960,
      "step": 4371,
      "time_per_iteration": 2.668612241744995
    },
    {
      "auxiliary_loss_clip": 0.01186881,
      "auxiliary_loss_mlp": 0.01027439,
      "balance_loss_clip": 1.0465349,
      "balance_loss_mlp": 1.01926112,
      "epoch": 0.5257019178741057,
      "flos": 14136595320960.0,
      "grad_norm": 2.7408516776729868,
      "language_loss": 0.73727274,
      "learning_rate": 1.930685406847242e-06,
      "loss": 0.75941598,
      "num_input_tokens_seen": 94461535,
      "step": 4372,
      "time_per_iteration": 3.648305654525757
    },
    {
      "auxiliary_loss_clip": 0.01179326,
      "auxiliary_loss_mlp": 0.01025679,
      "balance_loss_clip": 1.04822326,
      "balance_loss_mlp": 1.01800811,
      "epoch": 0.5258221607647448,
      "flos": 23548961145600.0,
      "grad_norm": 1.4462643737858902,
      "language_loss": 0.8198846,
      "learning_rate": 1.9299069084400734e-06,
      "loss": 0.84193468,
      "num_input_tokens_seen": 94482395,
      "step": 4373,
      "time_per_iteration": 2.6677215099334717
    },
    {
      "auxiliary_loss_clip": 0.01185966,
      "auxiliary_loss_mlp": 0.0102771,
      "balance_loss_clip": 1.05090678,
      "balance_loss_mlp": 1.01915061,
      "epoch": 0.5259424036553839,
      "flos": 24966103403520.0,
      "grad_norm": 2.2491738858981694,
      "language_loss": 0.69605374,
      "learning_rate": 1.9291284206659717e-06,
      "loss": 0.71819049,
      "num_input_tokens_seen": 94500580,
      "step": 4374,
      "time_per_iteration": 2.6578853130340576
    },
    {
      "auxiliary_loss_clip": 0.01183478,
      "auxiliary_loss_mlp": 0.01024272,
      "balance_loss_clip": 1.05595374,
      "balance_loss_mlp": 1.01577234,
      "epoch": 0.526062646546023,
      "flos": 28763908295040.0,
      "grad_norm": 1.8319319932018456,
      "language_loss": 0.71779811,
      "learning_rate": 1.928349943643032e-06,
      "loss": 0.73987567,
      "num_input_tokens_seen": 94519680,
      "step": 4375,
      "time_per_iteration": 2.7099287509918213
    },
    {
      "auxiliary_loss_clip": 0.01177869,
      "auxiliary_loss_mlp": 0.0102713,
      "balance_loss_clip": 1.05257487,
      "balance_loss_mlp": 1.01897597,
      "epoch": 0.526182889436662,
      "flos": 22821375254400.0,
      "grad_norm": 2.5950448571267053,
      "language_loss": 0.81972098,
      "learning_rate": 1.9275714774893493e-06,
      "loss": 0.84177101,
      "num_input_tokens_seen": 94539135,
      "step": 4376,
      "time_per_iteration": 2.661550521850586
    },
    {
      "auxiliary_loss_clip": 0.01175087,
      "auxiliary_loss_mlp": 0.01026254,
      "balance_loss_clip": 1.04509163,
      "balance_loss_mlp": 1.01771235,
      "epoch": 0.5263031323273012,
      "flos": 22929466256640.0,
      "grad_norm": 2.0299732607252334,
      "language_loss": 0.73229957,
      "learning_rate": 1.9267930223230154e-06,
      "loss": 0.75431299,
      "num_input_tokens_seen": 94557610,
      "step": 4377,
      "time_per_iteration": 2.6578357219696045
    },
    {
      "auxiliary_loss_clip": 0.01188016,
      "auxiliary_loss_mlp": 0.01028249,
      "balance_loss_clip": 1.05157113,
      "balance_loss_mlp": 1.02034235,
      "epoch": 0.5264233752179402,
      "flos": 17748634049280.0,
      "grad_norm": 2.3201184135496953,
      "language_loss": 0.77665627,
      "learning_rate": 1.9260145782621224e-06,
      "loss": 0.79881895,
      "num_input_tokens_seen": 94575390,
      "step": 4378,
      "time_per_iteration": 2.5844943523406982
    },
    {
      "auxiliary_loss_clip": 0.01183163,
      "auxiliary_loss_mlp": 0.01026497,
      "balance_loss_clip": 1.05302811,
      "balance_loss_mlp": 1.01856947,
      "epoch": 0.5265436181085793,
      "flos": 24421626069120.0,
      "grad_norm": 1.944222411107852,
      "language_loss": 0.88137656,
      "learning_rate": 1.925236145424758e-06,
      "loss": 0.90347314,
      "num_input_tokens_seen": 94594210,
      "step": 4379,
      "time_per_iteration": 2.650576591491699
    },
    {
      "auxiliary_loss_clip": 0.0108615,
      "auxiliary_loss_mlp": 0.01000931,
      "balance_loss_clip": 1.01880622,
      "balance_loss_mlp": 1.00027585,
      "epoch": 0.5266638609992185,
      "flos": 69207298156800.0,
      "grad_norm": 0.7043198412945602,
      "language_loss": 0.57577533,
      "learning_rate": 1.924457723929012e-06,
      "loss": 0.59664613,
      "num_input_tokens_seen": 94665020,
      "step": 4380,
      "time_per_iteration": 3.3475282192230225
    },
    {
      "auxiliary_loss_clip": 0.01182202,
      "auxiliary_loss_mlp": 0.01022186,
      "balance_loss_clip": 1.05281615,
      "balance_loss_mlp": 1.01421368,
      "epoch": 0.5267841038898575,
      "flos": 20738699850240.0,
      "grad_norm": 1.5248672293898,
      "language_loss": 0.8323828,
      "learning_rate": 1.9236793138929685e-06,
      "loss": 0.85442662,
      "num_input_tokens_seen": 94684290,
      "step": 4381,
      "time_per_iteration": 2.735473394393921
    },
    {
      "auxiliary_loss_clip": 0.01185545,
      "auxiliary_loss_mlp": 0.01024741,
      "balance_loss_clip": 1.05101871,
      "balance_loss_mlp": 1.0166471,
      "epoch": 0.5269043467804966,
      "flos": 17234392988160.0,
      "grad_norm": 2.1805775768309896,
      "language_loss": 0.81346464,
      "learning_rate": 1.9229009154347133e-06,
      "loss": 0.83556747,
      "num_input_tokens_seen": 94701880,
      "step": 4382,
      "time_per_iteration": 2.597158432006836
    },
    {
      "auxiliary_loss_clip": 0.01174326,
      "auxiliary_loss_mlp": 0.01208057,
      "balance_loss_clip": 1.04463685,
      "balance_loss_mlp": 1.00082541,
      "epoch": 0.5270245896711357,
      "flos": 18223157646720.0,
      "grad_norm": 3.376147110229698,
      "language_loss": 0.81285453,
      "learning_rate": 1.922122528672327e-06,
      "loss": 0.83667833,
      "num_input_tokens_seen": 94720545,
      "step": 4383,
      "time_per_iteration": 2.726196765899658
    },
    {
      "auxiliary_loss_clip": 0.011743,
      "auxiliary_loss_mlp": 0.010215,
      "balance_loss_clip": 1.05161023,
      "balance_loss_mlp": 1.01385248,
      "epoch": 0.5271448325617748,
      "flos": 21287558643840.0,
      "grad_norm": 2.4002544921986426,
      "language_loss": 0.788638,
      "learning_rate": 1.9213441537238914e-06,
      "loss": 0.81059599,
      "num_input_tokens_seen": 94737420,
      "step": 4384,
      "time_per_iteration": 2.617640256881714
    },
    {
      "auxiliary_loss_clip": 0.01095821,
      "auxiliary_loss_mlp": 0.0100366,
      "balance_loss_clip": 1.02112198,
      "balance_loss_mlp": 1.00289679,
      "epoch": 0.5272650754524139,
      "flos": 65495497403520.0,
      "grad_norm": 0.8320313345790393,
      "language_loss": 0.57304096,
      "learning_rate": 1.920565790707485e-06,
      "loss": 0.59403574,
      "num_input_tokens_seen": 94802810,
      "step": 4385,
      "time_per_iteration": 3.327061653137207
    },
    {
      "auxiliary_loss_clip": 0.01196999,
      "auxiliary_loss_mlp": 0.0103026,
      "balance_loss_clip": 1.04725313,
      "balance_loss_mlp": 1.02167082,
      "epoch": 0.527385318343053,
      "flos": 19676426008320.0,
      "grad_norm": 2.1178470295387846,
      "language_loss": 0.66013074,
      "learning_rate": 1.9197874397411853e-06,
      "loss": 0.68240327,
      "num_input_tokens_seen": 94819440,
      "step": 4386,
      "time_per_iteration": 2.63649320602417
    },
    {
      "auxiliary_loss_clip": 0.01187341,
      "auxiliary_loss_mlp": 0.01030188,
      "balance_loss_clip": 1.04579628,
      "balance_loss_mlp": 1.02131879,
      "epoch": 0.5275055612336921,
      "flos": 12712018947840.0,
      "grad_norm": 3.381541302917638,
      "language_loss": 0.65902984,
      "learning_rate": 1.919009100943067e-06,
      "loss": 0.68120509,
      "num_input_tokens_seen": 94835130,
      "step": 4387,
      "time_per_iteration": 2.6210103034973145
    },
    {
      "auxiliary_loss_clip": 0.01201611,
      "auxiliary_loss_mlp": 0.01025955,
      "balance_loss_clip": 1.04957879,
      "balance_loss_mlp": 1.01728296,
      "epoch": 0.5276258041243311,
      "flos": 17749029098880.0,
      "grad_norm": 2.0789980414293163,
      "language_loss": 0.65646183,
      "learning_rate": 1.9182307744312043e-06,
      "loss": 0.67873752,
      "num_input_tokens_seen": 94852235,
      "step": 4388,
      "time_per_iteration": 3.576903820037842
    },
    {
      "auxiliary_loss_clip": 0.0118916,
      "auxiliary_loss_mlp": 0.0103095,
      "balance_loss_clip": 1.05075932,
      "balance_loss_mlp": 1.02301025,
      "epoch": 0.5277460470149702,
      "flos": 22710447077760.0,
      "grad_norm": 1.8559012304721558,
      "language_loss": 0.76772135,
      "learning_rate": 1.9174524603236676e-06,
      "loss": 0.78992248,
      "num_input_tokens_seen": 94871185,
      "step": 4389,
      "time_per_iteration": 2.6462273597717285
    },
    {
      "auxiliary_loss_clip": 0.01186787,
      "auxiliary_loss_mlp": 0.01023286,
      "balance_loss_clip": 1.05150986,
      "balance_loss_mlp": 1.01463699,
      "epoch": 0.5278662899056094,
      "flos": 19902699734400.0,
      "grad_norm": 1.9530606661601893,
      "language_loss": 0.76713228,
      "learning_rate": 1.916674158738527e-06,
      "loss": 0.78923297,
      "num_input_tokens_seen": 94890090,
      "step": 4390,
      "time_per_iteration": 3.6750526428222656
    },
    {
      "auxiliary_loss_clip": 0.01178474,
      "auxiliary_loss_mlp": 0.01209217,
      "balance_loss_clip": 1.04809737,
      "balance_loss_mlp": 1.00081122,
      "epoch": 0.5279865327962484,
      "flos": 18005215875840.0,
      "grad_norm": 2.123040196739957,
      "language_loss": 0.60588038,
      "learning_rate": 1.9158958697938506e-06,
      "loss": 0.62975734,
      "num_input_tokens_seen": 94908470,
      "step": 4391,
      "time_per_iteration": 2.619487762451172
    },
    {
      "auxiliary_loss_clip": 0.01179186,
      "auxiliary_loss_mlp": 0.01027931,
      "balance_loss_clip": 1.04892182,
      "balance_loss_mlp": 1.01977134,
      "epoch": 0.5281067756868875,
      "flos": 15924443892480.0,
      "grad_norm": 5.245341416141367,
      "language_loss": 0.85804677,
      "learning_rate": 1.9151175936077032e-06,
      "loss": 0.88011789,
      "num_input_tokens_seen": 94923440,
      "step": 4392,
      "time_per_iteration": 2.5774056911468506
    },
    {
      "auxiliary_loss_clip": 0.01176144,
      "auxiliary_loss_mlp": 0.01026825,
      "balance_loss_clip": 1.0510639,
      "balance_loss_mlp": 1.01827168,
      "epoch": 0.5282270185775266,
      "flos": 19426488197760.0,
      "grad_norm": 1.6643744942066536,
      "language_loss": 0.7941345,
      "learning_rate": 1.9143393302981507e-06,
      "loss": 0.8161642,
      "num_input_tokens_seen": 94941125,
      "step": 4393,
      "time_per_iteration": 3.529024124145508
    },
    {
      "auxiliary_loss_clip": 0.01187664,
      "auxiliary_loss_mlp": 0.01023848,
      "balance_loss_clip": 1.05113411,
      "balance_loss_mlp": 1.01573586,
      "epoch": 0.5283472614681657,
      "flos": 16399613934720.0,
      "grad_norm": 1.986290608807514,
      "language_loss": 0.83435273,
      "learning_rate": 1.913561079983252e-06,
      "loss": 0.8564679,
      "num_input_tokens_seen": 94959950,
      "step": 4394,
      "time_per_iteration": 2.6849637031555176
    },
    {
      "auxiliary_loss_clip": 0.01188998,
      "auxiliary_loss_mlp": 0.01033911,
      "balance_loss_clip": 1.05032659,
      "balance_loss_mlp": 1.02453542,
      "epoch": 0.5284675043588047,
      "flos": 26760524163840.0,
      "grad_norm": 3.7801664425808257,
      "language_loss": 0.74855685,
      "learning_rate": 1.9127828427810693e-06,
      "loss": 0.77078599,
      "num_input_tokens_seen": 94980515,
      "step": 4395,
      "time_per_iteration": 2.66575026512146
    },
    {
      "auxiliary_loss_clip": 0.01195203,
      "auxiliary_loss_mlp": 0.01027307,
      "balance_loss_clip": 1.0499773,
      "balance_loss_mlp": 1.0189209,
      "epoch": 0.5285877472494439,
      "flos": 19899898473600.0,
      "grad_norm": 3.3001205301800938,
      "language_loss": 0.81301427,
      "learning_rate": 1.9120046188096607e-06,
      "loss": 0.83523941,
      "num_input_tokens_seen": 94998560,
      "step": 4396,
      "time_per_iteration": 2.6695947647094727
    },
    {
      "auxiliary_loss_clip": 0.01182935,
      "auxiliary_loss_mlp": 0.01035408,
      "balance_loss_clip": 1.05273294,
      "balance_loss_mlp": 1.02724171,
      "epoch": 0.528707990140083,
      "flos": 20011257613440.0,
      "grad_norm": 2.455717905700701,
      "language_loss": 0.73970866,
      "learning_rate": 1.9112264081870804e-06,
      "loss": 0.76189208,
      "num_input_tokens_seen": 95016950,
      "step": 4397,
      "time_per_iteration": 2.644587278366089
    },
    {
      "auxiliary_loss_clip": 0.01186799,
      "auxiliary_loss_mlp": 0.01027845,
      "balance_loss_clip": 1.05173421,
      "balance_loss_mlp": 1.01908946,
      "epoch": 0.528828233030722,
      "flos": 20667956014080.0,
      "grad_norm": 2.2960845379395174,
      "language_loss": 0.75615203,
      "learning_rate": 1.9104482110313843e-06,
      "loss": 0.77829844,
      "num_input_tokens_seen": 95036540,
      "step": 4398,
      "time_per_iteration": 3.573491096496582
    },
    {
      "auxiliary_loss_clip": 0.01180059,
      "auxiliary_loss_mlp": 0.0102616,
      "balance_loss_clip": 1.05273795,
      "balance_loss_mlp": 1.01853025,
      "epoch": 0.5289484759213612,
      "flos": 25192448956800.0,
      "grad_norm": 1.801295387570374,
      "language_loss": 0.74571157,
      "learning_rate": 1.909670027460623e-06,
      "loss": 0.76777375,
      "num_input_tokens_seen": 95053840,
      "step": 4399,
      "time_per_iteration": 2.6443865299224854
    },
    {
      "auxiliary_loss_clip": 0.01181704,
      "auxiliary_loss_mlp": 0.0102852,
      "balance_loss_clip": 1.05223441,
      "balance_loss_mlp": 1.02009225,
      "epoch": 0.5290687188120002,
      "flos": 31139255715840.0,
      "grad_norm": 2.659236011667943,
      "language_loss": 0.71698511,
      "learning_rate": 1.908891857592847e-06,
      "loss": 0.73908734,
      "num_input_tokens_seen": 95074910,
      "step": 4400,
      "time_per_iteration": 2.69376802444458
    },
    {
      "auxiliary_loss_clip": 0.01180856,
      "auxiliary_loss_mlp": 0.01025199,
      "balance_loss_clip": 1.05114341,
      "balance_loss_mlp": 1.01637793,
      "epoch": 0.5291889617026393,
      "flos": 20119851406080.0,
      "grad_norm": 2.364882671841727,
      "language_loss": 0.90082598,
      "learning_rate": 1.9081137015461034e-06,
      "loss": 0.92288655,
      "num_input_tokens_seen": 95090985,
      "step": 4401,
      "time_per_iteration": 2.652279853820801
    },
    {
      "auxiliary_loss_clip": 0.01182379,
      "auxiliary_loss_mlp": 0.01026081,
      "balance_loss_clip": 1.0476532,
      "balance_loss_mlp": 1.01829636,
      "epoch": 0.5293092045932785,
      "flos": 19643747610240.0,
      "grad_norm": 1.8587395416375845,
      "language_loss": 0.90656227,
      "learning_rate": 1.9073355594384383e-06,
      "loss": 0.92864686,
      "num_input_tokens_seen": 95109225,
      "step": 4402,
      "time_per_iteration": 2.7029106616973877
    },
    {
      "auxiliary_loss_clip": 0.01179716,
      "auxiliary_loss_mlp": 0.01030068,
      "balance_loss_clip": 1.05044007,
      "balance_loss_mlp": 1.02221763,
      "epoch": 0.5294294474839175,
      "flos": 24317736958080.0,
      "grad_norm": 2.2697495766149043,
      "language_loss": 0.80524009,
      "learning_rate": 1.906557431387895e-06,
      "loss": 0.82733786,
      "num_input_tokens_seen": 95128215,
      "step": 4403,
      "time_per_iteration": 2.684020519256592
    },
    {
      "auxiliary_loss_clip": 0.01182795,
      "auxiliary_loss_mlp": 0.0103174,
      "balance_loss_clip": 1.05213988,
      "balance_loss_mlp": 1.02254903,
      "epoch": 0.5295496903745566,
      "flos": 18875941464960.0,
      "grad_norm": 2.2362374512981766,
      "language_loss": 0.79024005,
      "learning_rate": 1.905779317512516e-06,
      "loss": 0.81238538,
      "num_input_tokens_seen": 95145760,
      "step": 4404,
      "time_per_iteration": 2.6017448902130127
    },
    {
      "auxiliary_loss_clip": 0.01179525,
      "auxiliary_loss_mlp": 0.01027425,
      "balance_loss_clip": 1.05173516,
      "balance_loss_mlp": 1.01952446,
      "epoch": 0.5296699332651957,
      "flos": 20923101296640.0,
      "grad_norm": 2.718758645110497,
      "language_loss": 0.80662644,
      "learning_rate": 1.9050012179303385e-06,
      "loss": 0.82869595,
      "num_input_tokens_seen": 95164270,
      "step": 4405,
      "time_per_iteration": 2.6412508487701416
    },
    {
      "auxiliary_loss_clip": 0.01183351,
      "auxiliary_loss_mlp": 0.01028838,
      "balance_loss_clip": 1.0513804,
      "balance_loss_mlp": 1.02032638,
      "epoch": 0.5297901761558348,
      "flos": 22046745525120.0,
      "grad_norm": 2.1819872004613767,
      "language_loss": 0.69394839,
      "learning_rate": 1.904223132759401e-06,
      "loss": 0.71607029,
      "num_input_tokens_seen": 95182870,
      "step": 4406,
      "time_per_iteration": 2.580453872680664
    },
    {
      "auxiliary_loss_clip": 0.0118339,
      "auxiliary_loss_mlp": 0.01031753,
      "balance_loss_clip": 1.05222821,
      "balance_loss_mlp": 1.02309251,
      "epoch": 0.5299104190464738,
      "flos": 21798495653760.0,
      "grad_norm": 2.868106165432142,
      "language_loss": 0.69233632,
      "learning_rate": 1.9034450621177383e-06,
      "loss": 0.71448779,
      "num_input_tokens_seen": 95201190,
      "step": 4407,
      "time_per_iteration": 2.558574676513672
    },
    {
      "auxiliary_loss_clip": 0.01182226,
      "auxiliary_loss_mlp": 0.01037589,
      "balance_loss_clip": 1.05451095,
      "balance_loss_mlp": 1.02874947,
      "epoch": 0.530030661937113,
      "flos": 14720790119040.0,
      "grad_norm": 2.0658968403186293,
      "language_loss": 0.70184731,
      "learning_rate": 1.9026670061233824e-06,
      "loss": 0.72404546,
      "num_input_tokens_seen": 95218625,
      "step": 4408,
      "time_per_iteration": 2.506326198577881
    },
    {
      "auxiliary_loss_clip": 0.01178678,
      "auxiliary_loss_mlp": 0.01025173,
      "balance_loss_clip": 1.05197167,
      "balance_loss_mlp": 1.01717985,
      "epoch": 0.5301509048277521,
      "flos": 21251504367360.0,
      "grad_norm": 1.652839947581564,
      "language_loss": 0.80547041,
      "learning_rate": 1.901888964894365e-06,
      "loss": 0.82750893,
      "num_input_tokens_seen": 95237665,
      "step": 4409,
      "time_per_iteration": 2.5430920124053955
    },
    {
      "auxiliary_loss_clip": 0.01180364,
      "auxiliary_loss_mlp": 0.01027856,
      "balance_loss_clip": 1.05326104,
      "balance_loss_mlp": 1.01992285,
      "epoch": 0.5302711477183911,
      "flos": 25957058791680.0,
      "grad_norm": 2.7613498362249382,
      "language_loss": 0.67713737,
      "learning_rate": 1.9011109385487134e-06,
      "loss": 0.69921964,
      "num_input_tokens_seen": 95258915,
      "step": 4410,
      "time_per_iteration": 2.5025336742401123
    },
    {
      "auxiliary_loss_clip": 0.01180172,
      "auxiliary_loss_mlp": 0.01031114,
      "balance_loss_clip": 1.05280721,
      "balance_loss_mlp": 1.0223279,
      "epoch": 0.5303913906090303,
      "flos": 22273126992000.0,
      "grad_norm": 2.6026022458950675,
      "language_loss": 0.66388696,
      "learning_rate": 1.900332927204454e-06,
      "loss": 0.68599981,
      "num_input_tokens_seen": 95277365,
      "step": 4411,
      "time_per_iteration": 2.533661127090454
    },
    {
      "auxiliary_loss_clip": 0.0119086,
      "auxiliary_loss_mlp": 0.01025272,
      "balance_loss_clip": 1.05192614,
      "balance_loss_mlp": 1.01717186,
      "epoch": 0.5305116334996693,
      "flos": 24936010784640.0,
      "grad_norm": 2.765441116483154,
      "language_loss": 0.76884323,
      "learning_rate": 1.8995549309796097e-06,
      "loss": 0.79100454,
      "num_input_tokens_seen": 95296670,
      "step": 4412,
      "time_per_iteration": 2.675184726715088
    },
    {
      "auxiliary_loss_clip": 0.01187036,
      "auxiliary_loss_mlp": 0.01031536,
      "balance_loss_clip": 1.05339813,
      "balance_loss_mlp": 1.02336407,
      "epoch": 0.5306318763903084,
      "flos": 20189338266240.0,
      "grad_norm": 1.7900950608579094,
      "language_loss": 0.7692318,
      "learning_rate": 1.8987769499922028e-06,
      "loss": 0.79141748,
      "num_input_tokens_seen": 95315640,
      "step": 4413,
      "time_per_iteration": 2.6202633380889893
    },
    {
      "auxiliary_loss_clip": 0.01180751,
      "auxiliary_loss_mlp": 0.01208334,
      "balance_loss_clip": 1.05220842,
      "balance_loss_mlp": 1.00096846,
      "epoch": 0.5307521192809476,
      "flos": 20266366982400.0,
      "grad_norm": 2.109774304396809,
      "language_loss": 0.70855588,
      "learning_rate": 1.897998984360252e-06,
      "loss": 0.73244673,
      "num_input_tokens_seen": 95334610,
      "step": 4414,
      "time_per_iteration": 3.559924364089966
    },
    {
      "auxiliary_loss_clip": 0.0118131,
      "auxiliary_loss_mlp": 0.01025062,
      "balance_loss_clip": 1.05067873,
      "balance_loss_mlp": 1.01700974,
      "epoch": 0.5308723621715866,
      "flos": 28844276976000.0,
      "grad_norm": 1.4205240143499316,
      "language_loss": 0.78485364,
      "learning_rate": 1.897221034201775e-06,
      "loss": 0.80691731,
      "num_input_tokens_seen": 95358350,
      "step": 4415,
      "time_per_iteration": 2.7307934761047363
    },
    {
      "auxiliary_loss_clip": 0.01185085,
      "auxiliary_loss_mlp": 0.0102588,
      "balance_loss_clip": 1.04663897,
      "balance_loss_mlp": 1.01830459,
      "epoch": 0.5309926050622257,
      "flos": 27457766040960.0,
      "grad_norm": 1.7175343551120161,
      "language_loss": 0.66846108,
      "learning_rate": 1.8964430996347842e-06,
      "loss": 0.69057071,
      "num_input_tokens_seen": 95379900,
      "step": 4416,
      "time_per_iteration": 2.7050187587738037
    },
    {
      "auxiliary_loss_clip": 0.01182776,
      "auxiliary_loss_mlp": 0.01026977,
      "balance_loss_clip": 1.05050337,
      "balance_loss_mlp": 1.01848984,
      "epoch": 0.5311128479528648,
      "flos": 20514545026560.0,
      "grad_norm": 1.9319942561844734,
      "language_loss": 0.82513213,
      "learning_rate": 1.8956651807772931e-06,
      "loss": 0.84722972,
      "num_input_tokens_seen": 95397935,
      "step": 4417,
      "time_per_iteration": 3.571707248687744
    },
    {
      "auxiliary_loss_clip": 0.01178416,
      "auxiliary_loss_mlp": 0.01024237,
      "balance_loss_clip": 1.05141616,
      "balance_loss_mlp": 1.0169946,
      "epoch": 0.5312330908435039,
      "flos": 21397660807680.0,
      "grad_norm": 2.0363060645519067,
      "language_loss": 0.84082741,
      "learning_rate": 1.8948872777473115e-06,
      "loss": 0.86285388,
      "num_input_tokens_seen": 95415890,
      "step": 4418,
      "time_per_iteration": 2.630171060562134
    },
    {
      "auxiliary_loss_clip": 0.01183702,
      "auxiliary_loss_mlp": 0.01026814,
      "balance_loss_clip": 1.05118692,
      "balance_loss_mlp": 1.01934588,
      "epoch": 0.531353333734143,
      "flos": 24717350741760.0,
      "grad_norm": 2.417088779177122,
      "language_loss": 0.63467836,
      "learning_rate": 1.8941093906628458e-06,
      "loss": 0.65678346,
      "num_input_tokens_seen": 95433675,
      "step": 4419,
      "time_per_iteration": 3.6583261489868164
    },
    {
      "auxiliary_loss_clip": 0.01177799,
      "auxiliary_loss_mlp": 0.0102155,
      "balance_loss_clip": 1.04836106,
      "balance_loss_mlp": 1.01400995,
      "epoch": 0.531473576624782,
      "flos": 30480689808000.0,
      "grad_norm": 1.6082153105724735,
      "language_loss": 0.70709932,
      "learning_rate": 1.893331519641902e-06,
      "loss": 0.72909284,
      "num_input_tokens_seen": 95455820,
      "step": 4420,
      "time_per_iteration": 2.677919387817383
    },
    {
      "auxiliary_loss_clip": 0.01172262,
      "auxiliary_loss_mlp": 0.01026047,
      "balance_loss_clip": 1.04333925,
      "balance_loss_mlp": 1.01806581,
      "epoch": 0.5315938195154212,
      "flos": 23002975440000.0,
      "grad_norm": 2.4332825597202503,
      "language_loss": 0.74135953,
      "learning_rate": 1.8925536648024815e-06,
      "loss": 0.76334262,
      "num_input_tokens_seen": 95473240,
      "step": 4421,
      "time_per_iteration": 2.635742425918579
    },
    {
      "auxiliary_loss_clip": 0.01179032,
      "auxiliary_loss_mlp": 0.01024652,
      "balance_loss_clip": 1.05306089,
      "balance_loss_mlp": 1.01619983,
      "epoch": 0.5317140624060602,
      "flos": 22748584343040.0,
      "grad_norm": 3.525937720846596,
      "language_loss": 0.75923645,
      "learning_rate": 1.8917758262625849e-06,
      "loss": 0.78127331,
      "num_input_tokens_seen": 95493480,
      "step": 4422,
      "time_per_iteration": 2.5627353191375732
    },
    {
      "auxiliary_loss_clip": 0.01177411,
      "auxiliary_loss_mlp": 0.01025202,
      "balance_loss_clip": 1.05006099,
      "balance_loss_mlp": 1.0175426,
      "epoch": 0.5318343052966993,
      "flos": 22821087945600.0,
      "grad_norm": 1.7628993271962767,
      "language_loss": 0.80924058,
      "learning_rate": 1.8909980041402089e-06,
      "loss": 0.83126676,
      "num_input_tokens_seen": 95512075,
      "step": 4423,
      "time_per_iteration": 2.6440649032592773
    },
    {
      "auxiliary_loss_clip": 0.01175609,
      "auxiliary_loss_mlp": 0.01026478,
      "balance_loss_clip": 1.04957128,
      "balance_loss_mlp": 1.01778781,
      "epoch": 0.5319545481873384,
      "flos": 13626089274240.0,
      "grad_norm": 2.1595759345182493,
      "language_loss": 0.65776819,
      "learning_rate": 1.8902201985533494e-06,
      "loss": 0.67978907,
      "num_input_tokens_seen": 95529340,
      "step": 4424,
      "time_per_iteration": 2.5644702911376953
    },
    {
      "auxiliary_loss_clip": 0.01182315,
      "auxiliary_loss_mlp": 0.01024125,
      "balance_loss_clip": 1.04994798,
      "balance_loss_mlp": 1.0166564,
      "epoch": 0.5320747910779775,
      "flos": 22162522037760.0,
      "grad_norm": 2.114757229189475,
      "language_loss": 0.74840045,
      "learning_rate": 1.8894424096199983e-06,
      "loss": 0.7704649,
      "num_input_tokens_seen": 95548545,
      "step": 4425,
      "time_per_iteration": 3.5584020614624023
    },
    {
      "auxiliary_loss_clip": 0.01184385,
      "auxiliary_loss_mlp": 0.01027947,
      "balance_loss_clip": 1.05443847,
      "balance_loss_mlp": 1.01933396,
      "epoch": 0.5321950339686166,
      "flos": 18588081870720.0,
      "grad_norm": 2.177441615501803,
      "language_loss": 0.85886079,
      "learning_rate": 1.8886646374581463e-06,
      "loss": 0.88098407,
      "num_input_tokens_seen": 95567770,
      "step": 4426,
      "time_per_iteration": 2.562253713607788
    },
    {
      "auxiliary_loss_clip": 0.01179933,
      "auxiliary_loss_mlp": 0.01027667,
      "balance_loss_clip": 1.05112445,
      "balance_loss_mlp": 1.01885724,
      "epoch": 0.5323152768592557,
      "flos": 22856818999680.0,
      "grad_norm": 1.883838548051615,
      "language_loss": 0.71175206,
      "learning_rate": 1.8878868821857795e-06,
      "loss": 0.73382807,
      "num_input_tokens_seen": 95587420,
      "step": 4427,
      "time_per_iteration": 2.633228302001953
    },
    {
      "auxiliary_loss_clip": 0.01185031,
      "auxiliary_loss_mlp": 0.01028299,
      "balance_loss_clip": 1.04465103,
      "balance_loss_mlp": 1.01927471,
      "epoch": 0.5324355197498948,
      "flos": 33948690998400.0,
      "grad_norm": 2.157475831966326,
      "language_loss": 0.75429726,
      "learning_rate": 1.8871091439208838e-06,
      "loss": 0.77643055,
      "num_input_tokens_seen": 95609030,
      "step": 4428,
      "time_per_iteration": 2.8194735050201416
    },
    {
      "auxiliary_loss_clip": 0.01185096,
      "auxiliary_loss_mlp": 0.01028491,
      "balance_loss_clip": 1.04472983,
      "balance_loss_mlp": 1.02000332,
      "epoch": 0.5325557626405338,
      "flos": 23256720092160.0,
      "grad_norm": 2.373252586305965,
      "language_loss": 0.77403176,
      "learning_rate": 1.8863314227814414e-06,
      "loss": 0.79616761,
      "num_input_tokens_seen": 95627340,
      "step": 4429,
      "time_per_iteration": 2.6569736003875732
    },
    {
      "auxiliary_loss_clip": 0.0118852,
      "auxiliary_loss_mlp": 0.01031224,
      "balance_loss_clip": 1.05444837,
      "balance_loss_mlp": 1.02251577,
      "epoch": 0.532676005531173,
      "flos": 26718687797760.0,
      "grad_norm": 2.562829550926964,
      "language_loss": 0.48604298,
      "learning_rate": 1.8855537188854313e-06,
      "loss": 0.50824046,
      "num_input_tokens_seen": 95646315,
      "step": 4430,
      "time_per_iteration": 2.642922878265381
    },
    {
      "auxiliary_loss_clip": 0.01183021,
      "auxiliary_loss_mlp": 0.01030086,
      "balance_loss_clip": 1.05011964,
      "balance_loss_mlp": 1.02200329,
      "epoch": 0.5327962484218121,
      "flos": 17894610921600.0,
      "grad_norm": 2.437844527045066,
      "language_loss": 0.78990024,
      "learning_rate": 1.8847760323508315e-06,
      "loss": 0.81203133,
      "num_input_tokens_seen": 95665220,
      "step": 4431,
      "time_per_iteration": 2.5888662338256836
    },
    {
      "auxiliary_loss_clip": 0.01178603,
      "auxiliary_loss_mlp": 0.01026497,
      "balance_loss_clip": 1.04956591,
      "balance_loss_mlp": 1.01894498,
      "epoch": 0.5329164913124511,
      "flos": 17925385898880.0,
      "grad_norm": 1.8839273671214543,
      "language_loss": 0.75804585,
      "learning_rate": 1.883998363295616e-06,
      "loss": 0.78009683,
      "num_input_tokens_seen": 95682700,
      "step": 4432,
      "time_per_iteration": 2.5755767822265625
    },
    {
      "auxiliary_loss_clip": 0.01085893,
      "auxiliary_loss_mlp": 0.01001331,
      "balance_loss_clip": 1.01636958,
      "balance_loss_mlp": 1.00053203,
      "epoch": 0.5330367342030903,
      "flos": 57254178781440.0,
      "grad_norm": 0.873782459169121,
      "language_loss": 0.62643659,
      "learning_rate": 1.8832207118377565e-06,
      "loss": 0.64730883,
      "num_input_tokens_seen": 95738070,
      "step": 4433,
      "time_per_iteration": 3.0748612880706787
    },
    {
      "auxiliary_loss_clip": 0.01176835,
      "auxiliary_loss_mlp": 0.01024894,
      "balance_loss_clip": 1.05248415,
      "balance_loss_mlp": 1.01734221,
      "epoch": 0.5331569770937293,
      "flos": 17420518287360.0,
      "grad_norm": 1.9326682504176003,
      "language_loss": 0.69339067,
      "learning_rate": 1.882443078095222e-06,
      "loss": 0.71540797,
      "num_input_tokens_seen": 95756950,
      "step": 4434,
      "time_per_iteration": 2.532721996307373
    },
    {
      "auxiliary_loss_clip": 0.01103883,
      "auxiliary_loss_mlp": 0.0100111,
      "balance_loss_clip": 1.01695395,
      "balance_loss_mlp": 1.00027525,
      "epoch": 0.5332772199843684,
      "flos": 56750783627520.0,
      "grad_norm": 0.857183065190089,
      "language_loss": 0.66794169,
      "learning_rate": 1.8816654621859794e-06,
      "loss": 0.68899155,
      "num_input_tokens_seen": 95816615,
      "step": 4435,
      "time_per_iteration": 3.046879291534424
    },
    {
      "auxiliary_loss_clip": 0.01177487,
      "auxiliary_loss_mlp": 0.01026115,
      "balance_loss_clip": 1.05286396,
      "balance_loss_mlp": 1.01772237,
      "epoch": 0.5333974628750076,
      "flos": 18697753071360.0,
      "grad_norm": 2.2804551086790052,
      "language_loss": 0.72532427,
      "learning_rate": 1.8808878642279915e-06,
      "loss": 0.74736023,
      "num_input_tokens_seen": 95832020,
      "step": 4436,
      "time_per_iteration": 2.5891945362091064
    },
    {
      "auxiliary_loss_clip": 0.01187595,
      "auxiliary_loss_mlp": 0.0103539,
      "balance_loss_clip": 1.04557586,
      "balance_loss_mlp": 1.02667606,
      "epoch": 0.5335177057656466,
      "flos": 23805507058560.0,
      "grad_norm": 2.537267281449272,
      "language_loss": 0.65586364,
      "learning_rate": 1.8801102843392209e-06,
      "loss": 0.67809349,
      "num_input_tokens_seen": 95851425,
      "step": 4437,
      "time_per_iteration": 2.6468424797058105
    },
    {
      "auxiliary_loss_clip": 0.01183642,
      "auxiliary_loss_mlp": 0.01026563,
      "balance_loss_clip": 1.04578531,
      "balance_loss_mlp": 1.01842666,
      "epoch": 0.5336379486562857,
      "flos": 25078683605760.0,
      "grad_norm": 1.5466953828952104,
      "language_loss": 0.85262847,
      "learning_rate": 1.8793327226376238e-06,
      "loss": 0.87473047,
      "num_input_tokens_seen": 95870745,
      "step": 4438,
      "time_per_iteration": 2.737468957901001
    },
    {
      "auxiliary_loss_clip": 0.01191937,
      "auxiliary_loss_mlp": 0.01029336,
      "balance_loss_clip": 1.05073333,
      "balance_loss_mlp": 1.02090251,
      "epoch": 0.5337581915469248,
      "flos": 21396691140480.0,
      "grad_norm": 2.627867729360724,
      "language_loss": 0.80190665,
      "learning_rate": 1.8785551792411569e-06,
      "loss": 0.82411939,
      "num_input_tokens_seen": 95889755,
      "step": 4439,
      "time_per_iteration": 2.6497693061828613
    },
    {
      "auxiliary_loss_clip": 0.01185847,
      "auxiliary_loss_mlp": 0.01029021,
      "balance_loss_clip": 1.05225003,
      "balance_loss_mlp": 1.02140391,
      "epoch": 0.5338784344375639,
      "flos": 14865905064960.0,
      "grad_norm": 2.1510445674927756,
      "language_loss": 0.82341611,
      "learning_rate": 1.8777776542677733e-06,
      "loss": 0.84556478,
      "num_input_tokens_seen": 95907805,
      "step": 4440,
      "time_per_iteration": 2.664858102798462
    },
    {
      "auxiliary_loss_clip": 0.01184873,
      "auxiliary_loss_mlp": 0.01022582,
      "balance_loss_clip": 1.0457015,
      "balance_loss_mlp": 1.0142076,
      "epoch": 0.5339986773282029,
      "flos": 20813501923200.0,
      "grad_norm": 2.186707544014366,
      "language_loss": 0.73033637,
      "learning_rate": 1.8770001478354216e-06,
      "loss": 0.75241089,
      "num_input_tokens_seen": 95927480,
      "step": 4441,
      "time_per_iteration": 3.616687059402466
    },
    {
      "auxiliary_loss_clip": 0.01178345,
      "auxiliary_loss_mlp": 0.01033332,
      "balance_loss_clip": 1.0502727,
      "balance_loss_mlp": 1.02451611,
      "epoch": 0.5341189202188421,
      "flos": 17969089772160.0,
      "grad_norm": 2.3550818743413156,
      "language_loss": 0.84466004,
      "learning_rate": 1.8762226600620504e-06,
      "loss": 0.86677676,
      "num_input_tokens_seen": 95946095,
      "step": 4442,
      "time_per_iteration": 2.5701754093170166
    },
    {
      "auxiliary_loss_clip": 0.0119095,
      "auxiliary_loss_mlp": 0.01029439,
      "balance_loss_clip": 1.04943871,
      "balance_loss_mlp": 1.02052164,
      "epoch": 0.5342391631094812,
      "flos": 11031866328960.0,
      "grad_norm": 2.40614945461551,
      "language_loss": 0.59177828,
      "learning_rate": 1.8754451910656031e-06,
      "loss": 0.6139822,
      "num_input_tokens_seen": 95959995,
      "step": 4443,
      "time_per_iteration": 2.614464521408081
    },
    {
      "auxiliary_loss_clip": 0.01196867,
      "auxiliary_loss_mlp": 0.0102884,
      "balance_loss_clip": 1.04664016,
      "balance_loss_mlp": 1.02018237,
      "epoch": 0.5343594060001202,
      "flos": 15339135772800.0,
      "grad_norm": 1.8735030508951258,
      "language_loss": 0.8281377,
      "learning_rate": 1.8746677409640212e-06,
      "loss": 0.85039479,
      "num_input_tokens_seen": 95977095,
      "step": 4444,
      "time_per_iteration": 3.6233203411102295
    },
    {
      "auxiliary_loss_clip": 0.01187224,
      "auxiliary_loss_mlp": 0.01028736,
      "balance_loss_clip": 1.05475831,
      "balance_loss_mlp": 1.02017093,
      "epoch": 0.5344796488907594,
      "flos": 26900898514560.0,
      "grad_norm": 2.065181084306954,
      "language_loss": 0.8490752,
      "learning_rate": 1.8738903098752432e-06,
      "loss": 0.87123477,
      "num_input_tokens_seen": 95996225,
      "step": 4445,
      "time_per_iteration": 2.6343231201171875
    },
    {
      "auxiliary_loss_clip": 0.01185255,
      "auxiliary_loss_mlp": 0.01031207,
      "balance_loss_clip": 1.05075908,
      "balance_loss_mlp": 1.02324951,
      "epoch": 0.5345998917813984,
      "flos": 25411216740480.0,
      "grad_norm": 2.3749753956576716,
      "language_loss": 0.73204494,
      "learning_rate": 1.8731128979172052e-06,
      "loss": 0.75420952,
      "num_input_tokens_seen": 96015425,
      "step": 4446,
      "time_per_iteration": 3.5423529148101807
    },
    {
      "auxiliary_loss_clip": 0.01179422,
      "auxiliary_loss_mlp": 0.01024252,
      "balance_loss_clip": 1.04917157,
      "balance_loss_mlp": 1.01643753,
      "epoch": 0.5347201346720375,
      "flos": 32853379622400.0,
      "grad_norm": 2.0730152223686953,
      "language_loss": 0.67398703,
      "learning_rate": 1.8723355052078394e-06,
      "loss": 0.69602376,
      "num_input_tokens_seen": 96035460,
      "step": 4447,
      "time_per_iteration": 2.6480023860931396
    },
    {
      "auxiliary_loss_clip": 0.0117961,
      "auxiliary_loss_mlp": 0.0103501,
      "balance_loss_clip": 1.04982686,
      "balance_loss_mlp": 1.02615237,
      "epoch": 0.5348403775626767,
      "flos": 17967940536960.0,
      "grad_norm": 2.8943821268289835,
      "language_loss": 0.77002501,
      "learning_rate": 1.8715581318650765e-06,
      "loss": 0.79217118,
      "num_input_tokens_seen": 96054515,
      "step": 4448,
      "time_per_iteration": 2.598759174346924
    },
    {
      "auxiliary_loss_clip": 0.01198104,
      "auxiliary_loss_mlp": 0.01028761,
      "balance_loss_clip": 1.05112541,
      "balance_loss_mlp": 1.01921189,
      "epoch": 0.5349606204533157,
      "flos": 17603339535360.0,
      "grad_norm": 3.9807017191633256,
      "language_loss": 0.82277912,
      "learning_rate": 1.8707807780068422e-06,
      "loss": 0.84504777,
      "num_input_tokens_seen": 96072330,
      "step": 4449,
      "time_per_iteration": 2.626394033432007
    },
    {
      "auxiliary_loss_clip": 0.01180221,
      "auxiliary_loss_mlp": 0.01031036,
      "balance_loss_clip": 1.04825413,
      "balance_loss_mlp": 1.02313197,
      "epoch": 0.5350808633439548,
      "flos": 29167831710720.0,
      "grad_norm": 3.212672830503573,
      "language_loss": 0.66436267,
      "learning_rate": 1.8700034437510611e-06,
      "loss": 0.68647528,
      "num_input_tokens_seen": 96092425,
      "step": 4450,
      "time_per_iteration": 2.7616653442382812
    },
    {
      "auxiliary_loss_clip": 0.01176815,
      "auxiliary_loss_mlp": 0.01027805,
      "balance_loss_clip": 1.04739428,
      "balance_loss_mlp": 1.01910889,
      "epoch": 0.5352011062345938,
      "flos": 19499997381120.0,
      "grad_norm": 2.1709548773637164,
      "language_loss": 0.80838168,
      "learning_rate": 1.8692261292156549e-06,
      "loss": 0.83042794,
      "num_input_tokens_seen": 96111660,
      "step": 4451,
      "time_per_iteration": 3.5486338138580322
    },
    {
      "auxiliary_loss_clip": 0.0118039,
      "auxiliary_loss_mlp": 0.01024867,
      "balance_loss_clip": 1.05676746,
      "balance_loss_mlp": 1.01657557,
      "epoch": 0.535321349125233,
      "flos": 23477642691840.0,
      "grad_norm": 2.0599407261654332,
      "language_loss": 0.8096267,
      "learning_rate": 1.8684488345185401e-06,
      "loss": 0.83167917,
      "num_input_tokens_seen": 96131835,
      "step": 4452,
      "time_per_iteration": 2.6326324939727783
    },
    {
      "auxiliary_loss_clip": 0.01183738,
      "auxiliary_loss_mlp": 0.01027618,
      "balance_loss_clip": 1.05658114,
      "balance_loss_mlp": 1.01895142,
      "epoch": 0.535441592015872,
      "flos": 20478059786880.0,
      "grad_norm": 2.5304555818610197,
      "language_loss": 0.78631264,
      "learning_rate": 1.8676715597776332e-06,
      "loss": 0.80842626,
      "num_input_tokens_seen": 96150180,
      "step": 4453,
      "time_per_iteration": 2.6403045654296875
    },
    {
      "auxiliary_loss_clip": 0.01181096,
      "auxiliary_loss_mlp": 0.01025396,
      "balance_loss_clip": 1.04462099,
      "balance_loss_mlp": 1.01737332,
      "epoch": 0.5355618349065111,
      "flos": 19573147428480.0,
      "grad_norm": 1.8679261422653968,
      "language_loss": 0.76418364,
      "learning_rate": 1.8668943051108455e-06,
      "loss": 0.78624856,
      "num_input_tokens_seen": 96167485,
      "step": 4454,
      "time_per_iteration": 2.6461799144744873
    },
    {
      "auxiliary_loss_clip": 0.01183375,
      "auxiliary_loss_mlp": 0.01030817,
      "balance_loss_clip": 1.04955721,
      "balance_loss_mlp": 1.02220988,
      "epoch": 0.5356820777971503,
      "flos": 24024633978240.0,
      "grad_norm": 2.1294708521269574,
      "language_loss": 0.76149392,
      "learning_rate": 1.8661170706360856e-06,
      "loss": 0.78363585,
      "num_input_tokens_seen": 96186650,
      "step": 4455,
      "time_per_iteration": 2.651857376098633
    },
    {
      "auxiliary_loss_clip": 0.01180396,
      "auxiliary_loss_mlp": 0.01024686,
      "balance_loss_clip": 1.05235672,
      "balance_loss_mlp": 1.01696742,
      "epoch": 0.5358023206877893,
      "flos": 20884676722560.0,
      "grad_norm": 1.5345602697402432,
      "language_loss": 0.81260872,
      "learning_rate": 1.8653398564712594e-06,
      "loss": 0.83465952,
      "num_input_tokens_seen": 96205595,
      "step": 4456,
      "time_per_iteration": 2.596808671951294
    },
    {
      "auxiliary_loss_clip": 0.01179365,
      "auxiliary_loss_mlp": 0.01024984,
      "balance_loss_clip": 1.05281293,
      "balance_loss_mlp": 1.01694369,
      "epoch": 0.5359225635784284,
      "flos": 22418996123520.0,
      "grad_norm": 1.6626982266227197,
      "language_loss": 0.82294506,
      "learning_rate": 1.8645626627342704e-06,
      "loss": 0.84498847,
      "num_input_tokens_seen": 96226360,
      "step": 4457,
      "time_per_iteration": 2.618734836578369
    },
    {
      "auxiliary_loss_clip": 0.01186912,
      "auxiliary_loss_mlp": 0.01029955,
      "balance_loss_clip": 1.05342877,
      "balance_loss_mlp": 1.02193785,
      "epoch": 0.5360428064690675,
      "flos": 24097784025600.0,
      "grad_norm": 3.7779373057266876,
      "language_loss": 0.8079133,
      "learning_rate": 1.8637854895430172e-06,
      "loss": 0.83008194,
      "num_input_tokens_seen": 96245625,
      "step": 4458,
      "time_per_iteration": 2.5819954872131348
    },
    {
      "auxiliary_loss_clip": 0.01177636,
      "auxiliary_loss_mlp": 0.01024777,
      "balance_loss_clip": 1.04654229,
      "balance_loss_mlp": 1.01578856,
      "epoch": 0.5361630493597066,
      "flos": 21434505183360.0,
      "grad_norm": 2.656608970086626,
      "language_loss": 0.69165754,
      "learning_rate": 1.8630083370153978e-06,
      "loss": 0.7136817,
      "num_input_tokens_seen": 96265265,
      "step": 4459,
      "time_per_iteration": 2.6300878524780273
    },
    {
      "auxiliary_loss_clip": 0.01112102,
      "auxiliary_loss_mlp": 0.01000813,
      "balance_loss_clip": 1.02030981,
      "balance_loss_mlp": 0.9999308,
      "epoch": 0.5362832922503457,
      "flos": 68888696520960.0,
      "grad_norm": 0.744857572877265,
      "language_loss": 0.55343038,
      "learning_rate": 1.8622312052693041e-06,
      "loss": 0.57455957,
      "num_input_tokens_seen": 96326445,
      "step": 4460,
      "time_per_iteration": 3.4199306964874268
    },
    {
      "auxiliary_loss_clip": 0.01173229,
      "auxiliary_loss_mlp": 0.01027228,
      "balance_loss_clip": 1.04754376,
      "balance_loss_mlp": 1.01924694,
      "epoch": 0.5364035351409848,
      "flos": 9793702563840.0,
      "grad_norm": 2.384147384585283,
      "language_loss": 0.7170248,
      "learning_rate": 1.8614540944226267e-06,
      "loss": 0.73902941,
      "num_input_tokens_seen": 96343115,
      "step": 4461,
      "time_per_iteration": 2.8384382724761963
    },
    {
      "auxiliary_loss_clip": 0.0118065,
      "auxiliary_loss_mlp": 0.01026747,
      "balance_loss_clip": 1.05142999,
      "balance_loss_mlp": 1.01951087,
      "epoch": 0.5365237780316239,
      "flos": 23290080848640.0,
      "grad_norm": 2.3792733169989377,
      "language_loss": 0.68297064,
      "learning_rate": 1.8606770045932537e-06,
      "loss": 0.70504463,
      "num_input_tokens_seen": 96362230,
      "step": 4462,
      "time_per_iteration": 2.6646900177001953
    },
    {
      "auxiliary_loss_clip": 0.01178916,
      "auxiliary_loss_mlp": 0.01029668,
      "balance_loss_clip": 1.04405224,
      "balance_loss_mlp": 1.02052474,
      "epoch": 0.5366440209222629,
      "flos": 26578133879040.0,
      "grad_norm": 1.7535917065141668,
      "language_loss": 0.8168782,
      "learning_rate": 1.859899935899068e-06,
      "loss": 0.83896405,
      "num_input_tokens_seen": 96382085,
      "step": 4463,
      "time_per_iteration": 2.7739949226379395
    },
    {
      "auxiliary_loss_clip": 0.01182822,
      "auxiliary_loss_mlp": 0.01028361,
      "balance_loss_clip": 1.05229712,
      "balance_loss_mlp": 1.02013564,
      "epoch": 0.5367642638129021,
      "flos": 19608052469760.0,
      "grad_norm": 1.6723784986028554,
      "language_loss": 0.79099143,
      "learning_rate": 1.8591228884579506e-06,
      "loss": 0.8131032,
      "num_input_tokens_seen": 96400580,
      "step": 4464,
      "time_per_iteration": 2.6182045936584473
    },
    {
      "auxiliary_loss_clip": 0.01190939,
      "auxiliary_loss_mlp": 0.01026802,
      "balance_loss_clip": 1.0484643,
      "balance_loss_mlp": 1.01858282,
      "epoch": 0.5368845067035412,
      "flos": 23915214172800.0,
      "grad_norm": 3.801899393012318,
      "language_loss": 0.82061279,
      "learning_rate": 1.8583458623877795e-06,
      "loss": 0.84279013,
      "num_input_tokens_seen": 96419680,
      "step": 4465,
      "time_per_iteration": 2.722968101501465
    },
    {
      "auxiliary_loss_clip": 0.01184842,
      "auxiliary_loss_mlp": 0.01028248,
      "balance_loss_clip": 1.05359173,
      "balance_loss_mlp": 1.020172,
      "epoch": 0.5370047495941802,
      "flos": 16873131951360.0,
      "grad_norm": 2.074706402226932,
      "language_loss": 0.74091923,
      "learning_rate": 1.8575688578064281e-06,
      "loss": 0.76305014,
      "num_input_tokens_seen": 96437805,
      "step": 4466,
      "time_per_iteration": 2.6909842491149902
    },
    {
      "auxiliary_loss_clip": 0.01184413,
      "auxiliary_loss_mlp": 0.01029081,
      "balance_loss_clip": 1.05384135,
      "balance_loss_mlp": 1.02082026,
      "epoch": 0.5371249924848194,
      "flos": 20740926493440.0,
      "grad_norm": 1.667374878081428,
      "language_loss": 0.77107662,
      "learning_rate": 1.8567918748317674e-06,
      "loss": 0.79321158,
      "num_input_tokens_seen": 96457155,
      "step": 4467,
      "time_per_iteration": 2.5952558517456055
    },
    {
      "auxiliary_loss_clip": 0.01185063,
      "auxiliary_loss_mlp": 0.01027847,
      "balance_loss_clip": 1.04547787,
      "balance_loss_mlp": 1.01931167,
      "epoch": 0.5372452353754584,
      "flos": 17968120104960.0,
      "grad_norm": 1.8594296595459818,
      "language_loss": 0.82902753,
      "learning_rate": 1.8560149135816659e-06,
      "loss": 0.85115659,
      "num_input_tokens_seen": 96473990,
      "step": 4468,
      "time_per_iteration": 3.687005043029785
    },
    {
      "auxiliary_loss_clip": 0.01176181,
      "auxiliary_loss_mlp": 0.01025222,
      "balance_loss_clip": 1.04839206,
      "balance_loss_mlp": 1.01723766,
      "epoch": 0.5373654782660975,
      "flos": 15377021642880.0,
      "grad_norm": 2.6369516723323634,
      "language_loss": 0.85051572,
      "learning_rate": 1.8552379741739873e-06,
      "loss": 0.87252975,
      "num_input_tokens_seen": 96491335,
      "step": 4469,
      "time_per_iteration": 2.556497097015381
    },
    {
      "auxiliary_loss_clip": 0.01100956,
      "auxiliary_loss_mlp": 0.01199545,
      "balance_loss_clip": 1.02409911,
      "balance_loss_mlp": 1.00033152,
      "epoch": 0.5374857211567367,
      "flos": 69000091574400.0,
      "grad_norm": 0.9097296473076789,
      "language_loss": 0.55632555,
      "learning_rate": 1.8544610567265935e-06,
      "loss": 0.57933056,
      "num_input_tokens_seen": 96545275,
      "step": 4470,
      "time_per_iteration": 3.1706645488739014
    },
    {
      "auxiliary_loss_clip": 0.01184303,
      "auxiliary_loss_mlp": 0.0120855,
      "balance_loss_clip": 1.05182624,
      "balance_loss_mlp": 1.00094461,
      "epoch": 0.5376059640473757,
      "flos": 15085355207040.0,
      "grad_norm": 3.6277590512271116,
      "language_loss": 0.83109874,
      "learning_rate": 1.853684161357341e-06,
      "loss": 0.85502726,
      "num_input_tokens_seen": 96562935,
      "step": 4471,
      "time_per_iteration": 3.5928187370300293
    },
    {
      "auxiliary_loss_clip": 0.01179517,
      "auxiliary_loss_mlp": 0.01208698,
      "balance_loss_clip": 1.05210447,
      "balance_loss_mlp": 1.00088859,
      "epoch": 0.5377262069380148,
      "flos": 19792597570560.0,
      "grad_norm": 1.7536208052403062,
      "language_loss": 0.7694515,
      "learning_rate": 1.852907288184085e-06,
      "loss": 0.79333365,
      "num_input_tokens_seen": 96581820,
      "step": 4472,
      "time_per_iteration": 3.631908893585205
    },
    {
      "auxiliary_loss_clip": 0.01194187,
      "auxiliary_loss_mlp": 0.01026214,
      "balance_loss_clip": 1.04801941,
      "balance_loss_mlp": 1.01685643,
      "epoch": 0.5378464498286539,
      "flos": 30003077640960.0,
      "grad_norm": 1.7776659612701156,
      "language_loss": 0.70113778,
      "learning_rate": 1.8521304373246762e-06,
      "loss": 0.72334182,
      "num_input_tokens_seen": 96602865,
      "step": 4473,
      "time_per_iteration": 2.7379424571990967
    },
    {
      "auxiliary_loss_clip": 0.01185902,
      "auxiliary_loss_mlp": 0.01027238,
      "balance_loss_clip": 1.05204308,
      "balance_loss_mlp": 1.01879811,
      "epoch": 0.537966692719293,
      "flos": 21251217058560.0,
      "grad_norm": 2.6772808027495185,
      "language_loss": 0.88605738,
      "learning_rate": 1.8513536088969626e-06,
      "loss": 0.90818876,
      "num_input_tokens_seen": 96620530,
      "step": 4474,
      "time_per_iteration": 2.63271164894104
    },
    {
      "auxiliary_loss_clip": 0.01183954,
      "auxiliary_loss_mlp": 0.01034167,
      "balance_loss_clip": 1.05342913,
      "balance_loss_mlp": 1.02488613,
      "epoch": 0.538086935609932,
      "flos": 21543170803200.0,
      "grad_norm": 1.9720265585483703,
      "language_loss": 0.80276936,
      "learning_rate": 1.8505768030187884e-06,
      "loss": 0.82495058,
      "num_input_tokens_seen": 96640660,
      "step": 4475,
      "time_per_iteration": 2.6382741928100586
    },
    {
      "auxiliary_loss_clip": 0.0118154,
      "auxiliary_loss_mlp": 0.01024864,
      "balance_loss_clip": 1.05207169,
      "balance_loss_mlp": 1.01694822,
      "epoch": 0.5382071785005712,
      "flos": 22747219626240.0,
      "grad_norm": 1.5558115460570179,
      "language_loss": 0.79893655,
      "learning_rate": 1.849800019807995e-06,
      "loss": 0.82100058,
      "num_input_tokens_seen": 96661885,
      "step": 4476,
      "time_per_iteration": 2.6865382194519043
    },
    {
      "auxiliary_loss_clip": 0.01184442,
      "auxiliary_loss_mlp": 0.01027595,
      "balance_loss_clip": 1.04872227,
      "balance_loss_mlp": 1.01923251,
      "epoch": 0.5383274213912103,
      "flos": 24934574240640.0,
      "grad_norm": 2.7808763635697487,
      "language_loss": 0.71529448,
      "learning_rate": 1.8490232593824186e-06,
      "loss": 0.7374149,
      "num_input_tokens_seen": 96678340,
      "step": 4477,
      "time_per_iteration": 2.644817590713501
    },
    {
      "auxiliary_loss_clip": 0.01182584,
      "auxiliary_loss_mlp": 0.01024823,
      "balance_loss_clip": 1.05186856,
      "balance_loss_mlp": 1.01758075,
      "epoch": 0.5384476642818493,
      "flos": 22310186849280.0,
      "grad_norm": 1.542290297219973,
      "language_loss": 0.84849668,
      "learning_rate": 1.8482465218598935e-06,
      "loss": 0.87057078,
      "num_input_tokens_seen": 96698285,
      "step": 4478,
      "time_per_iteration": 3.566389322280884
    },
    {
      "auxiliary_loss_clip": 0.01188149,
      "auxiliary_loss_mlp": 0.01030139,
      "balance_loss_clip": 1.04951096,
      "balance_loss_mlp": 1.02148485,
      "epoch": 0.5385679071724885,
      "flos": 22711021695360.0,
      "grad_norm": 1.8830686876988105,
      "language_loss": 0.83454037,
      "learning_rate": 1.8474698073582508e-06,
      "loss": 0.85672325,
      "num_input_tokens_seen": 96719655,
      "step": 4479,
      "time_per_iteration": 2.6818063259124756
    },
    {
      "auxiliary_loss_clip": 0.01191818,
      "auxiliary_loss_mlp": 0.01023012,
      "balance_loss_clip": 1.04751301,
      "balance_loss_mlp": 1.01463771,
      "epoch": 0.5386881500631275,
      "flos": 15953746412160.0,
      "grad_norm": 2.0852529049360107,
      "language_loss": 0.87327838,
      "learning_rate": 1.8466931159953166e-06,
      "loss": 0.89542675,
      "num_input_tokens_seen": 96736290,
      "step": 4480,
      "time_per_iteration": 2.6198501586914062
    },
    {
      "auxiliary_loss_clip": 0.01189808,
      "auxiliary_loss_mlp": 0.0103034,
      "balance_loss_clip": 1.0536834,
      "balance_loss_mlp": 1.02201355,
      "epoch": 0.5388083929537666,
      "flos": 24060041809920.0,
      "grad_norm": 1.9290578016198687,
      "language_loss": 0.84293449,
      "learning_rate": 1.8459164478889158e-06,
      "loss": 0.86513603,
      "num_input_tokens_seen": 96757685,
      "step": 4481,
      "time_per_iteration": 2.8157684803009033
    },
    {
      "auxiliary_loss_clip": 0.01176664,
      "auxiliary_loss_mlp": 0.01025954,
      "balance_loss_clip": 1.04452908,
      "balance_loss_mlp": 1.01846194,
      "epoch": 0.5389286358444056,
      "flos": 22236893147520.0,
      "grad_norm": 1.7868870499380527,
      "language_loss": 0.7588712,
      "learning_rate": 1.8451398031568663e-06,
      "loss": 0.78089738,
      "num_input_tokens_seen": 96777310,
      "step": 4482,
      "time_per_iteration": 2.6855921745300293
    },
    {
      "auxiliary_loss_clip": 0.01188723,
      "auxiliary_loss_mlp": 0.01025884,
      "balance_loss_clip": 1.05038631,
      "balance_loss_mlp": 1.0175097,
      "epoch": 0.5390488787350448,
      "flos": 24281718595200.0,
      "grad_norm": 1.6195050419428765,
      "language_loss": 0.74608189,
      "learning_rate": 1.844363181916986e-06,
      "loss": 0.76822799,
      "num_input_tokens_seen": 96798035,
      "step": 4483,
      "time_per_iteration": 2.716245174407959
    },
    {
      "auxiliary_loss_clip": 0.01180824,
      "auxiliary_loss_mlp": 0.01029961,
      "balance_loss_clip": 1.05138791,
      "balance_loss_mlp": 1.02187264,
      "epoch": 0.5391691216256839,
      "flos": 16581393688320.0,
      "grad_norm": 1.9767617590253368,
      "language_loss": 0.83110917,
      "learning_rate": 1.8435865842870868e-06,
      "loss": 0.85321689,
      "num_input_tokens_seen": 96815975,
      "step": 4484,
      "time_per_iteration": 2.597062587738037
    },
    {
      "auxiliary_loss_clip": 0.01175195,
      "auxiliary_loss_mlp": 0.01208828,
      "balance_loss_clip": 1.04687595,
      "balance_loss_mlp": 1.00093269,
      "epoch": 0.5392893645163229,
      "flos": 23330049707520.0,
      "grad_norm": 1.8702776683683462,
      "language_loss": 0.71974397,
      "learning_rate": 1.8428100103849787e-06,
      "loss": 0.74358416,
      "num_input_tokens_seen": 96835770,
      "step": 4485,
      "time_per_iteration": 2.7087197303771973
    },
    {
      "auxiliary_loss_clip": 0.01186793,
      "auxiliary_loss_mlp": 0.01031029,
      "balance_loss_clip": 1.05447149,
      "balance_loss_mlp": 1.02217162,
      "epoch": 0.5394096074069621,
      "flos": 15669801400320.0,
      "grad_norm": 2.295881790241933,
      "language_loss": 0.73636764,
      "learning_rate": 1.842033460328467e-06,
      "loss": 0.75854582,
      "num_input_tokens_seen": 96854490,
      "step": 4486,
      "time_per_iteration": 2.6402766704559326
    },
    {
      "auxiliary_loss_clip": 0.01184397,
      "auxiliary_loss_mlp": 0.01208134,
      "balance_loss_clip": 1.04792035,
      "balance_loss_mlp": 1.0009613,
      "epoch": 0.5395298502976011,
      "flos": 22893447893760.0,
      "grad_norm": 1.7260725335913325,
      "language_loss": 0.75256646,
      "learning_rate": 1.8412569342353541e-06,
      "loss": 0.77649176,
      "num_input_tokens_seen": 96874645,
      "step": 4487,
      "time_per_iteration": 2.7102866172790527
    },
    {
      "auxiliary_loss_clip": 0.01191543,
      "auxiliary_loss_mlp": 0.01032556,
      "balance_loss_clip": 1.05371809,
      "balance_loss_mlp": 1.02340674,
      "epoch": 0.5396500931882402,
      "flos": 23842135952640.0,
      "grad_norm": 3.829217923986189,
      "language_loss": 0.84694749,
      "learning_rate": 1.840480432223438e-06,
      "loss": 0.86918855,
      "num_input_tokens_seen": 96893650,
      "step": 4488,
      "time_per_iteration": 2.679185628890991
    },
    {
      "auxiliary_loss_clip": 0.01186708,
      "auxiliary_loss_mlp": 0.01031761,
      "balance_loss_clip": 1.04996932,
      "balance_loss_mlp": 1.0234288,
      "epoch": 0.5397703360788794,
      "flos": 26322988596480.0,
      "grad_norm": 2.2645060109105617,
      "language_loss": 0.77825868,
      "learning_rate": 1.8397039544105131e-06,
      "loss": 0.80044335,
      "num_input_tokens_seen": 96912735,
      "step": 4489,
      "time_per_iteration": 2.715629816055298
    },
    {
      "auxiliary_loss_clip": 0.01179382,
      "auxiliary_loss_mlp": 0.01026259,
      "balance_loss_clip": 1.04710877,
      "balance_loss_mlp": 1.01783061,
      "epoch": 0.5398905789695184,
      "flos": 21214588164480.0,
      "grad_norm": 3.7488837974777844,
      "language_loss": 0.7029494,
      "learning_rate": 1.8389275009143711e-06,
      "loss": 0.72500575,
      "num_input_tokens_seen": 96932475,
      "step": 4490,
      "time_per_iteration": 2.626880645751953
    },
    {
      "auxiliary_loss_clip": 0.01178219,
      "auxiliary_loss_mlp": 0.01027829,
      "balance_loss_clip": 1.05316687,
      "balance_loss_mlp": 1.0199132,
      "epoch": 0.5400108218601575,
      "flos": 25080335631360.0,
      "grad_norm": 1.7171101361302101,
      "language_loss": 0.73617417,
      "learning_rate": 1.8381510718527988e-06,
      "loss": 0.75823462,
      "num_input_tokens_seen": 96952085,
      "step": 4491,
      "time_per_iteration": 2.6381735801696777
    },
    {
      "auxiliary_loss_clip": 0.01188168,
      "auxiliary_loss_mlp": 0.01030517,
      "balance_loss_clip": 1.04921961,
      "balance_loss_mlp": 1.02206516,
      "epoch": 0.5401310647507966,
      "flos": 26357498588160.0,
      "grad_norm": 2.0985713419657865,
      "language_loss": 0.63810223,
      "learning_rate": 1.8373746673435812e-06,
      "loss": 0.66028905,
      "num_input_tokens_seen": 96973110,
      "step": 4492,
      "time_per_iteration": 2.675340175628662
    },
    {
      "auxiliary_loss_clip": 0.01182315,
      "auxiliary_loss_mlp": 0.01026864,
      "balance_loss_clip": 1.05567098,
      "balance_loss_mlp": 1.01840651,
      "epoch": 0.5402513076414357,
      "flos": 27855332749440.0,
      "grad_norm": 1.7120277643164972,
      "language_loss": 0.79178858,
      "learning_rate": 1.8365982875044964e-06,
      "loss": 0.81388044,
      "num_input_tokens_seen": 96993420,
      "step": 4493,
      "time_per_iteration": 2.6485092639923096
    },
    {
      "auxiliary_loss_clip": 0.01187005,
      "auxiliary_loss_mlp": 0.01209015,
      "balance_loss_clip": 1.0527494,
      "balance_loss_mlp": 1.00090289,
      "epoch": 0.5403715505320748,
      "flos": 22893771116160.0,
      "grad_norm": 2.408144725566726,
      "language_loss": 0.75880003,
      "learning_rate": 1.8358219324533217e-06,
      "loss": 0.78276026,
      "num_input_tokens_seen": 97013685,
      "step": 4494,
      "time_per_iteration": 2.6257073879241943
    },
    {
      "auxiliary_loss_clip": 0.01178382,
      "auxiliary_loss_mlp": 0.01025266,
      "balance_loss_clip": 1.04855454,
      "balance_loss_mlp": 1.01785135,
      "epoch": 0.5404917934227139,
      "flos": 30224143895040.0,
      "grad_norm": 1.566986159586225,
      "language_loss": 0.70230186,
      "learning_rate": 1.8350456023078292e-06,
      "loss": 0.72433841,
      "num_input_tokens_seen": 97036060,
      "step": 4495,
      "time_per_iteration": 3.621500015258789
    },
    {
      "auxiliary_loss_clip": 0.01186213,
      "auxiliary_loss_mlp": 0.01032193,
      "balance_loss_clip": 1.0560149,
      "balance_loss_mlp": 1.02353859,
      "epoch": 0.540612036313353,
      "flos": 19938502615680.0,
      "grad_norm": 2.090712225172776,
      "language_loss": 0.78337127,
      "learning_rate": 1.8342692971857874e-06,
      "loss": 0.80555534,
      "num_input_tokens_seen": 97055260,
      "step": 4496,
      "time_per_iteration": 2.650125503540039
    },
    {
      "auxiliary_loss_clip": 0.01183976,
      "auxiliary_loss_mlp": 0.01028404,
      "balance_loss_clip": 1.05188084,
      "balance_loss_mlp": 1.02043509,
      "epoch": 0.540732279203992,
      "flos": 24279599692800.0,
      "grad_norm": 3.147372000038409,
      "language_loss": 0.71816909,
      "learning_rate": 1.833493017204962e-06,
      "loss": 0.74029291,
      "num_input_tokens_seen": 97075365,
      "step": 4497,
      "time_per_iteration": 3.6209874153137207
    },
    {
      "auxiliary_loss_clip": 0.01180519,
      "auxiliary_loss_mlp": 0.01032425,
      "balance_loss_clip": 1.05383134,
      "balance_loss_mlp": 1.02449751,
      "epoch": 0.5408525220946312,
      "flos": 20193216935040.0,
      "grad_norm": 1.7376304457449263,
      "language_loss": 0.78129852,
      "learning_rate": 1.8327167624831134e-06,
      "loss": 0.80342793,
      "num_input_tokens_seen": 97093095,
      "step": 4498,
      "time_per_iteration": 3.537209987640381
    },
    {
      "auxiliary_loss_clip": 0.01178615,
      "auxiliary_loss_mlp": 0.01024564,
      "balance_loss_clip": 1.05410171,
      "balance_loss_mlp": 1.0164578,
      "epoch": 0.5409727649852702,
      "flos": 24134448833280.0,
      "grad_norm": 8.265244598062289,
      "language_loss": 0.70823383,
      "learning_rate": 1.831940533137999e-06,
      "loss": 0.73026556,
      "num_input_tokens_seen": 97112000,
      "step": 4499,
      "time_per_iteration": 2.657712697982788
    },
    {
      "auxiliary_loss_clip": 0.01179181,
      "auxiliary_loss_mlp": 0.01025504,
      "balance_loss_clip": 1.05380595,
      "balance_loss_mlp": 1.01765442,
      "epoch": 0.5410930078759093,
      "flos": 23912700220800.0,
      "grad_norm": 1.9654225219276082,
      "language_loss": 0.72892118,
      "learning_rate": 1.8311643292873718e-06,
      "loss": 0.75096804,
      "num_input_tokens_seen": 97130820,
      "step": 4500,
      "time_per_iteration": 2.6513004302978516
    },
    {
      "auxiliary_loss_clip": 0.01179756,
      "auxiliary_loss_mlp": 0.01027988,
      "balance_loss_clip": 1.0531069,
      "balance_loss_mlp": 1.02030528,
      "epoch": 0.5412132507665485,
      "flos": 21105132445440.0,
      "grad_norm": 1.8738374561829063,
      "language_loss": 0.88343298,
      "learning_rate": 1.8303881510489818e-06,
      "loss": 0.90551043,
      "num_input_tokens_seen": 97149210,
      "step": 4501,
      "time_per_iteration": 2.6354713439941406
    },
    {
      "auxiliary_loss_clip": 0.01185976,
      "auxiliary_loss_mlp": 0.0103101,
      "balance_loss_clip": 1.05200934,
      "balance_loss_mlp": 1.02144921,
      "epoch": 0.5413334936571875,
      "flos": 30227340205440.0,
      "grad_norm": 2.857604182795852,
      "language_loss": 0.6928851,
      "learning_rate": 1.829611998540574e-06,
      "loss": 0.71505499,
      "num_input_tokens_seen": 97170415,
      "step": 4502,
      "time_per_iteration": 2.712336778640747
    },
    {
      "auxiliary_loss_clip": 0.01183393,
      "auxiliary_loss_mlp": 0.01208669,
      "balance_loss_clip": 1.05110836,
      "balance_loss_mlp": 1.00089228,
      "epoch": 0.5414537365478266,
      "flos": 24279635606400.0,
      "grad_norm": 3.2812813192924533,
      "language_loss": 0.79812288,
      "learning_rate": 1.8288358718798914e-06,
      "loss": 0.82204354,
      "num_input_tokens_seen": 97189605,
      "step": 4503,
      "time_per_iteration": 2.682473659515381
    },
    {
      "auxiliary_loss_clip": 0.01177551,
      "auxiliary_loss_mlp": 0.01208604,
      "balance_loss_clip": 1.05296516,
      "balance_loss_mlp": 1.00091195,
      "epoch": 0.5415739794384657,
      "flos": 16654543735680.0,
      "grad_norm": 1.9578297922836512,
      "language_loss": 0.72732747,
      "learning_rate": 1.8280597711846703e-06,
      "loss": 0.75118899,
      "num_input_tokens_seen": 97207845,
      "step": 4504,
      "time_per_iteration": 2.598109006881714
    },
    {
      "auxiliary_loss_clip": 0.01179148,
      "auxiliary_loss_mlp": 0.01025953,
      "balance_loss_clip": 1.05367506,
      "balance_loss_mlp": 1.01782906,
      "epoch": 0.5416942223291048,
      "flos": 23185724860800.0,
      "grad_norm": 1.8496554595728167,
      "language_loss": 0.8385216,
      "learning_rate": 1.8272836965726455e-06,
      "loss": 0.86057258,
      "num_input_tokens_seen": 97226780,
      "step": 4505,
      "time_per_iteration": 3.563589572906494
    },
    {
      "auxiliary_loss_clip": 0.01193819,
      "auxiliary_loss_mlp": 0.01031187,
      "balance_loss_clip": 1.04527164,
      "balance_loss_mlp": 1.02229416,
      "epoch": 0.5418144652197439,
      "flos": 20303247271680.0,
      "grad_norm": 1.7370155942691334,
      "language_loss": 0.78499138,
      "learning_rate": 1.8265076481615461e-06,
      "loss": 0.8072415,
      "num_input_tokens_seen": 97246695,
      "step": 4506,
      "time_per_iteration": 2.724717855453491
    },
    {
      "auxiliary_loss_clip": 0.01184012,
      "auxiliary_loss_mlp": 0.01029459,
      "balance_loss_clip": 1.05268824,
      "balance_loss_mlp": 1.02072692,
      "epoch": 0.541934708110383,
      "flos": 12458633431680.0,
      "grad_norm": 2.7241654387739853,
      "language_loss": 0.8769151,
      "learning_rate": 1.8257316260690987e-06,
      "loss": 0.89904976,
      "num_input_tokens_seen": 97264480,
      "step": 4507,
      "time_per_iteration": 2.623145818710327
    },
    {
      "auxiliary_loss_clip": 0.01182589,
      "auxiliary_loss_mlp": 0.0102013,
      "balance_loss_clip": 1.05206132,
      "balance_loss_mlp": 1.01284003,
      "epoch": 0.5420549510010221,
      "flos": 21253802837760.0,
      "grad_norm": 1.6152437026936206,
      "language_loss": 0.76038885,
      "learning_rate": 1.8249556304130254e-06,
      "loss": 0.78241605,
      "num_input_tokens_seen": 97285760,
      "step": 4508,
      "time_per_iteration": 2.680795431137085
    },
    {
      "auxiliary_loss_clip": 0.01174331,
      "auxiliary_loss_mlp": 0.01025405,
      "balance_loss_clip": 1.04844046,
      "balance_loss_mlp": 1.0168215,
      "epoch": 0.5421751938916611,
      "flos": 29490524519040.0,
      "grad_norm": 2.6107105613199058,
      "language_loss": 0.69090259,
      "learning_rate": 1.824179661311044e-06,
      "loss": 0.71289992,
      "num_input_tokens_seen": 97304510,
      "step": 4509,
      "time_per_iteration": 2.6671853065490723
    },
    {
      "auxiliary_loss_clip": 0.0118818,
      "auxiliary_loss_mlp": 0.01028568,
      "balance_loss_clip": 1.04401696,
      "balance_loss_mlp": 1.02011597,
      "epoch": 0.5422954367823003,
      "flos": 18734238311040.0,
      "grad_norm": 1.8155943480694103,
      "language_loss": 0.79395533,
      "learning_rate": 1.823403718880868e-06,
      "loss": 0.81612289,
      "num_input_tokens_seen": 97323270,
      "step": 4510,
      "time_per_iteration": 2.6987357139587402
    },
    {
      "auxiliary_loss_clip": 0.01183163,
      "auxiliary_loss_mlp": 0.01035035,
      "balance_loss_clip": 1.04705095,
      "balance_loss_mlp": 1.02646375,
      "epoch": 0.5424156796729394,
      "flos": 39969006940800.0,
      "grad_norm": 1.7861187454054306,
      "language_loss": 0.66648024,
      "learning_rate": 1.822627803240207e-06,
      "loss": 0.68866223,
      "num_input_tokens_seen": 97345600,
      "step": 4511,
      "time_per_iteration": 2.804327964782715
    },
    {
      "auxiliary_loss_clip": 0.01192824,
      "auxiliary_loss_mlp": 0.0103113,
      "balance_loss_clip": 1.05146825,
      "balance_loss_mlp": 1.02333069,
      "epoch": 0.5425359225635784,
      "flos": 11546538353280.0,
      "grad_norm": 2.7889354921515115,
      "language_loss": 0.85344362,
      "learning_rate": 1.8218519145067675e-06,
      "loss": 0.87568307,
      "num_input_tokens_seen": 97361220,
      "step": 4512,
      "time_per_iteration": 2.665969133377075
    },
    {
      "auxiliary_loss_clip": 0.01180872,
      "auxiliary_loss_mlp": 0.01030189,
      "balance_loss_clip": 1.04697394,
      "balance_loss_mlp": 1.02201748,
      "epoch": 0.5426561654542175,
      "flos": 20229702174720.0,
      "grad_norm": 1.920131604107165,
      "language_loss": 0.89481783,
      "learning_rate": 1.8210760527982508e-06,
      "loss": 0.91692847,
      "num_input_tokens_seen": 97381505,
      "step": 4513,
      "time_per_iteration": 2.6740846633911133
    },
    {
      "auxiliary_loss_clip": 0.01185783,
      "auxiliary_loss_mlp": 0.01208612,
      "balance_loss_clip": 1.05091548,
      "balance_loss_mlp": 1.00078511,
      "epoch": 0.5427764083448566,
      "flos": 21871681614720.0,
      "grad_norm": 1.87571985325986,
      "language_loss": 0.75531149,
      "learning_rate": 1.8203002182323552e-06,
      "loss": 0.77925545,
      "num_input_tokens_seen": 97399060,
      "step": 4514,
      "time_per_iteration": 2.6819729804992676
    },
    {
      "auxiliary_loss_clip": 0.01186024,
      "auxiliary_loss_mlp": 0.01024907,
      "balance_loss_clip": 1.05193889,
      "balance_loss_mlp": 1.01637745,
      "epoch": 0.5428966512354957,
      "flos": 19640946349440.0,
      "grad_norm": 2.0847009202783413,
      "language_loss": 0.76088279,
      "learning_rate": 1.819524410926773e-06,
      "loss": 0.78299212,
      "num_input_tokens_seen": 97416740,
      "step": 4515,
      "time_per_iteration": 2.582648515701294
    },
    {
      "auxiliary_loss_clip": 0.01191673,
      "auxiliary_loss_mlp": 0.01027536,
      "balance_loss_clip": 1.04799736,
      "balance_loss_mlp": 1.01906586,
      "epoch": 0.5430168941261347,
      "flos": 22382187661440.0,
      "grad_norm": 1.4996904146621413,
      "language_loss": 0.77044487,
      "learning_rate": 1.8187486309991944e-06,
      "loss": 0.79263693,
      "num_input_tokens_seen": 97437620,
      "step": 4516,
      "time_per_iteration": 2.7650413513183594
    },
    {
      "auxiliary_loss_clip": 0.01188328,
      "auxiliary_loss_mlp": 0.01033298,
      "balance_loss_clip": 1.05489385,
      "balance_loss_mlp": 1.02560925,
      "epoch": 0.5431371370167739,
      "flos": 18764187275520.0,
      "grad_norm": 1.6841226831902596,
      "language_loss": 0.7762478,
      "learning_rate": 1.817972878567304e-06,
      "loss": 0.79846406,
      "num_input_tokens_seen": 97456275,
      "step": 4517,
      "time_per_iteration": 2.6289970874786377
    },
    {
      "auxiliary_loss_clip": 0.01188553,
      "auxiliary_loss_mlp": 0.01028364,
      "balance_loss_clip": 1.04977238,
      "balance_loss_mlp": 1.02060318,
      "epoch": 0.543257379907413,
      "flos": 18806023641600.0,
      "grad_norm": 1.8086193738392096,
      "language_loss": 0.76429355,
      "learning_rate": 1.8171971537487834e-06,
      "loss": 0.78646278,
      "num_input_tokens_seen": 97474925,
      "step": 4518,
      "time_per_iteration": 2.645268201828003
    },
    {
      "auxiliary_loss_clip": 0.01179825,
      "auxiliary_loss_mlp": 0.01029813,
      "balance_loss_clip": 1.05262876,
      "balance_loss_mlp": 1.0210247,
      "epoch": 0.543377622798052,
      "flos": 17493381025920.0,
      "grad_norm": 1.9443379214226766,
      "language_loss": 0.80706894,
      "learning_rate": 1.8164214566613093e-06,
      "loss": 0.82916534,
      "num_input_tokens_seen": 97493550,
      "step": 4519,
      "time_per_iteration": 2.5847971439361572
    },
    {
      "auxiliary_loss_clip": 0.01179813,
      "auxiliary_loss_mlp": 0.01024541,
      "balance_loss_clip": 1.05409145,
      "balance_loss_mlp": 1.01639855,
      "epoch": 0.5434978656886912,
      "flos": 18989311766400.0,
      "grad_norm": 3.0478112073488837,
      "language_loss": 0.6598022,
      "learning_rate": 1.8156457874225547e-06,
      "loss": 0.68184578,
      "num_input_tokens_seen": 97512010,
      "step": 4520,
      "time_per_iteration": 2.5471816062927246
    },
    {
      "auxiliary_loss_clip": 0.011728,
      "auxiliary_loss_mlp": 0.0102436,
      "balance_loss_clip": 1.04942274,
      "balance_loss_mlp": 1.01634884,
      "epoch": 0.5436181085793302,
      "flos": 17274936464640.0,
      "grad_norm": 1.9547034886057437,
      "language_loss": 0.81201345,
      "learning_rate": 1.814870146150187e-06,
      "loss": 0.83398509,
      "num_input_tokens_seen": 97530120,
      "step": 4521,
      "time_per_iteration": 3.5166192054748535
    },
    {
      "auxiliary_loss_clip": 0.01191935,
      "auxiliary_loss_mlp": 0.0103378,
      "balance_loss_clip": 1.05058217,
      "balance_loss_mlp": 1.02550101,
      "epoch": 0.5437383514699693,
      "flos": 19098587917440.0,
      "grad_norm": 2.5962583091591602,
      "language_loss": 0.79091787,
      "learning_rate": 1.814094532961871e-06,
      "loss": 0.81317502,
      "num_input_tokens_seen": 97548695,
      "step": 4522,
      "time_per_iteration": 2.6385138034820557
    },
    {
      "auxiliary_loss_clip": 0.01191697,
      "auxiliary_loss_mlp": 0.0103135,
      "balance_loss_clip": 1.04621339,
      "balance_loss_mlp": 1.02285075,
      "epoch": 0.5438585943606085,
      "flos": 22602715211520.0,
      "grad_norm": 2.0660206646731374,
      "language_loss": 0.83744097,
      "learning_rate": 1.8133189479752666e-06,
      "loss": 0.85967147,
      "num_input_tokens_seen": 97567625,
      "step": 4523,
      "time_per_iteration": 3.5648200511932373
    },
    {
      "auxiliary_loss_clip": 0.01177449,
      "auxiliary_loss_mlp": 0.01026433,
      "balance_loss_clip": 1.05309546,
      "balance_loss_mlp": 1.01895547,
      "epoch": 0.5439788372512475,
      "flos": 21798495653760.0,
      "grad_norm": 2.0483600065889056,
      "language_loss": 0.81702542,
      "learning_rate": 1.8125433913080292e-06,
      "loss": 0.83906424,
      "num_input_tokens_seen": 97585325,
      "step": 4524,
      "time_per_iteration": 2.603389024734497
    },
    {
      "auxiliary_loss_clip": 0.01188762,
      "auxiliary_loss_mlp": 0.01028382,
      "balance_loss_clip": 1.03973961,
      "balance_loss_mlp": 1.02144134,
      "epoch": 0.5440990801418866,
      "flos": 16399362539520.0,
      "grad_norm": 2.4264883608333,
      "language_loss": 0.82753068,
      "learning_rate": 1.811767863077811e-06,
      "loss": 0.84970212,
      "num_input_tokens_seen": 97604275,
      "step": 4525,
      "time_per_iteration": 3.6755549907684326
    },
    {
      "auxiliary_loss_clip": 0.01181681,
      "auxiliary_loss_mlp": 0.01028677,
      "balance_loss_clip": 1.04575014,
      "balance_loss_mlp": 1.02055871,
      "epoch": 0.5442193230325257,
      "flos": 21615638492160.0,
      "grad_norm": 1.609159726396947,
      "language_loss": 0.78588122,
      "learning_rate": 1.8109923634022577e-06,
      "loss": 0.80798483,
      "num_input_tokens_seen": 97624300,
      "step": 4526,
      "time_per_iteration": 2.7738327980041504
    },
    {
      "auxiliary_loss_clip": 0.0118493,
      "auxiliary_loss_mlp": 0.01027595,
      "balance_loss_clip": 1.05661941,
      "balance_loss_mlp": 1.01922679,
      "epoch": 0.5443395659231648,
      "flos": 15481198062720.0,
      "grad_norm": 2.0085337414061026,
      "language_loss": 0.86612844,
      "learning_rate": 1.8102168923990128e-06,
      "loss": 0.88825369,
      "num_input_tokens_seen": 97637845,
      "step": 4527,
      "time_per_iteration": 2.5896048545837402
    },
    {
      "auxiliary_loss_clip": 0.01185798,
      "auxiliary_loss_mlp": 0.01208043,
      "balance_loss_clip": 1.05466175,
      "balance_loss_mlp": 1.00086284,
      "epoch": 0.5444598088138038,
      "flos": 18770436241920.0,
      "grad_norm": 2.1558340198734034,
      "language_loss": 0.80243814,
      "learning_rate": 1.809441450185714e-06,
      "loss": 0.82637656,
      "num_input_tokens_seen": 97656330,
      "step": 4528,
      "time_per_iteration": 2.6047375202178955
    },
    {
      "auxiliary_loss_clip": 0.01189133,
      "auxiliary_loss_mlp": 0.01026222,
      "balance_loss_clip": 1.04924405,
      "balance_loss_mlp": 1.0181756,
      "epoch": 0.544580051704443,
      "flos": 21142335957120.0,
      "grad_norm": 2.534453233742027,
      "language_loss": 0.73393524,
      "learning_rate": 1.8086660368799958e-06,
      "loss": 0.75608885,
      "num_input_tokens_seen": 97674380,
      "step": 4529,
      "time_per_iteration": 2.622537136077881
    },
    {
      "auxiliary_loss_clip": 0.01188872,
      "auxiliary_loss_mlp": 0.01022944,
      "balance_loss_clip": 1.05347395,
      "balance_loss_mlp": 1.01442671,
      "epoch": 0.5447002945950821,
      "flos": 32491508054400.0,
      "grad_norm": 2.937984116359411,
      "language_loss": 0.77312279,
      "learning_rate": 1.807890652599488e-06,
      "loss": 0.79524088,
      "num_input_tokens_seen": 97698765,
      "step": 4530,
      "time_per_iteration": 2.743891954421997
    },
    {
      "auxiliary_loss_clip": 0.011793,
      "auxiliary_loss_mlp": 0.01027724,
      "balance_loss_clip": 1.05515325,
      "balance_loss_mlp": 1.02029681,
      "epoch": 0.5448205374857211,
      "flos": 11798307757440.0,
      "grad_norm": 2.216605649685207,
      "language_loss": 0.82994884,
      "learning_rate": 1.8071152974618156e-06,
      "loss": 0.85201907,
      "num_input_tokens_seen": 97716565,
      "step": 4531,
      "time_per_iteration": 3.5038883686065674
    },
    {
      "auxiliary_loss_clip": 0.01190502,
      "auxiliary_loss_mlp": 0.01208221,
      "balance_loss_clip": 1.04919124,
      "balance_loss_mlp": 1.00075912,
      "epoch": 0.5449407803763603,
      "flos": 24133766474880.0,
      "grad_norm": 2.0668201060048803,
      "language_loss": 0.78710186,
      "learning_rate": 1.806339971584599e-06,
      "loss": 0.81108916,
      "num_input_tokens_seen": 97733225,
      "step": 4532,
      "time_per_iteration": 2.6617836952209473
    },
    {
      "auxiliary_loss_clip": 0.0118135,
      "auxiliary_loss_mlp": 0.01025897,
      "balance_loss_clip": 1.05491245,
      "balance_loss_mlp": 1.01729012,
      "epoch": 0.5450610232669993,
      "flos": 23258551685760.0,
      "grad_norm": 2.0388020840713827,
      "language_loss": 0.8536756,
      "learning_rate": 1.8055646750854546e-06,
      "loss": 0.8757481,
      "num_input_tokens_seen": 97752735,
      "step": 4533,
      "time_per_iteration": 2.629239559173584
    },
    {
      "auxiliary_loss_clip": 0.0119021,
      "auxiliary_loss_mlp": 0.01024449,
      "balance_loss_clip": 1.0534507,
      "balance_loss_mlp": 1.01640213,
      "epoch": 0.5451812661576384,
      "flos": 17785083375360.0,
      "grad_norm": 2.237947814078708,
      "language_loss": 0.81517255,
      "learning_rate": 1.8047894080819945e-06,
      "loss": 0.83731914,
      "num_input_tokens_seen": 97769985,
      "step": 4534,
      "time_per_iteration": 2.6546924114227295
    },
    {
      "auxiliary_loss_clip": 0.01081545,
      "auxiliary_loss_mlp": 0.01002096,
      "balance_loss_clip": 1.02269113,
      "balance_loss_mlp": 1.00131536,
      "epoch": 0.5453015090482776,
      "flos": 71062586513280.0,
      "grad_norm": 0.7184129274400617,
      "language_loss": 0.63148332,
      "learning_rate": 1.8040141706918258e-06,
      "loss": 0.65231973,
      "num_input_tokens_seen": 97831225,
      "step": 4535,
      "time_per_iteration": 3.2873518466949463
    },
    {
      "auxiliary_loss_clip": 0.01183944,
      "auxiliary_loss_mlp": 0.01028829,
      "balance_loss_clip": 1.05131626,
      "balance_loss_mlp": 1.02042508,
      "epoch": 0.5454217519389166,
      "flos": 25552201622400.0,
      "grad_norm": 3.243441651049757,
      "language_loss": 0.76650804,
      "learning_rate": 1.8032389630325525e-06,
      "loss": 0.78863579,
      "num_input_tokens_seen": 97849975,
      "step": 4536,
      "time_per_iteration": 2.6902523040771484
    },
    {
      "auxiliary_loss_clip": 0.01183881,
      "auxiliary_loss_mlp": 0.01029775,
      "balance_loss_clip": 1.04802871,
      "balance_loss_mlp": 1.02124584,
      "epoch": 0.5455419948295557,
      "flos": 23658345037440.0,
      "grad_norm": 1.606328167214407,
      "language_loss": 0.75724912,
      "learning_rate": 1.8024637852217707e-06,
      "loss": 0.77938575,
      "num_input_tokens_seen": 97869700,
      "step": 4537,
      "time_per_iteration": 2.6673080921173096
    },
    {
      "auxiliary_loss_clip": 0.01185243,
      "auxiliary_loss_mlp": 0.01033424,
      "balance_loss_clip": 1.05189943,
      "balance_loss_mlp": 1.02465653,
      "epoch": 0.5456622377201948,
      "flos": 23403989854080.0,
      "grad_norm": 3.229253726743558,
      "language_loss": 0.84513092,
      "learning_rate": 1.8016886373770766e-06,
      "loss": 0.86731762,
      "num_input_tokens_seen": 97888215,
      "step": 4538,
      "time_per_iteration": 2.6177473068237305
    },
    {
      "auxiliary_loss_clip": 0.01186321,
      "auxiliary_loss_mlp": 0.01029804,
      "balance_loss_clip": 1.05231559,
      "balance_loss_mlp": 1.02138734,
      "epoch": 0.5457824806108339,
      "flos": 23988040997760.0,
      "grad_norm": 1.6907752843238333,
      "language_loss": 0.79155135,
      "learning_rate": 1.8009135196160579e-06,
      "loss": 0.8137126,
      "num_input_tokens_seen": 97907090,
      "step": 4539,
      "time_per_iteration": 2.714015245437622
    },
    {
      "auxiliary_loss_clip": 0.01185153,
      "auxiliary_loss_mlp": 0.01030763,
      "balance_loss_clip": 1.04896188,
      "balance_loss_mlp": 1.02368808,
      "epoch": 0.545902723501473,
      "flos": 22565870835840.0,
      "grad_norm": 1.7951679098888693,
      "language_loss": 0.84151936,
      "learning_rate": 1.8001384320563e-06,
      "loss": 0.86367857,
      "num_input_tokens_seen": 97927345,
      "step": 4540,
      "time_per_iteration": 2.7647132873535156
    },
    {
      "auxiliary_loss_clip": 0.01079609,
      "auxiliary_loss_mlp": 0.01002061,
      "balance_loss_clip": 1.02071714,
      "balance_loss_mlp": 1.00125051,
      "epoch": 0.5460229663921121,
      "flos": 55198399685760.0,
      "grad_norm": 0.7818077604480529,
      "language_loss": 0.5772779,
      "learning_rate": 1.7993633748153833e-06,
      "loss": 0.59809458,
      "num_input_tokens_seen": 97981950,
      "step": 4541,
      "time_per_iteration": 3.010251998901367
    },
    {
      "auxiliary_loss_clip": 0.01188322,
      "auxiliary_loss_mlp": 0.01031621,
      "balance_loss_clip": 1.05349827,
      "balance_loss_mlp": 1.02341342,
      "epoch": 0.5461432092827512,
      "flos": 15413866018560.0,
      "grad_norm": 2.1565792632367526,
      "language_loss": 0.73206985,
      "learning_rate": 1.7985883480108834e-06,
      "loss": 0.75426924,
      "num_input_tokens_seen": 97999585,
      "step": 4542,
      "time_per_iteration": 2.6558754444122314
    },
    {
      "auxiliary_loss_clip": 0.01179853,
      "auxiliary_loss_mlp": 0.01029641,
      "balance_loss_clip": 1.0522573,
      "balance_loss_mlp": 1.02116489,
      "epoch": 0.5462634521733902,
      "flos": 24024921287040.0,
      "grad_norm": 1.701365622030459,
      "language_loss": 0.72338158,
      "learning_rate": 1.797813351760371e-06,
      "loss": 0.74547648,
      "num_input_tokens_seen": 98021290,
      "step": 4543,
      "time_per_iteration": 2.662550926208496
    },
    {
      "auxiliary_loss_clip": 0.01183155,
      "auxiliary_loss_mlp": 0.01023826,
      "balance_loss_clip": 1.05618584,
      "balance_loss_mlp": 1.01531434,
      "epoch": 0.5463836950640293,
      "flos": 22820944291200.0,
      "grad_norm": 2.5751355652098176,
      "language_loss": 0.78145206,
      "learning_rate": 1.7970383861814116e-06,
      "loss": 0.80352187,
      "num_input_tokens_seen": 98041060,
      "step": 4544,
      "time_per_iteration": 2.657428026199341
    },
    {
      "auxiliary_loss_clip": 0.01183737,
      "auxiliary_loss_mlp": 0.01029535,
      "balance_loss_clip": 1.05454433,
      "balance_loss_mlp": 1.02052283,
      "epoch": 0.5465039379546685,
      "flos": 20448290390400.0,
      "grad_norm": 2.0369452326622133,
      "language_loss": 0.74101174,
      "learning_rate": 1.7962634513915684e-06,
      "loss": 0.76314443,
      "num_input_tokens_seen": 98058410,
      "step": 4545,
      "time_per_iteration": 2.6067121028900146
    },
    {
      "auxiliary_loss_clip": 0.01179869,
      "auxiliary_loss_mlp": 0.0102457,
      "balance_loss_clip": 1.05461764,
      "balance_loss_mlp": 1.01689577,
      "epoch": 0.5466241808453075,
      "flos": 17343310003200.0,
      "grad_norm": 1.6375498519566363,
      "language_loss": 0.79835153,
      "learning_rate": 1.7954885475083969e-06,
      "loss": 0.82039589,
      "num_input_tokens_seen": 98076080,
      "step": 4546,
      "time_per_iteration": 2.5939149856567383
    },
    {
      "auxiliary_loss_clip": 0.01184454,
      "auxiliary_loss_mlp": 0.01032392,
      "balance_loss_clip": 1.05638635,
      "balance_loss_mlp": 1.02403545,
      "epoch": 0.5467444237359466,
      "flos": 21617039122560.0,
      "grad_norm": 2.602282515426475,
      "language_loss": 0.73040587,
      "learning_rate": 1.7947136746494513e-06,
      "loss": 0.75257432,
      "num_input_tokens_seen": 98096995,
      "step": 4547,
      "time_per_iteration": 2.6445932388305664
    },
    {
      "auxiliary_loss_clip": 0.01182492,
      "auxiliary_loss_mlp": 0.01030229,
      "balance_loss_clip": 1.05299234,
      "balance_loss_mlp": 1.02207541,
      "epoch": 0.5468646666265857,
      "flos": 24170467196160.0,
      "grad_norm": 1.9052362529773363,
      "language_loss": 0.88116133,
      "learning_rate": 1.793938832932277e-06,
      "loss": 0.90328854,
      "num_input_tokens_seen": 98115105,
      "step": 4548,
      "time_per_iteration": 3.56528377532959
    },
    {
      "auxiliary_loss_clip": 0.01181367,
      "auxiliary_loss_mlp": 0.01024943,
      "balance_loss_clip": 1.05428362,
      "balance_loss_mlp": 1.01675987,
      "epoch": 0.5469849095172248,
      "flos": 27527001505920.0,
      "grad_norm": 3.9988093067324653,
      "language_loss": 0.7043221,
      "learning_rate": 1.7931640224744185e-06,
      "loss": 0.72638518,
      "num_input_tokens_seen": 98135655,
      "step": 4549,
      "time_per_iteration": 3.550666570663452
    },
    {
      "auxiliary_loss_clip": 0.01174298,
      "auxiliary_loss_mlp": 0.01025327,
      "balance_loss_clip": 1.04443574,
      "balance_loss_mlp": 1.0170536,
      "epoch": 0.5471051524078638,
      "flos": 27964680727680.0,
      "grad_norm": 2.5626100092489894,
      "language_loss": 0.73161471,
      "learning_rate": 1.7923892433934127e-06,
      "loss": 0.75361097,
      "num_input_tokens_seen": 98156730,
      "step": 4550,
      "time_per_iteration": 2.6862995624542236
    },
    {
      "auxiliary_loss_clip": 0.01190078,
      "auxiliary_loss_mlp": 0.01209026,
      "balance_loss_clip": 1.0530715,
      "balance_loss_mlp": 1.00072956,
      "epoch": 0.547225395298503,
      "flos": 18150510389760.0,
      "grad_norm": 2.1103056538299554,
      "language_loss": 0.78720045,
      "learning_rate": 1.7916144958067939e-06,
      "loss": 0.81119156,
      "num_input_tokens_seen": 98174590,
      "step": 4551,
      "time_per_iteration": 3.613301992416382
    },
    {
      "auxiliary_loss_clip": 0.01185264,
      "auxiliary_loss_mlp": 0.01024848,
      "balance_loss_clip": 1.05257416,
      "balance_loss_mlp": 1.01668775,
      "epoch": 0.5473456381891421,
      "flos": 21361498790400.0,
      "grad_norm": 1.7379700953103212,
      "language_loss": 0.78647679,
      "learning_rate": 1.7908397798320905e-06,
      "loss": 0.8085779,
      "num_input_tokens_seen": 98194325,
      "step": 4552,
      "time_per_iteration": 2.6532907485961914
    },
    {
      "auxiliary_loss_clip": 0.01182833,
      "auxiliary_loss_mlp": 0.01209061,
      "balance_loss_clip": 1.0528276,
      "balance_loss_mlp": 1.00073671,
      "epoch": 0.5474658810797811,
      "flos": 19932145908480.0,
      "grad_norm": 2.048090236057495,
      "language_loss": 0.7483592,
      "learning_rate": 1.7900650955868265e-06,
      "loss": 0.77227819,
      "num_input_tokens_seen": 98213970,
      "step": 4553,
      "time_per_iteration": 2.663530111312866
    },
    {
      "auxiliary_loss_clip": 0.01183633,
      "auxiliary_loss_mlp": 0.01208308,
      "balance_loss_clip": 1.05531704,
      "balance_loss_mlp": 1.00077081,
      "epoch": 0.5475861239704203,
      "flos": 50476217264640.0,
      "grad_norm": 1.7771384565991564,
      "language_loss": 0.76745415,
      "learning_rate": 1.7892904431885202e-06,
      "loss": 0.79137361,
      "num_input_tokens_seen": 98241145,
      "step": 4554,
      "time_per_iteration": 2.9076247215270996
    },
    {
      "auxiliary_loss_clip": 0.01188425,
      "auxiliary_loss_mlp": 0.01026557,
      "balance_loss_clip": 1.04373837,
      "balance_loss_mlp": 1.01865983,
      "epoch": 0.5477063668610593,
      "flos": 20705123612160.0,
      "grad_norm": 6.298581926546079,
      "language_loss": 0.75727373,
      "learning_rate": 1.788515822754686e-06,
      "loss": 0.77942353,
      "num_input_tokens_seen": 98261565,
      "step": 4555,
      "time_per_iteration": 2.6829097270965576
    },
    {
      "auxiliary_loss_clip": 0.01190617,
      "auxiliary_loss_mlp": 0.01031576,
      "balance_loss_clip": 1.04858685,
      "balance_loss_mlp": 1.02321362,
      "epoch": 0.5478266097516984,
      "flos": 19609740408960.0,
      "grad_norm": 2.8958668401303096,
      "language_loss": 0.78284073,
      "learning_rate": 1.7877412344028335e-06,
      "loss": 0.80506271,
      "num_input_tokens_seen": 98281370,
      "step": 4556,
      "time_per_iteration": 2.729125738143921
    },
    {
      "auxiliary_loss_clip": 0.01185108,
      "auxiliary_loss_mlp": 0.01021301,
      "balance_loss_clip": 1.0518105,
      "balance_loss_mlp": 1.013129,
      "epoch": 0.5479468526423376,
      "flos": 12896599962240.0,
      "grad_norm": 2.0657832653850785,
      "language_loss": 0.77011907,
      "learning_rate": 1.7869666782504668e-06,
      "loss": 0.79218322,
      "num_input_tokens_seen": 98297950,
      "step": 4557,
      "time_per_iteration": 2.582242250442505
    },
    {
      "auxiliary_loss_clip": 0.0117292,
      "auxiliary_loss_mlp": 0.01025874,
      "balance_loss_clip": 1.04763544,
      "balance_loss_mlp": 1.01761281,
      "epoch": 0.5480670955329766,
      "flos": 18588800142720.0,
      "grad_norm": 1.833929838914505,
      "language_loss": 0.69204515,
      "learning_rate": 1.7861921544150867e-06,
      "loss": 0.71403313,
      "num_input_tokens_seen": 98316800,
      "step": 4558,
      "time_per_iteration": 3.6400606632232666
    },
    {
      "auxiliary_loss_clip": 0.01180748,
      "auxiliary_loss_mlp": 0.0120851,
      "balance_loss_clip": 1.04599094,
      "balance_loss_mlp": 1.00062311,
      "epoch": 0.5481873384236157,
      "flos": 15954608338560.0,
      "grad_norm": 2.0988573708693714,
      "language_loss": 0.76306897,
      "learning_rate": 1.7854176630141856e-06,
      "loss": 0.78696162,
      "num_input_tokens_seen": 98333935,
      "step": 4559,
      "time_per_iteration": 2.68295955657959
    },
    {
      "auxiliary_loss_clip": 0.01185555,
      "auxiliary_loss_mlp": 0.01036004,
      "balance_loss_clip": 1.05608153,
      "balance_loss_mlp": 1.02756977,
      "epoch": 0.5483075813142548,
      "flos": 22783812606720.0,
      "grad_norm": 3.04694485022502,
      "language_loss": 0.84791267,
      "learning_rate": 1.784643204165255e-06,
      "loss": 0.87012827,
      "num_input_tokens_seen": 98353255,
      "step": 4560,
      "time_per_iteration": 2.6177685260772705
    },
    {
      "auxiliary_loss_clip": 0.01176548,
      "auxiliary_loss_mlp": 0.01025131,
      "balance_loss_clip": 1.05220985,
      "balance_loss_mlp": 1.01695883,
      "epoch": 0.5484278242048939,
      "flos": 19317212046720.0,
      "grad_norm": 1.8788370865532542,
      "language_loss": 0.76925099,
      "learning_rate": 1.7838687779857783e-06,
      "loss": 0.79126775,
      "num_input_tokens_seen": 98371130,
      "step": 4561,
      "time_per_iteration": 2.5637946128845215
    },
    {
      "auxiliary_loss_clip": 0.01178777,
      "auxiliary_loss_mlp": 0.01025124,
      "balance_loss_clip": 1.04930496,
      "balance_loss_mlp": 1.01631761,
      "epoch": 0.5485480670955329,
      "flos": 22816024128000.0,
      "grad_norm": 1.9061026604919582,
      "language_loss": 0.64097631,
      "learning_rate": 1.7830943845932366e-06,
      "loss": 0.66301525,
      "num_input_tokens_seen": 98390455,
      "step": 4562,
      "time_per_iteration": 2.6459403038024902
    },
    {
      "auxiliary_loss_clip": 0.01188752,
      "auxiliary_loss_mlp": 0.01030088,
      "balance_loss_clip": 1.05200779,
      "balance_loss_mlp": 1.02207708,
      "epoch": 0.5486683099861721,
      "flos": 22671304231680.0,
      "grad_norm": 1.756203658235334,
      "language_loss": 0.75294352,
      "learning_rate": 1.7823200241051044e-06,
      "loss": 0.77513194,
      "num_input_tokens_seen": 98409370,
      "step": 4563,
      "time_per_iteration": 2.7222354412078857
    },
    {
      "auxiliary_loss_clip": 0.01181686,
      "auxiliary_loss_mlp": 0.01024186,
      "balance_loss_clip": 1.05449271,
      "balance_loss_mlp": 1.01592505,
      "epoch": 0.5487885528768112,
      "flos": 23149383275520.0,
      "grad_norm": 1.8569579339899043,
      "language_loss": 0.80821115,
      "learning_rate": 1.7815456966388513e-06,
      "loss": 0.83026981,
      "num_input_tokens_seen": 98428465,
      "step": 4564,
      "time_per_iteration": 2.5665841102600098
    },
    {
      "auxiliary_loss_clip": 0.01187802,
      "auxiliary_loss_mlp": 0.010311,
      "balance_loss_clip": 1.04757416,
      "balance_loss_mlp": 1.02290988,
      "epoch": 0.5489087957674502,
      "flos": 22053928245120.0,
      "grad_norm": 2.4865457486630262,
      "language_loss": 0.81109333,
      "learning_rate": 1.780771402311943e-06,
      "loss": 0.83328235,
      "num_input_tokens_seen": 98447300,
      "step": 4565,
      "time_per_iteration": 2.631352186203003
    },
    {
      "auxiliary_loss_clip": 0.01186164,
      "auxiliary_loss_mlp": 0.01029157,
      "balance_loss_clip": 1.05229163,
      "balance_loss_mlp": 1.02102697,
      "epoch": 0.5490290386580894,
      "flos": 24315977191680.0,
      "grad_norm": 1.8047005151780569,
      "language_loss": 0.79118466,
      "learning_rate": 1.7799971412418374e-06,
      "loss": 0.81333792,
      "num_input_tokens_seen": 98468695,
      "step": 4566,
      "time_per_iteration": 2.6360654830932617
    },
    {
      "auxiliary_loss_clip": 0.01189718,
      "auxiliary_loss_mlp": 0.01027021,
      "balance_loss_clip": 1.04994667,
      "balance_loss_mlp": 1.01858652,
      "epoch": 0.5491492815487284,
      "flos": 18294942977280.0,
      "grad_norm": 4.420708896030683,
      "language_loss": 0.73840052,
      "learning_rate": 1.7792229135459918e-06,
      "loss": 0.7605679,
      "num_input_tokens_seen": 98485345,
      "step": 4567,
      "time_per_iteration": 2.669990301132202
    },
    {
      "auxiliary_loss_clip": 0.01100743,
      "auxiliary_loss_mlp": 0.01000506,
      "balance_loss_clip": 1.02397168,
      "balance_loss_mlp": 0.99969524,
      "epoch": 0.5492695244393675,
      "flos": 64550257050240.0,
      "grad_norm": 0.7315648282150671,
      "language_loss": 0.61580646,
      "learning_rate": 1.7784487193418538e-06,
      "loss": 0.63681895,
      "num_input_tokens_seen": 98543195,
      "step": 4568,
      "time_per_iteration": 3.128955364227295
    },
    {
      "auxiliary_loss_clip": 0.01173848,
      "auxiliary_loss_mlp": 0.01026547,
      "balance_loss_clip": 1.04512978,
      "balance_loss_mlp": 1.01775563,
      "epoch": 0.5493897673300067,
      "flos": 17379579761280.0,
      "grad_norm": 2.254478310574545,
      "language_loss": 0.61160254,
      "learning_rate": 1.7776745587468698e-06,
      "loss": 0.63360649,
      "num_input_tokens_seen": 98560620,
      "step": 4569,
      "time_per_iteration": 2.6512794494628906
    },
    {
      "auxiliary_loss_clip": 0.01180121,
      "auxiliary_loss_mlp": 0.01029236,
      "balance_loss_clip": 1.05288076,
      "balance_loss_mlp": 1.02132654,
      "epoch": 0.5495100102206457,
      "flos": 19901765980800.0,
      "grad_norm": 2.830995739495388,
      "language_loss": 0.82508177,
      "learning_rate": 1.7769004318784776e-06,
      "loss": 0.84717536,
      "num_input_tokens_seen": 98578265,
      "step": 4570,
      "time_per_iteration": 2.561492443084717
    },
    {
      "auxiliary_loss_clip": 0.01185513,
      "auxiliary_loss_mlp": 0.01022818,
      "balance_loss_clip": 1.05354476,
      "balance_loss_mlp": 1.01492608,
      "epoch": 0.5496302531112848,
      "flos": 16727190992640.0,
      "grad_norm": 1.7605992691235959,
      "language_loss": 0.80838531,
      "learning_rate": 1.776126338854113e-06,
      "loss": 0.83046865,
      "num_input_tokens_seen": 98596055,
      "step": 4571,
      "time_per_iteration": 2.628631353378296
    },
    {
      "auxiliary_loss_clip": 0.01177914,
      "auxiliary_loss_mlp": 0.01028743,
      "balance_loss_clip": 1.05298698,
      "balance_loss_mlp": 1.02097058,
      "epoch": 0.5497504960019239,
      "flos": 24572343536640.0,
      "grad_norm": 1.7020783202028609,
      "language_loss": 0.8470732,
      "learning_rate": 1.7753522797912044e-06,
      "loss": 0.86913979,
      "num_input_tokens_seen": 98616140,
      "step": 4572,
      "time_per_iteration": 2.721519708633423
    },
    {
      "auxiliary_loss_clip": 0.01192565,
      "auxiliary_loss_mlp": 0.01025358,
      "balance_loss_clip": 1.05024672,
      "balance_loss_mlp": 1.01698923,
      "epoch": 0.549870738892563,
      "flos": 15450494912640.0,
      "grad_norm": 3.152992238111176,
      "language_loss": 0.70014513,
      "learning_rate": 1.7745782548071765e-06,
      "loss": 0.72232437,
      "num_input_tokens_seen": 98633035,
      "step": 4573,
      "time_per_iteration": 2.614135980606079
    },
    {
      "auxiliary_loss_clip": 0.01185524,
      "auxiliary_loss_mlp": 0.01028075,
      "balance_loss_clip": 1.05457497,
      "balance_loss_mlp": 1.02032614,
      "epoch": 0.549990981783202,
      "flos": 21069114082560.0,
      "grad_norm": 1.6275381246885507,
      "language_loss": 0.74324226,
      "learning_rate": 1.7738042640194482e-06,
      "loss": 0.76537818,
      "num_input_tokens_seen": 98652700,
      "step": 4574,
      "time_per_iteration": 3.624460458755493
    },
    {
      "auxiliary_loss_clip": 0.0118158,
      "auxiliary_loss_mlp": 0.01027195,
      "balance_loss_clip": 1.05390358,
      "balance_loss_mlp": 1.01884413,
      "epoch": 0.5501112246738411,
      "flos": 21395901041280.0,
      "grad_norm": 1.9717545897357538,
      "language_loss": 0.70464826,
      "learning_rate": 1.7730303075454335e-06,
      "loss": 0.72673601,
      "num_input_tokens_seen": 98671590,
      "step": 4575,
      "time_per_iteration": 2.554893970489502
    },
    {
      "auxiliary_loss_clip": 0.01191987,
      "auxiliary_loss_mlp": 0.0103478,
      "balance_loss_clip": 1.04957485,
      "balance_loss_mlp": 1.02574968,
      "epoch": 0.5502314675644803,
      "flos": 17456931699840.0,
      "grad_norm": 3.0118956296860366,
      "language_loss": 0.85097128,
      "learning_rate": 1.7722563855025402e-06,
      "loss": 0.87323892,
      "num_input_tokens_seen": 98689620,
      "step": 4576,
      "time_per_iteration": 3.653860092163086
    },
    {
      "auxiliary_loss_clip": 0.01187301,
      "auxiliary_loss_mlp": 0.01026672,
      "balance_loss_clip": 1.04868388,
      "balance_loss_mlp": 1.01817298,
      "epoch": 0.5503517104551193,
      "flos": 24310410583680.0,
      "grad_norm": 2.054674799413722,
      "language_loss": 0.71067393,
      "learning_rate": 1.7714824980081721e-06,
      "loss": 0.73281372,
      "num_input_tokens_seen": 98708915,
      "step": 4577,
      "time_per_iteration": 2.6720292568206787
    },
    {
      "auxiliary_loss_clip": 0.01182487,
      "auxiliary_loss_mlp": 0.01025438,
      "balance_loss_clip": 1.05492115,
      "balance_loss_mlp": 1.01771617,
      "epoch": 0.5504719533457584,
      "flos": 22419427086720.0,
      "grad_norm": 1.6222083651819748,
      "language_loss": 0.73735893,
      "learning_rate": 1.7707086451797276e-06,
      "loss": 0.75943816,
      "num_input_tokens_seen": 98729790,
      "step": 4578,
      "time_per_iteration": 3.5651285648345947
    },
    {
      "auxiliary_loss_clip": 0.01097783,
      "auxiliary_loss_mlp": 0.01001272,
      "balance_loss_clip": 1.02096653,
      "balance_loss_mlp": 1.00039601,
      "epoch": 0.5505921962363975,
      "flos": 67294155968640.0,
      "grad_norm": 0.6985189120179832,
      "language_loss": 0.52316672,
      "learning_rate": 1.7699348271345993e-06,
      "loss": 0.54415727,
      "num_input_tokens_seen": 98792415,
      "step": 4579,
      "time_per_iteration": 3.1472725868225098
    },
    {
      "auxiliary_loss_clip": 0.01110873,
      "auxiliary_loss_mlp": 0.01001695,
      "balance_loss_clip": 1.02188575,
      "balance_loss_mlp": 1.00086021,
      "epoch": 0.5507124391270366,
      "flos": 45685125578880.0,
      "grad_norm": 0.7073108985222224,
      "language_loss": 0.54429412,
      "learning_rate": 1.7691610439901753e-06,
      "loss": 0.56541979,
      "num_input_tokens_seen": 98855350,
      "step": 4580,
      "time_per_iteration": 3.293635368347168
    },
    {
      "auxiliary_loss_clip": 0.01188265,
      "auxiliary_loss_mlp": 0.01030005,
      "balance_loss_clip": 1.05469692,
      "balance_loss_mlp": 1.02197039,
      "epoch": 0.5508326820176757,
      "flos": 22273845264000.0,
      "grad_norm": 1.7863051592057784,
      "language_loss": 0.75792152,
      "learning_rate": 1.7683872958638367e-06,
      "loss": 0.78010416,
      "num_input_tokens_seen": 98874230,
      "step": 4581,
      "time_per_iteration": 2.5744240283966064
    },
    {
      "auxiliary_loss_clip": 0.01184349,
      "auxiliary_loss_mlp": 0.01026977,
      "balance_loss_clip": 1.04975557,
      "balance_loss_mlp": 1.01868618,
      "epoch": 0.5509529249083148,
      "flos": 20012442762240.0,
      "grad_norm": 2.1687720858701875,
      "language_loss": 0.84243655,
      "learning_rate": 1.7676135828729614e-06,
      "loss": 0.86454982,
      "num_input_tokens_seen": 98893940,
      "step": 4582,
      "time_per_iteration": 2.685462474822998
    },
    {
      "auxiliary_loss_clip": 0.01184998,
      "auxiliary_loss_mlp": 0.01029173,
      "balance_loss_clip": 1.05486965,
      "balance_loss_mlp": 1.02071571,
      "epoch": 0.5510731677989539,
      "flos": 21834801325440.0,
      "grad_norm": 1.9764361685640175,
      "language_loss": 0.83264589,
      "learning_rate": 1.7668399051349205e-06,
      "loss": 0.85478759,
      "num_input_tokens_seen": 98913620,
      "step": 4583,
      "time_per_iteration": 2.609931230545044
    },
    {
      "auxiliary_loss_clip": 0.01186716,
      "auxiliary_loss_mlp": 0.01026017,
      "balance_loss_clip": 1.05027962,
      "balance_loss_mlp": 1.01767242,
      "epoch": 0.5511934106895929,
      "flos": 21467901853440.0,
      "grad_norm": 2.123865230142504,
      "language_loss": 0.83424783,
      "learning_rate": 1.766066262767081e-06,
      "loss": 0.8563751,
      "num_input_tokens_seen": 98931460,
      "step": 4584,
      "time_per_iteration": 2.660676956176758
    },
    {
      "auxiliary_loss_clip": 0.01184919,
      "auxiliary_loss_mlp": 0.01023963,
      "balance_loss_clip": 1.05376136,
      "balance_loss_mlp": 1.01577926,
      "epoch": 0.5513136535802321,
      "flos": 21068934514560.0,
      "grad_norm": 2.6054391295812045,
      "language_loss": 0.77225369,
      "learning_rate": 1.765292655886803e-06,
      "loss": 0.79434246,
      "num_input_tokens_seen": 98950105,
      "step": 4585,
      "time_per_iteration": 3.509443759918213
    },
    {
      "auxiliary_loss_clip": 0.01195575,
      "auxiliary_loss_mlp": 0.01026766,
      "balance_loss_clip": 1.05091119,
      "balance_loss_mlp": 1.01848662,
      "epoch": 0.5514338964708712,
      "flos": 27815004754560.0,
      "grad_norm": 1.9035101753863373,
      "language_loss": 0.70736831,
      "learning_rate": 1.764519084611443e-06,
      "loss": 0.72959173,
      "num_input_tokens_seen": 98970560,
      "step": 4586,
      "time_per_iteration": 2.6405656337738037
    },
    {
      "auxiliary_loss_clip": 0.01187346,
      "auxiliary_loss_mlp": 0.01027322,
      "balance_loss_clip": 1.05039668,
      "balance_loss_mlp": 1.01805317,
      "epoch": 0.5515541393615102,
      "flos": 21908525990400.0,
      "grad_norm": 1.7842747452859524,
      "language_loss": 0.78026426,
      "learning_rate": 1.7637455490583505e-06,
      "loss": 0.80241096,
      "num_input_tokens_seen": 98989885,
      "step": 4587,
      "time_per_iteration": 2.5776522159576416
    },
    {
      "auxiliary_loss_clip": 0.01185546,
      "auxiliary_loss_mlp": 0.01026792,
      "balance_loss_clip": 1.05542791,
      "balance_loss_mlp": 1.01892471,
      "epoch": 0.5516743822521494,
      "flos": 20485422074880.0,
      "grad_norm": 2.2532674392500875,
      "language_loss": 0.77729285,
      "learning_rate": 1.7629720493448701e-06,
      "loss": 0.79941618,
      "num_input_tokens_seen": 99007180,
      "step": 4588,
      "time_per_iteration": 2.6363112926483154
    },
    {
      "auxiliary_loss_clip": 0.01196035,
      "auxiliary_loss_mlp": 0.01030859,
      "balance_loss_clip": 1.05391121,
      "balance_loss_mlp": 1.022663,
      "epoch": 0.5517946251427884,
      "flos": 14940383915520.0,
      "grad_norm": 1.688433612868162,
      "language_loss": 0.85321784,
      "learning_rate": 1.7621985855883418e-06,
      "loss": 0.87548679,
      "num_input_tokens_seen": 99023880,
      "step": 4589,
      "time_per_iteration": 2.5592708587646484
    },
    {
      "auxiliary_loss_clip": 0.01179635,
      "auxiliary_loss_mlp": 0.0102696,
      "balance_loss_clip": 1.05058312,
      "balance_loss_mlp": 1.01844239,
      "epoch": 0.5519148680334275,
      "flos": 18404865573120.0,
      "grad_norm": 2.226787684626693,
      "language_loss": 0.72283727,
      "learning_rate": 1.7614251579060983e-06,
      "loss": 0.74490327,
      "num_input_tokens_seen": 99042475,
      "step": 4590,
      "time_per_iteration": 2.5730764865875244
    },
    {
      "auxiliary_loss_clip": 0.01191872,
      "auxiliary_loss_mlp": 0.01027837,
      "balance_loss_clip": 1.04944396,
      "balance_loss_mlp": 1.01897979,
      "epoch": 0.5520351109240667,
      "flos": 25113337251840.0,
      "grad_norm": 1.608448516662769,
      "language_loss": 0.84415591,
      "learning_rate": 1.76065176641547e-06,
      "loss": 0.86635303,
      "num_input_tokens_seen": 99065185,
      "step": 4591,
      "time_per_iteration": 2.6732254028320312
    },
    {
      "auxiliary_loss_clip": 0.01183081,
      "auxiliary_loss_mlp": 0.01024849,
      "balance_loss_clip": 1.05035353,
      "balance_loss_mlp": 1.01629615,
      "epoch": 0.5521553538147057,
      "flos": 21069545045760.0,
      "grad_norm": 2.0522192697813684,
      "language_loss": 0.78309798,
      "learning_rate": 1.759878411233777e-06,
      "loss": 0.80517733,
      "num_input_tokens_seen": 99083645,
      "step": 4592,
      "time_per_iteration": 2.6201627254486084
    },
    {
      "auxiliary_loss_clip": 0.0118259,
      "auxiliary_loss_mlp": 0.01027677,
      "balance_loss_clip": 1.05177462,
      "balance_loss_mlp": 1.01909399,
      "epoch": 0.5522755967053448,
      "flos": 18879999701760.0,
      "grad_norm": 2.4920194828961844,
      "language_loss": 0.76040536,
      "learning_rate": 1.7591050924783388e-06,
      "loss": 0.78250802,
      "num_input_tokens_seen": 99100835,
      "step": 4593,
      "time_per_iteration": 2.6062164306640625
    },
    {
      "auxiliary_loss_clip": 0.01103741,
      "auxiliary_loss_mlp": 0.01000705,
      "balance_loss_clip": 1.01882577,
      "balance_loss_mlp": 0.99979895,
      "epoch": 0.5523958395959839,
      "flos": 64675622494080.0,
      "grad_norm": 0.9437376566752576,
      "language_loss": 0.57955182,
      "learning_rate": 1.7583318102664661e-06,
      "loss": 0.60059631,
      "num_input_tokens_seen": 99168400,
      "step": 4594,
      "time_per_iteration": 3.2579710483551025
    },
    {
      "auxiliary_loss_clip": 0.01188021,
      "auxiliary_loss_mlp": 0.01029492,
      "balance_loss_clip": 1.05116951,
      "balance_loss_mlp": 1.02103424,
      "epoch": 0.552516082486623,
      "flos": 10889732211840.0,
      "grad_norm": 1.9053089728109203,
      "language_loss": 0.78893006,
      "learning_rate": 1.757558564715466e-06,
      "loss": 0.81110513,
      "num_input_tokens_seen": 99186475,
      "step": 4595,
      "time_per_iteration": 2.609553337097168
    },
    {
      "auxiliary_loss_clip": 0.01188176,
      "auxiliary_loss_mlp": 0.0102778,
      "balance_loss_clip": 1.05346823,
      "balance_loss_mlp": 1.01894045,
      "epoch": 0.552636325377262,
      "flos": 22199797376640.0,
      "grad_norm": 2.8522573265722238,
      "language_loss": 0.74232006,
      "learning_rate": 1.7567853559426386e-06,
      "loss": 0.76447964,
      "num_input_tokens_seen": 99203525,
      "step": 4596,
      "time_per_iteration": 2.643141508102417
    },
    {
      "auxiliary_loss_clip": 0.01187661,
      "auxiliary_loss_mlp": 0.01030666,
      "balance_loss_clip": 1.05317283,
      "balance_loss_mlp": 1.02257204,
      "epoch": 0.5527565682679012,
      "flos": 23988184652160.0,
      "grad_norm": 2.115745290221859,
      "language_loss": 0.7489323,
      "learning_rate": 1.7560121840652797e-06,
      "loss": 0.77111566,
      "num_input_tokens_seen": 99222910,
      "step": 4597,
      "time_per_iteration": 2.6660544872283936
    },
    {
      "auxiliary_loss_clip": 0.01179805,
      "auxiliary_loss_mlp": 0.01024778,
      "balance_loss_clip": 1.04919648,
      "balance_loss_mlp": 1.01635027,
      "epoch": 0.5528768111585403,
      "flos": 19719267955200.0,
      "grad_norm": 2.8405279175743505,
      "language_loss": 0.69049501,
      "learning_rate": 1.7552390492006782e-06,
      "loss": 0.71254086,
      "num_input_tokens_seen": 99241230,
      "step": 4598,
      "time_per_iteration": 2.648282051086426
    },
    {
      "auxiliary_loss_clip": 0.01197966,
      "auxiliary_loss_mlp": 0.01208599,
      "balance_loss_clip": 1.04621565,
      "balance_loss_mlp": 1.00068808,
      "epoch": 0.5529970540491793,
      "flos": 26215975002240.0,
      "grad_norm": 3.328030681791045,
      "language_loss": 0.65444636,
      "learning_rate": 1.7544659514661184e-06,
      "loss": 0.67851204,
      "num_input_tokens_seen": 99264320,
      "step": 4599,
      "time_per_iteration": 2.72969651222229
    },
    {
      "auxiliary_loss_clip": 0.01181055,
      "auxiliary_loss_mlp": 0.01022323,
      "balance_loss_clip": 1.04738796,
      "balance_loss_mlp": 1.01425231,
      "epoch": 0.5531172969398185,
      "flos": 24425971614720.0,
      "grad_norm": 2.1868497734976335,
      "language_loss": 0.8010118,
      "learning_rate": 1.7536928909788786e-06,
      "loss": 0.82304549,
      "num_input_tokens_seen": 99283625,
      "step": 4600,
      "time_per_iteration": 2.6612634658813477
    },
    {
      "auxiliary_loss_clip": 0.01109722,
      "auxiliary_loss_mlp": 0.01000303,
      "balance_loss_clip": 1.02147484,
      "balance_loss_mlp": 0.99946815,
      "epoch": 0.5532375398304575,
      "flos": 64907316195840.0,
      "grad_norm": 0.8837700065223985,
      "language_loss": 0.61906177,
      "learning_rate": 1.752919867856231e-06,
      "loss": 0.64016199,
      "num_input_tokens_seen": 99335270,
      "step": 4601,
      "time_per_iteration": 3.9792184829711914
    },
    {
      "auxiliary_loss_clip": 0.01176724,
      "auxiliary_loss_mlp": 0.01024131,
      "balance_loss_clip": 1.04704261,
      "balance_loss_mlp": 1.01609683,
      "epoch": 0.5533577827210966,
      "flos": 19683105937920.0,
      "grad_norm": 1.864491321933335,
      "language_loss": 0.79168355,
      "learning_rate": 1.7521468822154436e-06,
      "loss": 0.81369215,
      "num_input_tokens_seen": 99354185,
      "step": 4602,
      "time_per_iteration": 2.5914502143859863
    },
    {
      "auxiliary_loss_clip": 0.01178006,
      "auxiliary_loss_mlp": 0.01026628,
      "balance_loss_clip": 1.05040228,
      "balance_loss_mlp": 1.01914144,
      "epoch": 0.5534780256117358,
      "flos": 32306496076800.0,
      "grad_norm": 1.7759614266037909,
      "language_loss": 0.75438064,
      "learning_rate": 1.751373934173777e-06,
      "loss": 0.77642697,
      "num_input_tokens_seen": 99376930,
      "step": 4603,
      "time_per_iteration": 3.5981602668762207
    },
    {
      "auxiliary_loss_clip": 0.01183318,
      "auxiliary_loss_mlp": 0.01031376,
      "balance_loss_clip": 1.05409181,
      "balance_loss_mlp": 1.02290618,
      "epoch": 0.5535982685023748,
      "flos": 23222425582080.0,
      "grad_norm": 1.937486937999839,
      "language_loss": 0.73391759,
      "learning_rate": 1.750601023848487e-06,
      "loss": 0.75606453,
      "num_input_tokens_seen": 99397655,
      "step": 4604,
      "time_per_iteration": 2.7718143463134766
    },
    {
      "auxiliary_loss_clip": 0.01182605,
      "auxiliary_loss_mlp": 0.01208405,
      "balance_loss_clip": 1.05734897,
      "balance_loss_mlp": 1.00076914,
      "epoch": 0.5537185113930139,
      "flos": 24352534258560.0,
      "grad_norm": 2.5884395683258568,
      "language_loss": 0.73820138,
      "learning_rate": 1.749828151356823e-06,
      "loss": 0.76211149,
      "num_input_tokens_seen": 99417850,
      "step": 4605,
      "time_per_iteration": 3.589402198791504
    },
    {
      "auxiliary_loss_clip": 0.01186741,
      "auxiliary_loss_mlp": 0.01033714,
      "balance_loss_clip": 1.05144954,
      "balance_loss_mlp": 1.02579284,
      "epoch": 0.553838754283653,
      "flos": 23549068886400.0,
      "grad_norm": 1.7805030943579758,
      "language_loss": 0.75650299,
      "learning_rate": 1.7490553168160297e-06,
      "loss": 0.7787075,
      "num_input_tokens_seen": 99438920,
      "step": 4606,
      "time_per_iteration": 2.687401056289673
    },
    {
      "auxiliary_loss_clip": 0.01184385,
      "auxiliary_loss_mlp": 0.01026324,
      "balance_loss_clip": 1.05023694,
      "balance_loss_mlp": 1.01749611,
      "epoch": 0.5539589971742921,
      "flos": 17275044205440.0,
      "grad_norm": 2.144783852844175,
      "language_loss": 0.76517719,
      "learning_rate": 1.748282520343345e-06,
      "loss": 0.78728431,
      "num_input_tokens_seen": 99457950,
      "step": 4607,
      "time_per_iteration": 2.6454830169677734
    },
    {
      "auxiliary_loss_clip": 0.01194586,
      "auxiliary_loss_mlp": 0.0102565,
      "balance_loss_clip": 1.05548823,
      "balance_loss_mlp": 1.01688778,
      "epoch": 0.5540792400649311,
      "flos": 27564169104000.0,
      "grad_norm": 1.7988970123072205,
      "language_loss": 0.78836703,
      "learning_rate": 1.7475097620560023e-06,
      "loss": 0.81056941,
      "num_input_tokens_seen": 99478015,
      "step": 4608,
      "time_per_iteration": 2.687105417251587
    },
    {
      "auxiliary_loss_clip": 0.01181502,
      "auxiliary_loss_mlp": 0.01024659,
      "balance_loss_clip": 1.05514753,
      "balance_loss_mlp": 1.01676714,
      "epoch": 0.5541994829555702,
      "flos": 23878657105920.0,
      "grad_norm": 1.7730512975396375,
      "language_loss": 0.71173805,
      "learning_rate": 1.746737042071228e-06,
      "loss": 0.7337997,
      "num_input_tokens_seen": 99496520,
      "step": 4609,
      "time_per_iteration": 2.609800338745117
    },
    {
      "auxiliary_loss_clip": 0.01182882,
      "auxiliary_loss_mlp": 0.01028164,
      "balance_loss_clip": 1.05100131,
      "balance_loss_mlp": 1.01964653,
      "epoch": 0.5543197258462094,
      "flos": 20115721342080.0,
      "grad_norm": 1.7891547394832887,
      "language_loss": 0.79171288,
      "learning_rate": 1.7459643605062424e-06,
      "loss": 0.81382328,
      "num_input_tokens_seen": 99513780,
      "step": 4610,
      "time_per_iteration": 2.5905861854553223
    },
    {
      "auxiliary_loss_clip": 0.01186767,
      "auxiliary_loss_mlp": 0.01027543,
      "balance_loss_clip": 1.04769063,
      "balance_loss_mlp": 1.0189718,
      "epoch": 0.5544399687368484,
      "flos": 20916565021440.0,
      "grad_norm": 1.932414857551738,
      "language_loss": 0.81200141,
      "learning_rate": 1.745191717478262e-06,
      "loss": 0.83414447,
      "num_input_tokens_seen": 99532360,
      "step": 4611,
      "time_per_iteration": 3.5954155921936035
    },
    {
      "auxiliary_loss_clip": 0.01177922,
      "auxiliary_loss_mlp": 0.01026178,
      "balance_loss_clip": 1.04915619,
      "balance_loss_mlp": 1.01786304,
      "epoch": 0.5545602116274875,
      "flos": 25518661297920.0,
      "grad_norm": 2.620219912586263,
      "language_loss": 0.79843593,
      "learning_rate": 1.7444191131044948e-06,
      "loss": 0.82047695,
      "num_input_tokens_seen": 99552635,
      "step": 4612,
      "time_per_iteration": 2.692126750946045
    },
    {
      "auxiliary_loss_clip": 0.011894,
      "auxiliary_loss_mlp": 0.0103022,
      "balance_loss_clip": 1.05456483,
      "balance_loss_mlp": 1.02104723,
      "epoch": 0.5546804545181266,
      "flos": 20995568985600.0,
      "grad_norm": 1.7675561462346703,
      "language_loss": 0.73239589,
      "learning_rate": 1.7436465475021456e-06,
      "loss": 0.75459206,
      "num_input_tokens_seen": 99572685,
      "step": 4613,
      "time_per_iteration": 2.6624319553375244
    },
    {
      "auxiliary_loss_clip": 0.01180003,
      "auxiliary_loss_mlp": 0.01025219,
      "balance_loss_clip": 1.04978037,
      "balance_loss_mlp": 1.01656401,
      "epoch": 0.5548006974087657,
      "flos": 26833638297600.0,
      "grad_norm": 2.1434743491593387,
      "language_loss": 0.71458101,
      "learning_rate": 1.7428740207884111e-06,
      "loss": 0.73663324,
      "num_input_tokens_seen": 99593565,
      "step": 4614,
      "time_per_iteration": 2.671232223510742
    },
    {
      "auxiliary_loss_clip": 0.0119282,
      "auxiliary_loss_mlp": 0.01028139,
      "balance_loss_clip": 1.04967916,
      "balance_loss_mlp": 1.01960397,
      "epoch": 0.5549209402994048,
      "flos": 33656414031360.0,
      "grad_norm": 1.7522054607384034,
      "language_loss": 0.60918963,
      "learning_rate": 1.7421015330804833e-06,
      "loss": 0.63139921,
      "num_input_tokens_seen": 99613485,
      "step": 4615,
      "time_per_iteration": 2.6892879009246826
    },
    {
      "auxiliary_loss_clip": 0.01182326,
      "auxiliary_loss_mlp": 0.01029094,
      "balance_loss_clip": 1.05514979,
      "balance_loss_mlp": 1.02074897,
      "epoch": 0.5550411831900439,
      "flos": 23769524609280.0,
      "grad_norm": 1.8439013204295112,
      "language_loss": 0.72318059,
      "learning_rate": 1.7413290844955475e-06,
      "loss": 0.74529481,
      "num_input_tokens_seen": 99633515,
      "step": 4616,
      "time_per_iteration": 2.5967960357666016
    },
    {
      "auxiliary_loss_clip": 0.01178984,
      "auxiliary_loss_mlp": 0.0103116,
      "balance_loss_clip": 1.05547285,
      "balance_loss_mlp": 1.02314305,
      "epoch": 0.555161426080683,
      "flos": 21651189978240.0,
      "grad_norm": 1.8449605284979078,
      "language_loss": 0.78443074,
      "learning_rate": 1.7405566751507843e-06,
      "loss": 0.80653214,
      "num_input_tokens_seen": 99651560,
      "step": 4617,
      "time_per_iteration": 2.619091510772705
    },
    {
      "auxiliary_loss_clip": 0.01188532,
      "auxiliary_loss_mlp": 0.01032229,
      "balance_loss_clip": 1.04958832,
      "balance_loss_mlp": 1.02417064,
      "epoch": 0.555281668971322,
      "flos": 49563116605440.0,
      "grad_norm": 3.178995722492089,
      "language_loss": 0.67737645,
      "learning_rate": 1.7397843051633668e-06,
      "loss": 0.69958407,
      "num_input_tokens_seen": 99674255,
      "step": 4618,
      "time_per_iteration": 2.8705480098724365
    },
    {
      "auxiliary_loss_clip": 0.01178659,
      "auxiliary_loss_mlp": 0.01028395,
      "balance_loss_clip": 1.05268645,
      "balance_loss_mlp": 1.01976371,
      "epoch": 0.5554019118619612,
      "flos": 20741608851840.0,
      "grad_norm": 1.8698216520286548,
      "language_loss": 0.71724987,
      "learning_rate": 1.739011974650464e-06,
      "loss": 0.7393204,
      "num_input_tokens_seen": 99693585,
      "step": 4619,
      "time_per_iteration": 2.5175318717956543
    },
    {
      "auxiliary_loss_clip": 0.01195948,
      "auxiliary_loss_mlp": 0.01027245,
      "balance_loss_clip": 1.04955125,
      "balance_loss_mlp": 1.0184474,
      "epoch": 0.5555221547526003,
      "flos": 25483217552640.0,
      "grad_norm": 2.0475931031012897,
      "language_loss": 0.77037799,
      "learning_rate": 1.7382396837292365e-06,
      "loss": 0.79260993,
      "num_input_tokens_seen": 99714045,
      "step": 4620,
      "time_per_iteration": 2.6442158222198486
    },
    {
      "auxiliary_loss_clip": 0.01182826,
      "auxiliary_loss_mlp": 0.01024778,
      "balance_loss_clip": 1.05669951,
      "balance_loss_mlp": 1.01599848,
      "epoch": 0.5556423976432393,
      "flos": 21762513204480.0,
      "grad_norm": 1.6155848659860317,
      "language_loss": 0.73538011,
      "learning_rate": 1.737467432516841e-06,
      "loss": 0.75745618,
      "num_input_tokens_seen": 99734145,
      "step": 4621,
      "time_per_iteration": 2.5968010425567627
    },
    {
      "auxiliary_loss_clip": 0.01184636,
      "auxiliary_loss_mlp": 0.01026597,
      "balance_loss_clip": 1.04819512,
      "balance_loss_mlp": 1.01818061,
      "epoch": 0.5557626405338785,
      "flos": 24900171989760.0,
      "grad_norm": 2.1571485955247347,
      "language_loss": 0.74229777,
      "learning_rate": 1.7366952211304274e-06,
      "loss": 0.76441008,
      "num_input_tokens_seen": 99751990,
      "step": 4622,
      "time_per_iteration": 2.669602155685425
    },
    {
      "auxiliary_loss_clip": 0.01178881,
      "auxiliary_loss_mlp": 0.01027516,
      "balance_loss_clip": 1.04972625,
      "balance_loss_mlp": 1.01917148,
      "epoch": 0.5558828834245175,
      "flos": 18697501676160.0,
      "grad_norm": 2.1134305905856507,
      "language_loss": 0.83423674,
      "learning_rate": 1.735923049687139e-06,
      "loss": 0.85630065,
      "num_input_tokens_seen": 99768565,
      "step": 4623,
      "time_per_iteration": 2.615893602371216
    },
    {
      "auxiliary_loss_clip": 0.01180082,
      "auxiliary_loss_mlp": 0.01025508,
      "balance_loss_clip": 1.04953313,
      "balance_loss_mlp": 1.01747894,
      "epoch": 0.5560031263151566,
      "flos": 27272179445760.0,
      "grad_norm": 1.504014499246557,
      "language_loss": 0.73795557,
      "learning_rate": 1.7351509183041144e-06,
      "loss": 0.76001143,
      "num_input_tokens_seen": 99788895,
      "step": 4624,
      "time_per_iteration": 2.7069225311279297
    },
    {
      "auxiliary_loss_clip": 0.01183858,
      "auxiliary_loss_mlp": 0.01026838,
      "balance_loss_clip": 1.05510592,
      "balance_loss_mlp": 1.01871419,
      "epoch": 0.5561233692057957,
      "flos": 23403738458880.0,
      "grad_norm": 1.8567247670854163,
      "language_loss": 0.71863008,
      "learning_rate": 1.7343788270984852e-06,
      "loss": 0.74073702,
      "num_input_tokens_seen": 99808035,
      "step": 4625,
      "time_per_iteration": 2.6337318420410156
    },
    {
      "auxiliary_loss_clip": 0.01184186,
      "auxiliary_loss_mlp": 0.01025367,
      "balance_loss_clip": 1.05345821,
      "balance_loss_mlp": 1.0168196,
      "epoch": 0.5562436120964348,
      "flos": 37670867804160.0,
      "grad_norm": 3.6754477867756497,
      "language_loss": 0.74936771,
      "learning_rate": 1.7336067761873764e-06,
      "loss": 0.77146327,
      "num_input_tokens_seen": 99830460,
      "step": 4626,
      "time_per_iteration": 2.8649260997772217
    },
    {
      "auxiliary_loss_clip": 0.0119184,
      "auxiliary_loss_mlp": 0.0103126,
      "balance_loss_clip": 1.05356956,
      "balance_loss_mlp": 1.02243888,
      "epoch": 0.5563638549870739,
      "flos": 25155245445120.0,
      "grad_norm": 1.8748155876090917,
      "language_loss": 0.76342183,
      "learning_rate": 1.7328347656879076e-06,
      "loss": 0.78565288,
      "num_input_tokens_seen": 99850320,
      "step": 4627,
      "time_per_iteration": 3.5841543674468994
    },
    {
      "auxiliary_loss_clip": 0.01188329,
      "auxiliary_loss_mlp": 0.01025769,
      "balance_loss_clip": 1.04903603,
      "balance_loss_mlp": 1.01697755,
      "epoch": 0.556484097877713,
      "flos": 13581810783360.0,
      "grad_norm": 2.5920745515425723,
      "language_loss": 0.6821723,
      "learning_rate": 1.7320627957171927e-06,
      "loss": 0.70431328,
      "num_input_tokens_seen": 99864980,
      "step": 4628,
      "time_per_iteration": 2.7506942749023438
    },
    {
      "auxiliary_loss_clip": 0.01183962,
      "auxiliary_loss_mlp": 0.01024667,
      "balance_loss_clip": 1.05595744,
      "balance_loss_mlp": 1.01669168,
      "epoch": 0.5566043407683521,
      "flos": 24681368292480.0,
      "grad_norm": 1.828642780057197,
      "language_loss": 0.81489795,
      "learning_rate": 1.7312908663923382e-06,
      "loss": 0.83698422,
      "num_input_tokens_seen": 99881155,
      "step": 4629,
      "time_per_iteration": 2.655766725540161
    },
    {
      "auxiliary_loss_clip": 0.01177561,
      "auxiliary_loss_mlp": 0.01025089,
      "balance_loss_clip": 1.05015659,
      "balance_loss_mlp": 1.01626754,
      "epoch": 0.5567245836589911,
      "flos": 20588161950720.0,
      "grad_norm": 2.6146250594092355,
      "language_loss": 0.67184812,
      "learning_rate": 1.7305189778304463e-06,
      "loss": 0.69387466,
      "num_input_tokens_seen": 99899330,
      "step": 4630,
      "time_per_iteration": 3.4812681674957275
    },
    {
      "auxiliary_loss_clip": 0.01185306,
      "auxiliary_loss_mlp": 0.01031219,
      "balance_loss_clip": 1.0532757,
      "balance_loss_mlp": 1.02277291,
      "epoch": 0.5568448265496303,
      "flos": 20704189858560.0,
      "grad_norm": 2.0987596864838705,
      "language_loss": 0.80245471,
      "learning_rate": 1.729747130148611e-06,
      "loss": 0.82462001,
      "num_input_tokens_seen": 99918525,
      "step": 4631,
      "time_per_iteration": 3.592193603515625
    },
    {
      "auxiliary_loss_clip": 0.01194778,
      "auxiliary_loss_mlp": 0.01029437,
      "balance_loss_clip": 1.05064583,
      "balance_loss_mlp": 1.02026403,
      "epoch": 0.5569650694402694,
      "flos": 25302910256640.0,
      "grad_norm": 2.1011290310961903,
      "language_loss": 0.76847571,
      "learning_rate": 1.7289753234639208e-06,
      "loss": 0.79071784,
      "num_input_tokens_seen": 99937500,
      "step": 4632,
      "time_per_iteration": 2.6942708492279053
    },
    {
      "auxiliary_loss_clip": 0.01189508,
      "auxiliary_loss_mlp": 0.01029748,
      "balance_loss_clip": 1.05518031,
      "balance_loss_mlp": 1.02097404,
      "epoch": 0.5570853123309084,
      "flos": 19712623939200.0,
      "grad_norm": 3.211149177714875,
      "language_loss": 0.76678199,
      "learning_rate": 1.7282035578934592e-06,
      "loss": 0.78897452,
      "num_input_tokens_seen": 99955665,
      "step": 4633,
      "time_per_iteration": 2.5921695232391357
    },
    {
      "auxiliary_loss_clip": 0.01176874,
      "auxiliary_loss_mlp": 0.01034803,
      "balance_loss_clip": 1.05231631,
      "balance_loss_mlp": 1.02682829,
      "epoch": 0.5572055552215476,
      "flos": 16108091153280.0,
      "grad_norm": 1.7465605349326023,
      "language_loss": 0.79031694,
      "learning_rate": 1.727431833554301e-06,
      "loss": 0.81243372,
      "num_input_tokens_seen": 99974140,
      "step": 4634,
      "time_per_iteration": 2.6276912689208984
    },
    {
      "auxiliary_loss_clip": 0.0120161,
      "auxiliary_loss_mlp": 0.01027215,
      "balance_loss_clip": 1.04828,
      "balance_loss_mlp": 1.01886761,
      "epoch": 0.5573257981121866,
      "flos": 17128815937920.0,
      "grad_norm": 2.135990181169772,
      "language_loss": 0.77347291,
      "learning_rate": 1.7266601505635175e-06,
      "loss": 0.79576111,
      "num_input_tokens_seen": 99991480,
      "step": 4635,
      "time_per_iteration": 2.667940616607666
    },
    {
      "auxiliary_loss_clip": 0.01186228,
      "auxiliary_loss_mlp": 0.01029414,
      "balance_loss_clip": 1.05558312,
      "balance_loss_mlp": 1.02083743,
      "epoch": 0.5574460410028257,
      "flos": 18807029222400.0,
      "grad_norm": 1.916513097483379,
      "language_loss": 0.76010835,
      "learning_rate": 1.7258885090381717e-06,
      "loss": 0.78226477,
      "num_input_tokens_seen": 100009520,
      "step": 4636,
      "time_per_iteration": 2.602064609527588
    },
    {
      "auxiliary_loss_clip": 0.01188367,
      "auxiliary_loss_mlp": 0.01029408,
      "balance_loss_clip": 1.0513649,
      "balance_loss_mlp": 1.02129006,
      "epoch": 0.5575662838934649,
      "flos": 29642678530560.0,
      "grad_norm": 1.7398370940301582,
      "language_loss": 0.78797841,
      "learning_rate": 1.7251169090953213e-06,
      "loss": 0.81015617,
      "num_input_tokens_seen": 100029995,
      "step": 4637,
      "time_per_iteration": 2.6800498962402344
    },
    {
      "auxiliary_loss_clip": 0.0118374,
      "auxiliary_loss_mlp": 0.01024318,
      "balance_loss_clip": 1.05195546,
      "balance_loss_mlp": 1.01550245,
      "epoch": 0.5576865267841039,
      "flos": 22054466949120.0,
      "grad_norm": 3.1213613891872227,
      "language_loss": 0.76274174,
      "learning_rate": 1.7243453508520168e-06,
      "loss": 0.78482234,
      "num_input_tokens_seen": 100046980,
      "step": 4638,
      "time_per_iteration": 3.5763654708862305
    },
    {
      "auxiliary_loss_clip": 0.01187324,
      "auxiliary_loss_mlp": 0.01028872,
      "balance_loss_clip": 1.05148029,
      "balance_loss_mlp": 1.02034879,
      "epoch": 0.557806769674743,
      "flos": 17196040241280.0,
      "grad_norm": 1.904640189953558,
      "language_loss": 0.84406638,
      "learning_rate": 1.7235738344253038e-06,
      "loss": 0.8662284,
      "num_input_tokens_seen": 100060610,
      "step": 4639,
      "time_per_iteration": 2.660076379776001
    },
    {
      "auxiliary_loss_clip": 0.0118245,
      "auxiliary_loss_mlp": 0.01028768,
      "balance_loss_clip": 1.05505824,
      "balance_loss_mlp": 1.02000022,
      "epoch": 0.557927012565382,
      "flos": 24712717887360.0,
      "grad_norm": 2.652058446403555,
      "language_loss": 0.82715744,
      "learning_rate": 1.72280235993222e-06,
      "loss": 0.84926963,
      "num_input_tokens_seen": 100078915,
      "step": 4640,
      "time_per_iteration": 2.663775682449341
    },
    {
      "auxiliary_loss_clip": 0.01181376,
      "auxiliary_loss_mlp": 0.01209149,
      "balance_loss_clip": 1.05321145,
      "balance_loss_mlp": 1.00085711,
      "epoch": 0.5580472554560212,
      "flos": 16983090460800.0,
      "grad_norm": 2.502752832458773,
      "language_loss": 0.69812953,
      "learning_rate": 1.722030927489798e-06,
      "loss": 0.72203481,
      "num_input_tokens_seen": 100096195,
      "step": 4641,
      "time_per_iteration": 2.6260437965393066
    },
    {
      "auxiliary_loss_clip": 0.01191299,
      "auxiliary_loss_mlp": 0.01026433,
      "balance_loss_clip": 1.05193651,
      "balance_loss_mlp": 1.01769459,
      "epoch": 0.5581674983466602,
      "flos": 23509100027520.0,
      "grad_norm": 1.708884051992637,
      "language_loss": 0.74573231,
      "learning_rate": 1.7212595372150634e-06,
      "loss": 0.76790959,
      "num_input_tokens_seen": 100116175,
      "step": 4642,
      "time_per_iteration": 2.679630756378174
    },
    {
      "auxiliary_loss_clip": 0.01182978,
      "auxiliary_loss_mlp": 0.01025467,
      "balance_loss_clip": 1.05575442,
      "balance_loss_mlp": 1.01748013,
      "epoch": 0.5582877412372993,
      "flos": 13480291969920.0,
      "grad_norm": 2.182085546064244,
      "language_loss": 0.72905552,
      "learning_rate": 1.720488189225035e-06,
      "loss": 0.75114,
      "num_input_tokens_seen": 100133875,
      "step": 4643,
      "time_per_iteration": 2.611619472503662
    },
    {
      "auxiliary_loss_clip": 0.01186811,
      "auxiliary_loss_mlp": 0.01025876,
      "balance_loss_clip": 1.05359769,
      "balance_loss_mlp": 1.01773381,
      "epoch": 0.5584079841279385,
      "flos": 21903605827200.0,
      "grad_norm": 2.7450455074376925,
      "language_loss": 0.79042017,
      "learning_rate": 1.7197168836367265e-06,
      "loss": 0.81254709,
      "num_input_tokens_seen": 100150685,
      "step": 4644,
      "time_per_iteration": 2.576539993286133
    },
    {
      "auxiliary_loss_clip": 0.01180292,
      "auxiliary_loss_mlp": 0.0120851,
      "balance_loss_clip": 1.05149269,
      "balance_loss_mlp": 1.00084209,
      "epoch": 0.5585282270185775,
      "flos": 18843550375680.0,
      "grad_norm": 1.8457887745086616,
      "language_loss": 0.81892383,
      "learning_rate": 1.7189456205671433e-06,
      "loss": 0.84281182,
      "num_input_tokens_seen": 100169530,
      "step": 4645,
      "time_per_iteration": 2.6082839965820312
    },
    {
      "auxiliary_loss_clip": 0.01191989,
      "auxiliary_loss_mlp": 0.01026551,
      "balance_loss_clip": 1.0544138,
      "balance_loss_mlp": 1.0177356,
      "epoch": 0.5586484699092166,
      "flos": 21868449390720.0,
      "grad_norm": 1.8802824743786697,
      "language_loss": 0.82237881,
      "learning_rate": 1.7181744001332866e-06,
      "loss": 0.8445642,
      "num_input_tokens_seen": 100188140,
      "step": 4646,
      "time_per_iteration": 2.766840934753418
    },
    {
      "auxiliary_loss_clip": 0.01182317,
      "auxiliary_loss_mlp": 0.01026956,
      "balance_loss_clip": 1.05702198,
      "balance_loss_mlp": 1.01854539,
      "epoch": 0.5587687127998557,
      "flos": 22893232412160.0,
      "grad_norm": 1.9188976998843585,
      "language_loss": 0.63408339,
      "learning_rate": 1.7174032224521493e-06,
      "loss": 0.65617609,
      "num_input_tokens_seen": 100206850,
      "step": 4647,
      "time_per_iteration": 2.6319611072540283
    },
    {
      "auxiliary_loss_clip": 0.01184933,
      "auxiliary_loss_mlp": 0.01030786,
      "balance_loss_clip": 1.05390072,
      "balance_loss_mlp": 1.02258182,
      "epoch": 0.5588889556904948,
      "flos": 20303067703680.0,
      "grad_norm": 1.6651159590632374,
      "language_loss": 0.69708055,
      "learning_rate": 1.7166320876407184e-06,
      "loss": 0.71923774,
      "num_input_tokens_seen": 100226270,
      "step": 4648,
      "time_per_iteration": 2.5792195796966553
    },
    {
      "auxiliary_loss_clip": 0.01185562,
      "auxiliary_loss_mlp": 0.01208994,
      "balance_loss_clip": 1.05696964,
      "balance_loss_mlp": 1.00102997,
      "epoch": 0.5590091985811338,
      "flos": 16472153450880.0,
      "grad_norm": 2.1030085838443893,
      "language_loss": 0.67780256,
      "learning_rate": 1.7158609958159742e-06,
      "loss": 0.70174813,
      "num_input_tokens_seen": 100243675,
      "step": 4649,
      "time_per_iteration": 2.589214563369751
    },
    {
      "auxiliary_loss_clip": 0.01201869,
      "auxiliary_loss_mlp": 0.01027,
      "balance_loss_clip": 1.04710603,
      "balance_loss_mlp": 1.01883435,
      "epoch": 0.559129441471773,
      "flos": 14532186781440.0,
      "grad_norm": 2.016158329828184,
      "language_loss": 0.7817508,
      "learning_rate": 1.7150899470948911e-06,
      "loss": 0.80403954,
      "num_input_tokens_seen": 100258940,
      "step": 4650,
      "time_per_iteration": 2.719865322113037
    },
    {
      "auxiliary_loss_clip": 0.01090182,
      "auxiliary_loss_mlp": 0.01003511,
      "balance_loss_clip": 1.01975203,
      "balance_loss_mlp": 1.002653,
      "epoch": 0.5592496843624121,
      "flos": 60521009852160.0,
      "grad_norm": 1.0179840601817285,
      "language_loss": 0.56643516,
      "learning_rate": 1.7143189415944365e-06,
      "loss": 0.58737206,
      "num_input_tokens_seen": 100323400,
      "step": 4651,
      "time_per_iteration": 3.279416799545288
    },
    {
      "auxiliary_loss_clip": 0.01184426,
      "auxiliary_loss_mlp": 0.01027065,
      "balance_loss_clip": 1.05472994,
      "balance_loss_mlp": 1.01823163,
      "epoch": 0.5593699272530511,
      "flos": 20886256920960.0,
      "grad_norm": 4.950985371621367,
      "language_loss": 0.76335251,
      "learning_rate": 1.7135479794315714e-06,
      "loss": 0.78546745,
      "num_input_tokens_seen": 100340355,
      "step": 4652,
      "time_per_iteration": 2.694735288619995
    },
    {
      "auxiliary_loss_clip": 0.01187134,
      "auxiliary_loss_mlp": 0.01024776,
      "balance_loss_clip": 1.05070829,
      "balance_loss_mlp": 1.01691413,
      "epoch": 0.5594901701436903,
      "flos": 12896743616640.0,
      "grad_norm": 1.8669081108123682,
      "language_loss": 0.7930724,
      "learning_rate": 1.7127770607232502e-06,
      "loss": 0.81519151,
      "num_input_tokens_seen": 100358900,
      "step": 4653,
      "time_per_iteration": 2.7576141357421875
    },
    {
      "auxiliary_loss_clip": 0.01195479,
      "auxiliary_loss_mlp": 0.01025878,
      "balance_loss_clip": 1.05115902,
      "balance_loss_mlp": 1.01744962,
      "epoch": 0.5596104130343293,
      "flos": 23112107936640.0,
      "grad_norm": 2.0778303453170617,
      "language_loss": 0.80189186,
      "learning_rate": 1.7120061855864204e-06,
      "loss": 0.82410544,
      "num_input_tokens_seen": 100378910,
      "step": 4654,
      "time_per_iteration": 3.6024868488311768
    },
    {
      "auxiliary_loss_clip": 0.01185235,
      "auxiliary_loss_mlp": 0.01029351,
      "balance_loss_clip": 1.05525291,
      "balance_loss_mlp": 1.02081513,
      "epoch": 0.5597306559249684,
      "flos": 25957812977280.0,
      "grad_norm": 2.1613043315425053,
      "language_loss": 0.71180165,
      "learning_rate": 1.7112353541380233e-06,
      "loss": 0.73394746,
      "num_input_tokens_seen": 100398770,
      "step": 4655,
      "time_per_iteration": 2.694761276245117
    },
    {
      "auxiliary_loss_clip": 0.01190207,
      "auxiliary_loss_mlp": 0.0103116,
      "balance_loss_clip": 1.05366302,
      "balance_loss_mlp": 1.02156401,
      "epoch": 0.5598508988156076,
      "flos": 22492289825280.0,
      "grad_norm": 1.5085483798265253,
      "language_loss": 0.722534,
      "learning_rate": 1.7104645664949931e-06,
      "loss": 0.74474764,
      "num_input_tokens_seen": 100421240,
      "step": 4656,
      "time_per_iteration": 3.5753047466278076
    },
    {
      "auxiliary_loss_clip": 0.01188434,
      "auxiliary_loss_mlp": 0.01029636,
      "balance_loss_clip": 1.0494684,
      "balance_loss_mlp": 1.02057624,
      "epoch": 0.5599711417062466,
      "flos": 23112538899840.0,
      "grad_norm": 2.850225898854265,
      "language_loss": 0.71289945,
      "learning_rate": 1.7096938227742584e-06,
      "loss": 0.73508012,
      "num_input_tokens_seen": 100442370,
      "step": 4657,
      "time_per_iteration": 2.6665408611297607
    },
    {
      "auxiliary_loss_clip": 0.01182679,
      "auxiliary_loss_mlp": 0.01027097,
      "balance_loss_clip": 1.05521369,
      "balance_loss_mlp": 1.01857352,
      "epoch": 0.5600913845968857,
      "flos": 22339345714560.0,
      "grad_norm": 2.1965927618751486,
      "language_loss": 0.84617949,
      "learning_rate": 1.70892312309274e-06,
      "loss": 0.86827719,
      "num_input_tokens_seen": 100460260,
      "step": 4658,
      "time_per_iteration": 3.5105178356170654
    },
    {
      "auxiliary_loss_clip": 0.01187356,
      "auxiliary_loss_mlp": 0.01026516,
      "balance_loss_clip": 1.04595578,
      "balance_loss_mlp": 1.01784909,
      "epoch": 0.5602116274875248,
      "flos": 17633791290240.0,
      "grad_norm": 2.3753468443216597,
      "language_loss": 0.68335092,
      "learning_rate": 1.7081524675673523e-06,
      "loss": 0.70548964,
      "num_input_tokens_seen": 100475750,
      "step": 4659,
      "time_per_iteration": 2.6951146125793457
    },
    {
      "auxiliary_loss_clip": 0.01091608,
      "auxiliary_loss_mlp": 0.01003738,
      "balance_loss_clip": 1.01900959,
      "balance_loss_mlp": 1.00289178,
      "epoch": 0.5603318703781639,
      "flos": 70115945529600.0,
      "grad_norm": 0.7786646060826212,
      "language_loss": 0.59615302,
      "learning_rate": 1.7073818563150026e-06,
      "loss": 0.61710644,
      "num_input_tokens_seen": 100537830,
      "step": 4660,
      "time_per_iteration": 3.296513795852661
    },
    {
      "auxiliary_loss_clip": 0.01180161,
      "auxiliary_loss_mlp": 0.01026438,
      "balance_loss_clip": 1.05140626,
      "balance_loss_mlp": 1.01799786,
      "epoch": 0.560452113268803,
      "flos": 18545850455040.0,
      "grad_norm": 3.234275453860782,
      "language_loss": 0.86986375,
      "learning_rate": 1.7066112894525935e-06,
      "loss": 0.89192969,
      "num_input_tokens_seen": 100555910,
      "step": 4661,
      "time_per_iteration": 2.5926122665405273
    },
    {
      "auxiliary_loss_clip": 0.01179604,
      "auxiliary_loss_mlp": 0.01033626,
      "balance_loss_clip": 1.05121374,
      "balance_loss_mlp": 1.02499223,
      "epoch": 0.5605723561594421,
      "flos": 25264665250560.0,
      "grad_norm": 1.864404748540004,
      "language_loss": 0.7299009,
      "learning_rate": 1.7058407670970177e-06,
      "loss": 0.75203317,
      "num_input_tokens_seen": 100577385,
      "step": 4662,
      "time_per_iteration": 2.824634313583374
    },
    {
      "auxiliary_loss_clip": 0.01189823,
      "auxiliary_loss_mlp": 0.01029729,
      "balance_loss_clip": 1.05274808,
      "balance_loss_mlp": 1.02107406,
      "epoch": 0.5606925990500812,
      "flos": 20594949621120.0,
      "grad_norm": 1.7998416771709527,
      "language_loss": 0.61704832,
      "learning_rate": 1.7050702893651643e-06,
      "loss": 0.63924384,
      "num_input_tokens_seen": 100596965,
      "step": 4663,
      "time_per_iteration": 2.6698148250579834
    },
    {
      "auxiliary_loss_clip": 0.0118455,
      "auxiliary_loss_mlp": 0.0102819,
      "balance_loss_clip": 1.05388749,
      "balance_loss_mlp": 1.01935709,
      "epoch": 0.5608128419407202,
      "flos": 35006044677120.0,
      "grad_norm": 7.5654244396908705,
      "language_loss": 0.75382119,
      "learning_rate": 1.7042998563739134e-06,
      "loss": 0.77594864,
      "num_input_tokens_seen": 100615315,
      "step": 4664,
      "time_per_iteration": 2.8368642330169678
    },
    {
      "auxiliary_loss_clip": 0.01194772,
      "auxiliary_loss_mlp": 0.01034423,
      "balance_loss_clip": 1.04986668,
      "balance_loss_mlp": 1.02542305,
      "epoch": 0.5609330848313594,
      "flos": 24639819235200.0,
      "grad_norm": 1.9744663963522977,
      "language_loss": 0.71595061,
      "learning_rate": 1.703529468240139e-06,
      "loss": 0.73824263,
      "num_input_tokens_seen": 100634185,
      "step": 4665,
      "time_per_iteration": 3.547125816345215
    },
    {
      "auxiliary_loss_clip": 0.01180847,
      "auxiliary_loss_mlp": 0.01027729,
      "balance_loss_clip": 1.05177057,
      "balance_loss_mlp": 1.0191524,
      "epoch": 0.5610533277219985,
      "flos": 18762894385920.0,
      "grad_norm": 2.4326393392555237,
      "language_loss": 0.73600143,
      "learning_rate": 1.7027591250807088e-06,
      "loss": 0.75808716,
      "num_input_tokens_seen": 100651360,
      "step": 4666,
      "time_per_iteration": 2.753309488296509
    },
    {
      "auxiliary_loss_clip": 0.0118571,
      "auxiliary_loss_mlp": 0.01028082,
      "balance_loss_clip": 1.05697548,
      "balance_loss_mlp": 1.01923621,
      "epoch": 0.5611735706126375,
      "flos": 15012384727680.0,
      "grad_norm": 2.2935560305814016,
      "language_loss": 0.84677458,
      "learning_rate": 1.7019888270124825e-06,
      "loss": 0.86891258,
      "num_input_tokens_seen": 100668525,
      "step": 4667,
      "time_per_iteration": 2.6374130249023438
    },
    {
      "auxiliary_loss_clip": 0.01188983,
      "auxiliary_loss_mlp": 0.01032312,
      "balance_loss_clip": 1.05481648,
      "balance_loss_mlp": 1.02323461,
      "epoch": 0.5612938135032767,
      "flos": 16468167041280.0,
      "grad_norm": 1.91053860123085,
      "language_loss": 0.82033336,
      "learning_rate": 1.7012185741523147e-06,
      "loss": 0.84254628,
      "num_input_tokens_seen": 100684850,
      "step": 4668,
      "time_per_iteration": 2.5448319911956787
    },
    {
      "auxiliary_loss_clip": 0.01185111,
      "auxiliary_loss_mlp": 0.01030514,
      "balance_loss_clip": 1.05759656,
      "balance_loss_mlp": 1.02231801,
      "epoch": 0.5614140563939157,
      "flos": 25666433850240.0,
      "grad_norm": 2.0132160759832023,
      "language_loss": 0.62685525,
      "learning_rate": 1.7004483666170514e-06,
      "loss": 0.64901149,
      "num_input_tokens_seen": 100705345,
      "step": 4669,
      "time_per_iteration": 2.6349408626556396
    },
    {
      "auxiliary_loss_clip": 0.01184052,
      "auxiliary_loss_mlp": 0.01027662,
      "balance_loss_clip": 1.05283499,
      "balance_loss_mlp": 1.01981831,
      "epoch": 0.5615342992845548,
      "flos": 24717566223360.0,
      "grad_norm": 1.9013077231327031,
      "language_loss": 0.80320179,
      "learning_rate": 1.699678204523533e-06,
      "loss": 0.82531893,
      "num_input_tokens_seen": 100725210,
      "step": 4670,
      "time_per_iteration": 2.632131814956665
    },
    {
      "auxiliary_loss_clip": 0.01191699,
      "auxiliary_loss_mlp": 0.01027666,
      "balance_loss_clip": 1.0543201,
      "balance_loss_mlp": 1.018296,
      "epoch": 0.5616545421751938,
      "flos": 22015934634240.0,
      "grad_norm": 2.763663266733416,
      "language_loss": 0.68902904,
      "learning_rate": 1.6989080879885918e-06,
      "loss": 0.71122265,
      "num_input_tokens_seen": 100743070,
      "step": 4671,
      "time_per_iteration": 2.6172478199005127
    },
    {
      "auxiliary_loss_clip": 0.01096786,
      "auxiliary_loss_mlp": 0.01002783,
      "balance_loss_clip": 1.02048254,
      "balance_loss_mlp": 1.00202012,
      "epoch": 0.561774785065833,
      "flos": 53760358690560.0,
      "grad_norm": 0.8902751695782674,
      "language_loss": 0.61009109,
      "learning_rate": 1.6981380171290544e-06,
      "loss": 0.63108683,
      "num_input_tokens_seen": 100804095,
      "step": 4672,
      "time_per_iteration": 3.1390013694763184
    },
    {
      "auxiliary_loss_clip": 0.01180448,
      "auxiliary_loss_mlp": 0.01029626,
      "balance_loss_clip": 1.04735112,
      "balance_loss_mlp": 1.02116251,
      "epoch": 0.5618950279564721,
      "flos": 19750007018880.0,
      "grad_norm": 1.9102882409788988,
      "language_loss": 0.74381053,
      "learning_rate": 1.6973679920617396e-06,
      "loss": 0.76591134,
      "num_input_tokens_seen": 100821630,
      "step": 4673,
      "time_per_iteration": 2.6296563148498535
    },
    {
      "auxiliary_loss_clip": 0.01185979,
      "auxiliary_loss_mlp": 0.01030584,
      "balance_loss_clip": 1.05312419,
      "balance_loss_mlp": 1.02161968,
      "epoch": 0.5620152708471111,
      "flos": 16800592435200.0,
      "grad_norm": 2.4857375714892322,
      "language_loss": 0.85346431,
      "learning_rate": 1.6965980129034603e-06,
      "loss": 0.8756299,
      "num_input_tokens_seen": 100839015,
      "step": 4674,
      "time_per_iteration": 2.5249993801116943
    },
    {
      "auxiliary_loss_clip": 0.01188455,
      "auxiliary_loss_mlp": 0.01028197,
      "balance_loss_clip": 1.05406761,
      "balance_loss_mlp": 1.0201211,
      "epoch": 0.5621355137377503,
      "flos": 26797799502720.0,
      "grad_norm": 1.5490881479424938,
      "language_loss": 0.766222,
      "learning_rate": 1.6958280797710209e-06,
      "loss": 0.78838849,
      "num_input_tokens_seen": 100860940,
      "step": 4675,
      "time_per_iteration": 2.6603500843048096
    },
    {
      "auxiliary_loss_clip": 0.0109002,
      "auxiliary_loss_mlp": 0.01000511,
      "balance_loss_clip": 1.01922917,
      "balance_loss_mlp": 0.99968892,
      "epoch": 0.5622557566283893,
      "flos": 61207046686080.0,
      "grad_norm": 0.7198956782325162,
      "language_loss": 0.54779565,
      "learning_rate": 1.6950581927812198e-06,
      "loss": 0.56870097,
      "num_input_tokens_seen": 100920510,
      "step": 4676,
      "time_per_iteration": 3.1765003204345703
    },
    {
      "auxiliary_loss_clip": 0.01183529,
      "auxiliary_loss_mlp": 0.01026991,
      "balance_loss_clip": 1.0527972,
      "balance_loss_mlp": 1.01833606,
      "epoch": 0.5623759995190284,
      "flos": 26468534505600.0,
      "grad_norm": 2.136593468536588,
      "language_loss": 0.79700851,
      "learning_rate": 1.6942883520508486e-06,
      "loss": 0.81911373,
      "num_input_tokens_seen": 100939245,
      "step": 4677,
      "time_per_iteration": 2.629241943359375
    },
    {
      "auxiliary_loss_clip": 0.01185385,
      "auxiliary_loss_mlp": 0.01027678,
      "balance_loss_clip": 1.05307603,
      "balance_loss_mlp": 1.01918459,
      "epoch": 0.5624962424096676,
      "flos": 19390900798080.0,
      "grad_norm": 2.0987910285603473,
      "language_loss": 0.77030218,
      "learning_rate": 1.693518557696691e-06,
      "loss": 0.79243279,
      "num_input_tokens_seen": 100958385,
      "step": 4678,
      "time_per_iteration": 2.5896670818328857
    },
    {
      "auxiliary_loss_clip": 0.0118089,
      "auxiliary_loss_mlp": 0.01024673,
      "balance_loss_clip": 1.05064082,
      "balance_loss_mlp": 1.01620293,
      "epoch": 0.5626164853003066,
      "flos": 20667345482880.0,
      "grad_norm": 2.9772538984400074,
      "language_loss": 0.88885254,
      "learning_rate": 1.6927488098355252e-06,
      "loss": 0.9109081,
      "num_input_tokens_seen": 100976015,
      "step": 4679,
      "time_per_iteration": 2.601341485977173
    },
    {
      "auxiliary_loss_clip": 0.01108016,
      "auxiliary_loss_mlp": 0.01001481,
      "balance_loss_clip": 1.01831889,
      "balance_loss_mlp": 1.00067067,
      "epoch": 0.5627367281909457,
      "flos": 62766071665920.0,
      "grad_norm": 0.9114285118689147,
      "language_loss": 0.63178408,
      "learning_rate": 1.6919791085841201e-06,
      "loss": 0.65287906,
      "num_input_tokens_seen": 101033425,
      "step": 4680,
      "time_per_iteration": 4.033419370651245
    },
    {
      "auxiliary_loss_clip": 0.01178392,
      "auxiliary_loss_mlp": 0.01031945,
      "balance_loss_clip": 1.0496012,
      "balance_loss_mlp": 1.02274156,
      "epoch": 0.5628569710815848,
      "flos": 12787144243200.0,
      "grad_norm": 2.4126928980837303,
      "language_loss": 0.78973091,
      "learning_rate": 1.6912094540592396e-06,
      "loss": 0.81183428,
      "num_input_tokens_seen": 101048945,
      "step": 4681,
      "time_per_iteration": 2.515080690383911
    },
    {
      "auxiliary_loss_clip": 0.0118309,
      "auxiliary_loss_mlp": 0.0103081,
      "balance_loss_clip": 1.05172503,
      "balance_loss_mlp": 1.02244699,
      "epoch": 0.5629772139722239,
      "flos": 13762082165760.0,
      "grad_norm": 2.492773493360484,
      "language_loss": 0.8175683,
      "learning_rate": 1.6904398463776393e-06,
      "loss": 0.83970731,
      "num_input_tokens_seen": 101062745,
      "step": 4682,
      "time_per_iteration": 2.5585508346557617
    },
    {
      "auxiliary_loss_clip": 0.01183649,
      "auxiliary_loss_mlp": 0.01024148,
      "balance_loss_clip": 1.05031765,
      "balance_loss_mlp": 1.01575518,
      "epoch": 0.5630974568628629,
      "flos": 21467830026240.0,
      "grad_norm": 1.9219391613818642,
      "language_loss": 0.72894841,
      "learning_rate": 1.6896702856560683e-06,
      "loss": 0.75102639,
      "num_input_tokens_seen": 101081840,
      "step": 4683,
      "time_per_iteration": 3.521772623062134
    },
    {
      "auxiliary_loss_clip": 0.01185924,
      "auxiliary_loss_mlp": 0.01025972,
      "balance_loss_clip": 1.0451839,
      "balance_loss_mlp": 1.01761532,
      "epoch": 0.5632176997535021,
      "flos": 14245907385600.0,
      "grad_norm": 3.6168794629562284,
      "language_loss": 0.69395363,
      "learning_rate": 1.6889007720112677e-06,
      "loss": 0.71607262,
      "num_input_tokens_seen": 101099585,
      "step": 4684,
      "time_per_iteration": 3.5722920894622803
    },
    {
      "auxiliary_loss_clip": 0.01186147,
      "auxiliary_loss_mlp": 0.01026778,
      "balance_loss_clip": 1.05356479,
      "balance_loss_mlp": 1.0188477,
      "epoch": 0.5633379426441412,
      "flos": 20812244947200.0,
      "grad_norm": 1.7336398328297806,
      "language_loss": 0.77366281,
      "learning_rate": 1.6881313055599734e-06,
      "loss": 0.79579204,
      "num_input_tokens_seen": 101119515,
      "step": 4685,
      "time_per_iteration": 2.5224642753601074
    },
    {
      "auxiliary_loss_clip": 0.0117358,
      "auxiliary_loss_mlp": 0.01023523,
      "balance_loss_clip": 1.04718828,
      "balance_loss_mlp": 1.01460576,
      "epoch": 0.5634581855347802,
      "flos": 22600883617920.0,
      "grad_norm": 2.384979400048974,
      "language_loss": 0.82275975,
      "learning_rate": 1.6873618864189117e-06,
      "loss": 0.84473073,
      "num_input_tokens_seen": 101135285,
      "step": 4686,
      "time_per_iteration": 2.4994957447052
    },
    {
      "auxiliary_loss_clip": 0.01184377,
      "auxiliary_loss_mlp": 0.01032229,
      "balance_loss_clip": 1.05210364,
      "balance_loss_mlp": 1.02327609,
      "epoch": 0.5635784284254194,
      "flos": 21506972872320.0,
      "grad_norm": 2.681578240115915,
      "language_loss": 0.78030324,
      "learning_rate": 1.686592514704803e-06,
      "loss": 0.80246925,
      "num_input_tokens_seen": 101152680,
      "step": 4687,
      "time_per_iteration": 2.502624273300171
    },
    {
      "auxiliary_loss_clip": 0.01184882,
      "auxiliary_loss_mlp": 0.0102558,
      "balance_loss_clip": 1.05376089,
      "balance_loss_mlp": 1.01787329,
      "epoch": 0.5636986713160584,
      "flos": 19827466698240.0,
      "grad_norm": 2.5151038855189882,
      "language_loss": 0.71035057,
      "learning_rate": 1.685823190534361e-06,
      "loss": 0.73245519,
      "num_input_tokens_seen": 101170920,
      "step": 4688,
      "time_per_iteration": 2.535834550857544
    },
    {
      "auxiliary_loss_clip": 0.01184465,
      "auxiliary_loss_mlp": 0.01024592,
      "balance_loss_clip": 1.05429304,
      "balance_loss_mlp": 1.01565111,
      "epoch": 0.5638189142066975,
      "flos": 19792453916160.0,
      "grad_norm": 2.1749578448773264,
      "language_loss": 0.83632612,
      "learning_rate": 1.6850539140242907e-06,
      "loss": 0.85841668,
      "num_input_tokens_seen": 101190180,
      "step": 4689,
      "time_per_iteration": 2.5621895790100098
    },
    {
      "auxiliary_loss_clip": 0.01186149,
      "auxiliary_loss_mlp": 0.01029613,
      "balance_loss_clip": 1.05179262,
      "balance_loss_mlp": 1.02135205,
      "epoch": 0.5639391570973367,
      "flos": 22893771116160.0,
      "grad_norm": 1.8363013350760018,
      "language_loss": 0.82291651,
      "learning_rate": 1.684284685291292e-06,
      "loss": 0.84507406,
      "num_input_tokens_seen": 101211825,
      "step": 4690,
      "time_per_iteration": 2.5997800827026367
    },
    {
      "auxiliary_loss_clip": 0.01183324,
      "auxiliary_loss_mlp": 0.01030971,
      "balance_loss_clip": 1.0550406,
      "balance_loss_mlp": 1.02220309,
      "epoch": 0.5640593999879757,
      "flos": 23727077712000.0,
      "grad_norm": 3.047131627892479,
      "language_loss": 0.8120532,
      "learning_rate": 1.683515504452055e-06,
      "loss": 0.83419609,
      "num_input_tokens_seen": 101229200,
      "step": 4691,
      "time_per_iteration": 2.585092544555664
    },
    {
      "auxiliary_loss_clip": 0.01178214,
      "auxiliary_loss_mlp": 0.01033368,
      "balance_loss_clip": 1.04581594,
      "balance_loss_mlp": 1.02424276,
      "epoch": 0.5641796428786148,
      "flos": 22710123855360.0,
      "grad_norm": 4.279594227473624,
      "language_loss": 0.6645366,
      "learning_rate": 1.6827463716232648e-06,
      "loss": 0.68665242,
      "num_input_tokens_seen": 101249860,
      "step": 4692,
      "time_per_iteration": 3.5508463382720947
    },
    {
      "auxiliary_loss_clip": 0.01180311,
      "auxiliary_loss_mlp": 0.01208781,
      "balance_loss_clip": 1.05075622,
      "balance_loss_mlp": 1.00072229,
      "epoch": 0.5642998857692539,
      "flos": 19791987039360.0,
      "grad_norm": 1.6547492864796878,
      "language_loss": 0.75314498,
      "learning_rate": 1.6819772869215972e-06,
      "loss": 0.77703595,
      "num_input_tokens_seen": 101268940,
      "step": 4693,
      "time_per_iteration": 2.575258255004883
    },
    {
      "auxiliary_loss_clip": 0.01189074,
      "auxiliary_loss_mlp": 0.01026366,
      "balance_loss_clip": 1.05089617,
      "balance_loss_mlp": 1.0187428,
      "epoch": 0.564420128659893,
      "flos": 23185904428800.0,
      "grad_norm": 1.7332741233771154,
      "language_loss": 0.81946039,
      "learning_rate": 1.6812082504637228e-06,
      "loss": 0.84161478,
      "num_input_tokens_seen": 101290260,
      "step": 4694,
      "time_per_iteration": 2.6409389972686768
    },
    {
      "auxiliary_loss_clip": 0.01178947,
      "auxiliary_loss_mlp": 0.01027188,
      "balance_loss_clip": 1.05166411,
      "balance_loss_mlp": 1.0189867,
      "epoch": 0.564540371550532,
      "flos": 23258264376960.0,
      "grad_norm": 1.5638505074640676,
      "language_loss": 0.74236196,
      "learning_rate": 1.6804392623663025e-06,
      "loss": 0.76442331,
      "num_input_tokens_seen": 101311465,
      "step": 4695,
      "time_per_iteration": 2.6026320457458496
    },
    {
      "auxiliary_loss_clip": 0.0117442,
      "auxiliary_loss_mlp": 0.01025672,
      "balance_loss_clip": 1.04966021,
      "balance_loss_mlp": 1.01739883,
      "epoch": 0.5646606144411712,
      "flos": 25010058672000.0,
      "grad_norm": 2.9960678630389546,
      "language_loss": 0.78502893,
      "learning_rate": 1.6796703227459935e-06,
      "loss": 0.80702984,
      "num_input_tokens_seen": 101329420,
      "step": 4696,
      "time_per_iteration": 2.6289801597595215
    },
    {
      "auxiliary_loss_clip": 0.01181547,
      "auxiliary_loss_mlp": 0.01024445,
      "balance_loss_clip": 1.04460907,
      "balance_loss_mlp": 1.01606488,
      "epoch": 0.5647808573318103,
      "flos": 36539645806080.0,
      "grad_norm": 1.9498670987644164,
      "language_loss": 0.76239491,
      "learning_rate": 1.6789014317194407e-06,
      "loss": 0.78445488,
      "num_input_tokens_seen": 101350900,
      "step": 4697,
      "time_per_iteration": 2.7977771759033203
    },
    {
      "auxiliary_loss_clip": 0.01193621,
      "auxiliary_loss_mlp": 0.01027403,
      "balance_loss_clip": 1.05268455,
      "balance_loss_mlp": 1.01862931,
      "epoch": 0.5649011002224493,
      "flos": 22528451842560.0,
      "grad_norm": 2.843686996345989,
      "language_loss": 0.73007429,
      "learning_rate": 1.6781325894032853e-06,
      "loss": 0.75228453,
      "num_input_tokens_seen": 101369860,
      "step": 4698,
      "time_per_iteration": 2.671379804611206
    },
    {
      "auxiliary_loss_clip": 0.01177648,
      "auxiliary_loss_mlp": 0.01029422,
      "balance_loss_clip": 1.04963207,
      "balance_loss_mlp": 1.02123857,
      "epoch": 0.5650213431130885,
      "flos": 18515147304960.0,
      "grad_norm": 4.813829366857776,
      "language_loss": 0.92320991,
      "learning_rate": 1.6773637959141608e-06,
      "loss": 0.94528055,
      "num_input_tokens_seen": 101386835,
      "step": 4699,
      "time_per_iteration": 2.569213390350342
    },
    {
      "auxiliary_loss_clip": 0.01174418,
      "auxiliary_loss_mlp": 0.01028599,
      "balance_loss_clip": 1.04809272,
      "balance_loss_mlp": 1.02044499,
      "epoch": 0.5651415860037275,
      "flos": 17526310819200.0,
      "grad_norm": 2.3614134025706597,
      "language_loss": 0.66621196,
      "learning_rate": 1.6765950513686915e-06,
      "loss": 0.68824208,
      "num_input_tokens_seen": 101404945,
      "step": 4700,
      "time_per_iteration": 2.5967462062835693
    },
    {
      "auxiliary_loss_clip": 0.01189654,
      "auxiliary_loss_mlp": 0.01031452,
      "balance_loss_clip": 1.04376614,
      "balance_loss_mlp": 1.02238631,
      "epoch": 0.5652618288943666,
      "flos": 25520026014720.0,
      "grad_norm": 1.7242569845919766,
      "language_loss": 0.76207376,
      "learning_rate": 1.675826355883496e-06,
      "loss": 0.78428483,
      "num_input_tokens_seen": 101424160,
      "step": 4701,
      "time_per_iteration": 2.6427841186523438
    },
    {
      "auxiliary_loss_clip": 0.01176464,
      "auxiliary_loss_mlp": 0.010292,
      "balance_loss_clip": 1.04870439,
      "balance_loss_mlp": 1.02061725,
      "epoch": 0.5653820717850057,
      "flos": 19683105937920.0,
      "grad_norm": 1.993185812200807,
      "language_loss": 0.79478925,
      "learning_rate": 1.6750577095751848e-06,
      "loss": 0.81684589,
      "num_input_tokens_seen": 101443270,
      "step": 4702,
      "time_per_iteration": 2.6612625122070312
    },
    {
      "auxiliary_loss_clip": 0.01176256,
      "auxiliary_loss_mlp": 0.01030346,
      "balance_loss_clip": 1.05133605,
      "balance_loss_mlp": 1.02210891,
      "epoch": 0.5655023146756448,
      "flos": 26979722910720.0,
      "grad_norm": 1.7773102281343247,
      "language_loss": 0.73148108,
      "learning_rate": 1.6742891125603605e-06,
      "loss": 0.75354713,
      "num_input_tokens_seen": 101464175,
      "step": 4703,
      "time_per_iteration": 2.6103763580322266
    },
    {
      "auxiliary_loss_clip": 0.01179195,
      "auxiliary_loss_mlp": 0.01027318,
      "balance_loss_clip": 1.05080187,
      "balance_loss_mlp": 1.01846623,
      "epoch": 0.5656225575662839,
      "flos": 27669351104640.0,
      "grad_norm": 2.57329588086481,
      "language_loss": 0.72366375,
      "learning_rate": 1.6735205649556185e-06,
      "loss": 0.74572897,
      "num_input_tokens_seen": 101484045,
      "step": 4704,
      "time_per_iteration": 2.661700963973999
    },
    {
      "auxiliary_loss_clip": 0.01190452,
      "auxiliary_loss_mlp": 0.01026884,
      "balance_loss_clip": 1.0487864,
      "balance_loss_mlp": 1.01887953,
      "epoch": 0.5657428004569229,
      "flos": 24349732997760.0,
      "grad_norm": 2.1273268110830394,
      "language_loss": 0.84861028,
      "learning_rate": 1.6727520668775476e-06,
      "loss": 0.87078369,
      "num_input_tokens_seen": 101504330,
      "step": 4705,
      "time_per_iteration": 2.688451051712036
    },
    {
      "auxiliary_loss_clip": 0.01181069,
      "auxiliary_loss_mlp": 0.01030819,
      "balance_loss_clip": 1.0522151,
      "balance_loss_mlp": 1.0221051,
      "epoch": 0.5658630433475621,
      "flos": 21944041562880.0,
      "grad_norm": 1.742620778726295,
      "language_loss": 0.755234,
      "learning_rate": 1.6719836184427275e-06,
      "loss": 0.77735281,
      "num_input_tokens_seen": 101524635,
      "step": 4706,
      "time_per_iteration": 3.4804270267486572
    },
    {
      "auxiliary_loss_clip": 0.0118154,
      "auxiliary_loss_mlp": 0.01025283,
      "balance_loss_clip": 1.04743207,
      "balance_loss_mlp": 1.01746845,
      "epoch": 0.5659832862382012,
      "flos": 30409012218240.0,
      "grad_norm": 1.7954196524197354,
      "language_loss": 0.6420846,
      "learning_rate": 1.671215219767733e-06,
      "loss": 0.66415286,
      "num_input_tokens_seen": 101544095,
      "step": 4707,
      "time_per_iteration": 2.6685028076171875
    },
    {
      "auxiliary_loss_clip": 0.01189363,
      "auxiliary_loss_mlp": 0.01029029,
      "balance_loss_clip": 1.04651237,
      "balance_loss_mlp": 1.02100062,
      "epoch": 0.5661035291288402,
      "flos": 13188194570880.0,
      "grad_norm": 1.8982802654570425,
      "language_loss": 0.76499945,
      "learning_rate": 1.670446870969127e-06,
      "loss": 0.7871834,
      "num_input_tokens_seen": 101561760,
      "step": 4708,
      "time_per_iteration": 2.6184840202331543
    },
    {
      "auxiliary_loss_clip": 0.01187215,
      "auxiliary_loss_mlp": 0.01028522,
      "balance_loss_clip": 1.05012524,
      "balance_loss_mlp": 1.02034473,
      "epoch": 0.5662237720194794,
      "flos": 16143032108160.0,
      "grad_norm": 2.3472527238476024,
      "language_loss": 0.80449206,
      "learning_rate": 1.6696785721634685e-06,
      "loss": 0.82664943,
      "num_input_tokens_seen": 101576245,
      "step": 4709,
      "time_per_iteration": 2.6235063076019287
    },
    {
      "auxiliary_loss_clip": 0.01183417,
      "auxiliary_loss_mlp": 0.01035172,
      "balance_loss_clip": 1.05118525,
      "balance_loss_mlp": 1.02641296,
      "epoch": 0.5663440149101184,
      "flos": 17676848718720.0,
      "grad_norm": 2.366199717134748,
      "language_loss": 0.73901916,
      "learning_rate": 1.6689103234673086e-06,
      "loss": 0.76120502,
      "num_input_tokens_seen": 101594565,
      "step": 4710,
      "time_per_iteration": 3.47572660446167
    },
    {
      "auxiliary_loss_clip": 0.01181651,
      "auxiliary_loss_mlp": 0.01029317,
      "balance_loss_clip": 1.05059028,
      "balance_loss_mlp": 1.02090108,
      "epoch": 0.5664642578007575,
      "flos": 23368330627200.0,
      "grad_norm": 2.8048736492030595,
      "language_loss": 0.76841456,
      "learning_rate": 1.668142124997189e-06,
      "loss": 0.79052424,
      "num_input_tokens_seen": 101614225,
      "step": 4711,
      "time_per_iteration": 3.488166093826294
    },
    {
      "auxiliary_loss_clip": 0.01079078,
      "auxiliary_loss_mlp": 0.01000544,
      "balance_loss_clip": 1.01592135,
      "balance_loss_mlp": 0.99972719,
      "epoch": 0.5665845006913967,
      "flos": 65516470945920.0,
      "grad_norm": 0.7310641284759603,
      "language_loss": 0.59740782,
      "learning_rate": 1.6673739768696453e-06,
      "loss": 0.618204,
      "num_input_tokens_seen": 101680795,
      "step": 4712,
      "time_per_iteration": 3.1786587238311768
    },
    {
      "auxiliary_loss_clip": 0.01190129,
      "auxiliary_loss_mlp": 0.01026311,
      "balance_loss_clip": 1.04967809,
      "balance_loss_mlp": 1.01785898,
      "epoch": 0.5667047435820357,
      "flos": 26140885620480.0,
      "grad_norm": 1.6668257371685171,
      "language_loss": 0.77412778,
      "learning_rate": 1.6666058792012052e-06,
      "loss": 0.79629219,
      "num_input_tokens_seen": 101701680,
      "step": 4713,
      "time_per_iteration": 2.6389050483703613
    },
    {
      "auxiliary_loss_clip": 0.01082848,
      "auxiliary_loss_mlp": 0.01000983,
      "balance_loss_clip": 1.0156287,
      "balance_loss_mlp": 1.00022054,
      "epoch": 0.5668249864726748,
      "flos": 71866949725440.0,
      "grad_norm": 0.9182302656124544,
      "language_loss": 0.68761837,
      "learning_rate": 1.6658378321083878e-06,
      "loss": 0.7084567,
      "num_input_tokens_seen": 101766010,
      "step": 4714,
      "time_per_iteration": 3.183619499206543
    },
    {
      "auxiliary_loss_clip": 0.01195686,
      "auxiliary_loss_mlp": 0.01025557,
      "balance_loss_clip": 1.04580343,
      "balance_loss_mlp": 1.0178889,
      "epoch": 0.5669452293633139,
      "flos": 22195667312640.0,
      "grad_norm": 1.955255273961141,
      "language_loss": 0.82867759,
      "learning_rate": 1.6650698357077055e-06,
      "loss": 0.85088998,
      "num_input_tokens_seen": 101783055,
      "step": 4715,
      "time_per_iteration": 2.683086633682251
    },
    {
      "auxiliary_loss_clip": 0.0118763,
      "auxiliary_loss_mlp": 0.01033389,
      "balance_loss_clip": 1.04878235,
      "balance_loss_mlp": 1.02441847,
      "epoch": 0.567065472253953,
      "flos": 18223193560320.0,
      "grad_norm": 3.7650718244926944,
      "language_loss": 0.81121433,
      "learning_rate": 1.6643018901156632e-06,
      "loss": 0.83342451,
      "num_input_tokens_seen": 101802150,
      "step": 4716,
      "time_per_iteration": 2.5882887840270996
    },
    {
      "auxiliary_loss_clip": 0.01189919,
      "auxiliary_loss_mlp": 0.01023608,
      "balance_loss_clip": 1.05072808,
      "balance_loss_mlp": 1.01563859,
      "epoch": 0.567185715144592,
      "flos": 20371548983040.0,
      "grad_norm": 3.21674303930847,
      "language_loss": 0.79006058,
      "learning_rate": 1.6635339954487566e-06,
      "loss": 0.81219584,
      "num_input_tokens_seen": 101818025,
      "step": 4717,
      "time_per_iteration": 2.663834571838379
    },
    {
      "auxiliary_loss_clip": 0.0118813,
      "auxiliary_loss_mlp": 0.01025728,
      "balance_loss_clip": 1.0499835,
      "balance_loss_mlp": 1.01737714,
      "epoch": 0.5673059580352312,
      "flos": 23221348174080.0,
      "grad_norm": 3.7556658004562897,
      "language_loss": 0.82330853,
      "learning_rate": 1.6627661518234765e-06,
      "loss": 0.84544706,
      "num_input_tokens_seen": 101837280,
      "step": 4718,
      "time_per_iteration": 2.611722946166992
    },
    {
      "auxiliary_loss_clip": 0.01188977,
      "auxiliary_loss_mlp": 0.01025727,
      "balance_loss_clip": 1.04724145,
      "balance_loss_mlp": 1.01757014,
      "epoch": 0.5674262009258703,
      "flos": 21719599430400.0,
      "grad_norm": 2.7667778619534698,
      "language_loss": 0.85837269,
      "learning_rate": 1.661998359356302e-06,
      "loss": 0.88051975,
      "num_input_tokens_seen": 101856310,
      "step": 4719,
      "time_per_iteration": 3.5745673179626465
    },
    {
      "auxiliary_loss_clip": 0.01075101,
      "auxiliary_loss_mlp": 0.01000281,
      "balance_loss_clip": 1.01601529,
      "balance_loss_mlp": 0.99953312,
      "epoch": 0.5675464438165093,
      "flos": 67470369114240.0,
      "grad_norm": 0.7485886425207188,
      "language_loss": 0.55722374,
      "learning_rate": 1.6612306181637077e-06,
      "loss": 0.57797754,
      "num_input_tokens_seen": 101915635,
      "step": 4720,
      "time_per_iteration": 3.09881854057312
    },
    {
      "auxiliary_loss_clip": 0.01183133,
      "auxiliary_loss_mlp": 0.01030415,
      "balance_loss_clip": 1.04714525,
      "balance_loss_mlp": 1.02280378,
      "epoch": 0.5676666867071485,
      "flos": 18879173688960.0,
      "grad_norm": 2.9766839199333743,
      "language_loss": 0.6589973,
      "learning_rate": 1.6604629283621598e-06,
      "loss": 0.68113279,
      "num_input_tokens_seen": 101933565,
      "step": 4721,
      "time_per_iteration": 2.6013550758361816
    },
    {
      "auxiliary_loss_clip": 0.01181513,
      "auxiliary_loss_mlp": 0.01032767,
      "balance_loss_clip": 1.05292773,
      "balance_loss_mlp": 1.02414846,
      "epoch": 0.5677869295977875,
      "flos": 33546778744320.0,
      "grad_norm": 2.1352584123310874,
      "language_loss": 0.74418628,
      "learning_rate": 1.6596952900681152e-06,
      "loss": 0.76632911,
      "num_input_tokens_seen": 101954325,
      "step": 4722,
      "time_per_iteration": 2.7206389904022217
    },
    {
      "auxiliary_loss_clip": 0.01178866,
      "auxiliary_loss_mlp": 0.01028633,
      "balance_loss_clip": 1.04814124,
      "balance_loss_mlp": 1.01990724,
      "epoch": 0.5679071724884266,
      "flos": 28037256157440.0,
      "grad_norm": 2.3055330206415814,
      "language_loss": 0.81759036,
      "learning_rate": 1.658927703398025e-06,
      "loss": 0.83966535,
      "num_input_tokens_seen": 101974390,
      "step": 4723,
      "time_per_iteration": 2.7092576026916504
    },
    {
      "auxiliary_loss_clip": 0.01186703,
      "auxiliary_loss_mlp": 0.01025635,
      "balance_loss_clip": 1.04143667,
      "balance_loss_mlp": 1.01748085,
      "epoch": 0.5680274153790658,
      "flos": 23550110380800.0,
      "grad_norm": 2.7451841504258883,
      "language_loss": 0.77776015,
      "learning_rate": 1.6581601684683309e-06,
      "loss": 0.79988348,
      "num_input_tokens_seen": 101994815,
      "step": 4724,
      "time_per_iteration": 2.724271535873413
    },
    {
      "auxiliary_loss_clip": 0.01183113,
      "auxiliary_loss_mlp": 0.01028485,
      "balance_loss_clip": 1.0537014,
      "balance_loss_mlp": 1.02107608,
      "epoch": 0.5681476582697048,
      "flos": 22455158140800.0,
      "grad_norm": 2.599603721027836,
      "language_loss": 0.68574572,
      "learning_rate": 1.6573926853954674e-06,
      "loss": 0.70786166,
      "num_input_tokens_seen": 102012400,
      "step": 4725,
      "time_per_iteration": 2.607166290283203
    },
    {
      "auxiliary_loss_clip": 0.01173277,
      "auxiliary_loss_mlp": 0.01025091,
      "balance_loss_clip": 1.04367292,
      "balance_loss_mlp": 1.01669919,
      "epoch": 0.5682679011603439,
      "flos": 19536913584000.0,
      "grad_norm": 2.148001686247693,
      "language_loss": 0.83156419,
      "learning_rate": 1.6566252542958608e-06,
      "loss": 0.85354793,
      "num_input_tokens_seen": 102031900,
      "step": 4726,
      "time_per_iteration": 2.7089011669158936
    },
    {
      "auxiliary_loss_clip": 0.01178464,
      "auxiliary_loss_mlp": 0.01028868,
      "balance_loss_clip": 1.046731,
      "balance_loss_mlp": 1.02061009,
      "epoch": 0.568388144050983,
      "flos": 28765488493440.0,
      "grad_norm": 2.028161984285372,
      "language_loss": 0.78375149,
      "learning_rate": 1.6558578752859305e-06,
      "loss": 0.80582482,
      "num_input_tokens_seen": 102050860,
      "step": 4727,
      "time_per_iteration": 2.710399866104126
    },
    {
      "auxiliary_loss_clip": 0.01182956,
      "auxiliary_loss_mlp": 0.01024002,
      "balance_loss_clip": 1.04606116,
      "balance_loss_mlp": 1.01653337,
      "epoch": 0.5685083869416221,
      "flos": 21209452519680.0,
      "grad_norm": 1.82477845548556,
      "language_loss": 0.78566575,
      "learning_rate": 1.6550905484820865e-06,
      "loss": 0.80773532,
      "num_input_tokens_seen": 102069320,
      "step": 4728,
      "time_per_iteration": 2.6773691177368164
    },
    {
      "auxiliary_loss_clip": 0.01178466,
      "auxiliary_loss_mlp": 0.01028863,
      "balance_loss_clip": 1.0515697,
      "balance_loss_mlp": 1.02026248,
      "epoch": 0.5686286298322611,
      "flos": 24827021942400.0,
      "grad_norm": 2.3023150717615697,
      "language_loss": 0.79328197,
      "learning_rate": 1.6543232740007328e-06,
      "loss": 0.8153553,
      "num_input_tokens_seen": 102086435,
      "step": 4729,
      "time_per_iteration": 2.5839695930480957
    },
    {
      "auxiliary_loss_clip": 0.01181218,
      "auxiliary_loss_mlp": 0.01027714,
      "balance_loss_clip": 1.05038369,
      "balance_loss_mlp": 1.01982844,
      "epoch": 0.5687488727229003,
      "flos": 26615121909120.0,
      "grad_norm": 2.8237232519497533,
      "language_loss": 0.67256868,
      "learning_rate": 1.653556051958263e-06,
      "loss": 0.69465804,
      "num_input_tokens_seen": 102106115,
      "step": 4730,
      "time_per_iteration": 2.611128091812134
    },
    {
      "auxiliary_loss_clip": 0.01173801,
      "auxiliary_loss_mlp": 0.01027602,
      "balance_loss_clip": 1.04300404,
      "balance_loss_mlp": 1.01905477,
      "epoch": 0.5688691156135394,
      "flos": 20808725414400.0,
      "grad_norm": 1.7294253910580455,
      "language_loss": 0.73919845,
      "learning_rate": 1.6527888824710642e-06,
      "loss": 0.76121247,
      "num_input_tokens_seen": 102125715,
      "step": 4731,
      "time_per_iteration": 2.6286416053771973
    },
    {
      "auxiliary_loss_clip": 0.01178141,
      "auxiliary_loss_mlp": 0.01030446,
      "balance_loss_clip": 1.04443479,
      "balance_loss_mlp": 1.02171421,
      "epoch": 0.5689893585041784,
      "flos": 25880963829120.0,
      "grad_norm": 2.5326635814889924,
      "language_loss": 0.76975435,
      "learning_rate": 1.6520217656555166e-06,
      "loss": 0.7918402,
      "num_input_tokens_seen": 102145005,
      "step": 4732,
      "time_per_iteration": 2.56903076171875
    },
    {
      "auxiliary_loss_clip": 0.01169169,
      "auxiliary_loss_mlp": 0.01028451,
      "balance_loss_clip": 1.04536855,
      "balance_loss_mlp": 1.02029669,
      "epoch": 0.5691096013948175,
      "flos": 23477463123840.0,
      "grad_norm": 1.5136300774827245,
      "language_loss": 0.7109158,
      "learning_rate": 1.65125470162799e-06,
      "loss": 0.73289204,
      "num_input_tokens_seen": 102165360,
      "step": 4733,
      "time_per_iteration": 3.374347686767578
    },
    {
      "auxiliary_loss_clip": 0.01187069,
      "auxiliary_loss_mlp": 0.01030414,
      "balance_loss_clip": 1.04580152,
      "balance_loss_mlp": 1.02226067,
      "epoch": 0.5692298442854566,
      "flos": 18075600576000.0,
      "grad_norm": 8.931149539168958,
      "language_loss": 0.69539094,
      "learning_rate": 1.6504876905048485e-06,
      "loss": 0.71756577,
      "num_input_tokens_seen": 102182320,
      "step": 4734,
      "time_per_iteration": 2.503620147705078
    },
    {
      "auxiliary_loss_clip": 0.01177165,
      "auxiliary_loss_mlp": 0.01024266,
      "balance_loss_clip": 1.05354035,
      "balance_loss_mlp": 1.01707208,
      "epoch": 0.5693500871760957,
      "flos": 23039317025280.0,
      "grad_norm": 1.8492157070299728,
      "language_loss": 0.72086859,
      "learning_rate": 1.6497207324024464e-06,
      "loss": 0.74288291,
      "num_input_tokens_seen": 102201220,
      "step": 4735,
      "time_per_iteration": 2.67050838470459
    },
    {
      "auxiliary_loss_clip": 0.01188952,
      "auxiliary_loss_mlp": 0.01030551,
      "balance_loss_clip": 1.04921198,
      "balance_loss_mlp": 1.02273107,
      "epoch": 0.5694703300667348,
      "flos": 18989670902400.0,
      "grad_norm": 2.1601567546390927,
      "language_loss": 0.82970017,
      "learning_rate": 1.6489538274371305e-06,
      "loss": 0.85189521,
      "num_input_tokens_seen": 102219825,
      "step": 4736,
      "time_per_iteration": 3.543668746948242
    },
    {
      "auxiliary_loss_clip": 0.0117319,
      "auxiliary_loss_mlp": 0.0102729,
      "balance_loss_clip": 1.0512085,
      "balance_loss_mlp": 1.01973188,
      "epoch": 0.5695905729573739,
      "flos": 21908705558400.0,
      "grad_norm": 2.010174431788345,
      "language_loss": 0.83234131,
      "learning_rate": 1.6481869757252396e-06,
      "loss": 0.85434616,
      "num_input_tokens_seen": 102238160,
      "step": 4737,
      "time_per_iteration": 3.5492138862609863
    },
    {
      "auxiliary_loss_clip": 0.01179204,
      "auxiliary_loss_mlp": 0.01025561,
      "balance_loss_clip": 1.05157423,
      "balance_loss_mlp": 1.01799154,
      "epoch": 0.569710815848013,
      "flos": 28476659232000.0,
      "grad_norm": 1.9552779383044618,
      "language_loss": 0.72019398,
      "learning_rate": 1.647420177383105e-06,
      "loss": 0.74224162,
      "num_input_tokens_seen": 102261030,
      "step": 4738,
      "time_per_iteration": 2.665910005569458
    },
    {
      "auxiliary_loss_clip": 0.01177109,
      "auxiliary_loss_mlp": 0.01023794,
      "balance_loss_clip": 1.05418825,
      "balance_loss_mlp": 1.01617098,
      "epoch": 0.569831058738652,
      "flos": 28366162018560.0,
      "grad_norm": 1.7859727636713096,
      "language_loss": 0.724819,
      "learning_rate": 1.646653432527049e-06,
      "loss": 0.74682802,
      "num_input_tokens_seen": 102281670,
      "step": 4739,
      "time_per_iteration": 2.616001605987549
    },
    {
      "auxiliary_loss_clip": 0.01185773,
      "auxiliary_loss_mlp": 0.01025639,
      "balance_loss_clip": 1.04802668,
      "balance_loss_mlp": 1.01776564,
      "epoch": 0.5699513016292912,
      "flos": 25849973370240.0,
      "grad_norm": 1.4818947423485582,
      "language_loss": 0.74258333,
      "learning_rate": 1.645886741273387e-06,
      "loss": 0.76469743,
      "num_input_tokens_seen": 102303485,
      "step": 4740,
      "time_per_iteration": 2.6985204219818115
    },
    {
      "auxiliary_loss_clip": 0.01178794,
      "auxiliary_loss_mlp": 0.01032919,
      "balance_loss_clip": 1.04996705,
      "balance_loss_mlp": 1.02492642,
      "epoch": 0.5700715445199303,
      "flos": 18037858360320.0,
      "grad_norm": 5.288584291014718,
      "language_loss": 0.73555517,
      "learning_rate": 1.645120103738424e-06,
      "loss": 0.75767225,
      "num_input_tokens_seen": 102320995,
      "step": 4741,
      "time_per_iteration": 2.598025321960449
    },
    {
      "auxiliary_loss_clip": 0.01166148,
      "auxiliary_loss_mlp": 0.01207663,
      "balance_loss_clip": 1.04722822,
      "balance_loss_mlp": 1.00084925,
      "epoch": 0.5701917874105693,
      "flos": 11473352392320.0,
      "grad_norm": 2.445756078358452,
      "language_loss": 0.84356058,
      "learning_rate": 1.6443535200384591e-06,
      "loss": 0.86729866,
      "num_input_tokens_seen": 102339170,
      "step": 4742,
      "time_per_iteration": 2.6018221378326416
    },
    {
      "auxiliary_loss_clip": 0.01179835,
      "auxiliary_loss_mlp": 0.01029921,
      "balance_loss_clip": 1.05407357,
      "balance_loss_mlp": 1.02168393,
      "epoch": 0.5703120303012085,
      "flos": 21761759018880.0,
      "grad_norm": 2.6804733867113333,
      "language_loss": 0.70828521,
      "learning_rate": 1.6435869902897827e-06,
      "loss": 0.7303828,
      "num_input_tokens_seen": 102357750,
      "step": 4743,
      "time_per_iteration": 2.5868382453918457
    },
    {
      "auxiliary_loss_clip": 0.01093753,
      "auxiliary_loss_mlp": 0.01004944,
      "balance_loss_clip": 1.0169208,
      "balance_loss_mlp": 1.00412738,
      "epoch": 0.5704322731918475,
      "flos": 56746258513920.0,
      "grad_norm": 0.7962602230534644,
      "language_loss": 0.61991292,
      "learning_rate": 1.6428205146086764e-06,
      "loss": 0.6408999,
      "num_input_tokens_seen": 102419730,
      "step": 4744,
      "time_per_iteration": 3.2727231979370117
    },
    {
      "auxiliary_loss_clip": 0.01186772,
      "auxiliary_loss_mlp": 0.01022634,
      "balance_loss_clip": 1.04863691,
      "balance_loss_mlp": 1.0145936,
      "epoch": 0.5705525160824866,
      "flos": 20741141975040.0,
      "grad_norm": 1.605829880750977,
      "language_loss": 0.70638871,
      "learning_rate": 1.6420540931114142e-06,
      "loss": 0.72848284,
      "num_input_tokens_seen": 102440320,
      "step": 4745,
      "time_per_iteration": 3.60969877243042
    },
    {
      "auxiliary_loss_clip": 0.01187727,
      "auxiliary_loss_mlp": 0.01039268,
      "balance_loss_clip": 1.05159414,
      "balance_loss_mlp": 1.03123307,
      "epoch": 0.5706727589731257,
      "flos": 18771262254720.0,
      "grad_norm": 1.557487499915306,
      "language_loss": 0.79060197,
      "learning_rate": 1.6412877259142616e-06,
      "loss": 0.81287193,
      "num_input_tokens_seen": 102460240,
      "step": 4746,
      "time_per_iteration": 2.6203806400299072
    },
    {
      "auxiliary_loss_clip": 0.01180291,
      "auxiliary_loss_mlp": 0.01024493,
      "balance_loss_clip": 1.05046535,
      "balance_loss_mlp": 1.01662481,
      "epoch": 0.5707930018637648,
      "flos": 27634733372160.0,
      "grad_norm": 9.704136234741277,
      "language_loss": 0.73991692,
      "learning_rate": 1.6405214131334757e-06,
      "loss": 0.7619648,
      "num_input_tokens_seen": 102478765,
      "step": 4747,
      "time_per_iteration": 2.8504562377929688
    },
    {
      "auxiliary_loss_clip": 0.01183302,
      "auxiliary_loss_mlp": 0.01022992,
      "balance_loss_clip": 1.04678869,
      "balance_loss_mlp": 1.01486778,
      "epoch": 0.5709132447544039,
      "flos": 27597673514880.0,
      "grad_norm": 1.9625945208904234,
      "language_loss": 0.79904032,
      "learning_rate": 1.6397551548853052e-06,
      "loss": 0.82110327,
      "num_input_tokens_seen": 102496930,
      "step": 4748,
      "time_per_iteration": 2.7487263679504395
    },
    {
      "auxiliary_loss_clip": 0.01182882,
      "auxiliary_loss_mlp": 0.01025712,
      "balance_loss_clip": 1.05004287,
      "balance_loss_mlp": 1.01741529,
      "epoch": 0.571033487645043,
      "flos": 21686095019520.0,
      "grad_norm": 1.920112417577248,
      "language_loss": 0.70728338,
      "learning_rate": 1.6389889512859917e-06,
      "loss": 0.72936928,
      "num_input_tokens_seen": 102516590,
      "step": 4749,
      "time_per_iteration": 2.664929151535034
    },
    {
      "auxiliary_loss_clip": 0.01084589,
      "auxiliary_loss_mlp": 0.01000415,
      "balance_loss_clip": 1.01608777,
      "balance_loss_mlp": 0.99962246,
      "epoch": 0.5711537305356821,
      "flos": 70181445980160.0,
      "grad_norm": 0.8112090648721294,
      "language_loss": 0.60362959,
      "learning_rate": 1.638222802451767e-06,
      "loss": 0.62447965,
      "num_input_tokens_seen": 102578070,
      "step": 4750,
      "time_per_iteration": 3.165015459060669
    },
    {
      "auxiliary_loss_clip": 0.01171187,
      "auxiliary_loss_mlp": 0.01022796,
      "balance_loss_clip": 1.04988456,
      "balance_loss_mlp": 1.01514304,
      "epoch": 0.5712739734263211,
      "flos": 24717494396160.0,
      "grad_norm": 1.7131350265895569,
      "language_loss": 0.75416368,
      "learning_rate": 1.6374567084988561e-06,
      "loss": 0.7761035,
      "num_input_tokens_seen": 102599255,
      "step": 4751,
      "time_per_iteration": 2.7522501945495605
    },
    {
      "auxiliary_loss_clip": 0.01188915,
      "auxiliary_loss_mlp": 0.01027722,
      "balance_loss_clip": 1.05298173,
      "balance_loss_mlp": 1.01870954,
      "epoch": 0.5713942163169603,
      "flos": 26578169792640.0,
      "grad_norm": 2.050836688997342,
      "language_loss": 0.76588809,
      "learning_rate": 1.6366906695434738e-06,
      "loss": 0.78805453,
      "num_input_tokens_seen": 102621775,
      "step": 4752,
      "time_per_iteration": 2.614769458770752
    },
    {
      "auxiliary_loss_clip": 0.01181348,
      "auxiliary_loss_mlp": 0.01026077,
      "balance_loss_clip": 1.05339956,
      "balance_loss_mlp": 1.01870382,
      "epoch": 0.5715144592075994,
      "flos": 21142443697920.0,
      "grad_norm": 2.920192915826014,
      "language_loss": 0.85738319,
      "learning_rate": 1.6359246857018275e-06,
      "loss": 0.87945741,
      "num_input_tokens_seen": 102639305,
      "step": 4753,
      "time_per_iteration": 2.617947578430176
    },
    {
      "auxiliary_loss_clip": 0.01185559,
      "auxiliary_loss_mlp": 0.01024124,
      "balance_loss_clip": 1.04427099,
      "balance_loss_mlp": 1.01604795,
      "epoch": 0.5716347020982384,
      "flos": 23330265189120.0,
      "grad_norm": 1.9482799727405924,
      "language_loss": 0.78337079,
      "learning_rate": 1.6351587570901178e-06,
      "loss": 0.80546761,
      "num_input_tokens_seen": 102659430,
      "step": 4754,
      "time_per_iteration": 2.685180187225342
    },
    {
      "auxiliary_loss_clip": 0.01183838,
      "auxiliary_loss_mlp": 0.0102653,
      "balance_loss_clip": 1.04832673,
      "balance_loss_mlp": 1.01899576,
      "epoch": 0.5717549449888776,
      "flos": 17009555806080.0,
      "grad_norm": 2.361252338188896,
      "language_loss": 0.75541031,
      "learning_rate": 1.634392883824534e-06,
      "loss": 0.77751398,
      "num_input_tokens_seen": 102671430,
      "step": 4755,
      "time_per_iteration": 2.6044628620147705
    },
    {
      "auxiliary_loss_clip": 0.01187608,
      "auxiliary_loss_mlp": 0.01024223,
      "balance_loss_clip": 1.04481316,
      "balance_loss_mlp": 1.01623642,
      "epoch": 0.5718751878795166,
      "flos": 35518130922240.0,
      "grad_norm": 2.7602206590125107,
      "language_loss": 0.67760217,
      "learning_rate": 1.6336270660212595e-06,
      "loss": 0.6997205,
      "num_input_tokens_seen": 102693025,
      "step": 4756,
      "time_per_iteration": 2.78238844871521
    },
    {
      "auxiliary_loss_clip": 0.01177743,
      "auxiliary_loss_mlp": 0.01027703,
      "balance_loss_clip": 1.05124402,
      "balance_loss_mlp": 1.01886344,
      "epoch": 0.5719954307701557,
      "flos": 38613989255040.0,
      "grad_norm": 2.0192185752111946,
      "language_loss": 0.65875804,
      "learning_rate": 1.6328613037964676e-06,
      "loss": 0.68081248,
      "num_input_tokens_seen": 102716090,
      "step": 4757,
      "time_per_iteration": 2.7659010887145996
    },
    {
      "auxiliary_loss_clip": 0.01178185,
      "auxiliary_loss_mlp": 0.01024668,
      "balance_loss_clip": 1.04996276,
      "balance_loss_mlp": 1.01659179,
      "epoch": 0.5721156736607949,
      "flos": 20631111638400.0,
      "grad_norm": 2.5750156307016883,
      "language_loss": 0.67924768,
      "learning_rate": 1.6320955972663241e-06,
      "loss": 0.70127618,
      "num_input_tokens_seen": 102735685,
      "step": 4758,
      "time_per_iteration": 2.6310129165649414
    },
    {
      "auxiliary_loss_clip": 0.01179242,
      "auxiliary_loss_mlp": 0.01027535,
      "balance_loss_clip": 1.04985058,
      "balance_loss_mlp": 1.01985192,
      "epoch": 0.5722359165514339,
      "flos": 37415076076800.0,
      "grad_norm": 1.821266289766432,
      "language_loss": 0.65751624,
      "learning_rate": 1.6313299465469857e-06,
      "loss": 0.67958397,
      "num_input_tokens_seen": 102758415,
      "step": 4759,
      "time_per_iteration": 3.631452798843384
    },
    {
      "auxiliary_loss_clip": 0.0117703,
      "auxiliary_loss_mlp": 0.01027955,
      "balance_loss_clip": 1.05103588,
      "balance_loss_mlp": 1.01938415,
      "epoch": 0.572356159442073,
      "flos": 21972877205760.0,
      "grad_norm": 3.081503299867889,
      "language_loss": 0.79693812,
      "learning_rate": 1.6305643517546014e-06,
      "loss": 0.81898797,
      "num_input_tokens_seen": 102773795,
      "step": 4760,
      "time_per_iteration": 2.629124402999878
    },
    {
      "auxiliary_loss_clip": 0.01175319,
      "auxiliary_loss_mlp": 0.01034201,
      "balance_loss_clip": 1.05230427,
      "balance_loss_mlp": 1.02657747,
      "epoch": 0.5724764023327121,
      "flos": 19135540033920.0,
      "grad_norm": 2.2173335246293364,
      "language_loss": 0.851156,
      "learning_rate": 1.629798813005311e-06,
      "loss": 0.8732512,
      "num_input_tokens_seen": 102793515,
      "step": 4761,
      "time_per_iteration": 2.625985860824585
    },
    {
      "auxiliary_loss_clip": 0.01188887,
      "auxiliary_loss_mlp": 0.01029702,
      "balance_loss_clip": 1.04822183,
      "balance_loss_mlp": 1.02196515,
      "epoch": 0.5725966452233512,
      "flos": 22819759142400.0,
      "grad_norm": 1.9625846099370525,
      "language_loss": 0.71417254,
      "learning_rate": 1.6290333304152473e-06,
      "loss": 0.73635846,
      "num_input_tokens_seen": 102813390,
      "step": 4762,
      "time_per_iteration": 2.722397804260254
    },
    {
      "auxiliary_loss_clip": 0.01178698,
      "auxiliary_loss_mlp": 0.01030581,
      "balance_loss_clip": 1.05169058,
      "balance_loss_mlp": 1.02244496,
      "epoch": 0.5727168881139902,
      "flos": 41496610498560.0,
      "grad_norm": 1.9164271675492035,
      "language_loss": 0.57081902,
      "learning_rate": 1.6282679041005314e-06,
      "loss": 0.59291178,
      "num_input_tokens_seen": 102838980,
      "step": 4763,
      "time_per_iteration": 3.7205231189727783
    },
    {
      "auxiliary_loss_clip": 0.01173603,
      "auxiliary_loss_mlp": 0.01020965,
      "balance_loss_clip": 1.04558372,
      "balance_loss_mlp": 1.01302576,
      "epoch": 0.5728371310046293,
      "flos": 14647675985280.0,
      "grad_norm": 6.317504605225245,
      "language_loss": 0.8780303,
      "learning_rate": 1.6275025341772789e-06,
      "loss": 0.89997602,
      "num_input_tokens_seen": 102855285,
      "step": 4764,
      "time_per_iteration": 3.5335967540740967
    },
    {
      "auxiliary_loss_clip": 0.01182572,
      "auxiliary_loss_mlp": 0.01027318,
      "balance_loss_clip": 1.04827547,
      "balance_loss_mlp": 1.01857996,
      "epoch": 0.5729573738952685,
      "flos": 21506613736320.0,
      "grad_norm": 2.184544903024251,
      "language_loss": 0.81708795,
      "learning_rate": 1.626737220761596e-06,
      "loss": 0.83918679,
      "num_input_tokens_seen": 102872750,
      "step": 4765,
      "time_per_iteration": 2.6502933502197266
    },
    {
      "auxiliary_loss_clip": 0.01172991,
      "auxiliary_loss_mlp": 0.01028719,
      "balance_loss_clip": 1.04890931,
      "balance_loss_mlp": 1.02094102,
      "epoch": 0.5730776167859075,
      "flos": 23621680229760.0,
      "grad_norm": 3.902781551356214,
      "language_loss": 0.78949255,
      "learning_rate": 1.62597196396958e-06,
      "loss": 0.81150961,
      "num_input_tokens_seen": 102890920,
      "step": 4766,
      "time_per_iteration": 2.5939950942993164
    },
    {
      "auxiliary_loss_clip": 0.01177839,
      "auxiliary_loss_mlp": 0.01023788,
      "balance_loss_clip": 1.05030322,
      "balance_loss_mlp": 1.01578879,
      "epoch": 0.5731978596765466,
      "flos": 25739224761600.0,
      "grad_norm": 2.7503670456388387,
      "language_loss": 0.8566798,
      "learning_rate": 1.6252067639173197e-06,
      "loss": 0.87869608,
      "num_input_tokens_seen": 102912830,
      "step": 4767,
      "time_per_iteration": 2.6460680961608887
    },
    {
      "auxiliary_loss_clip": 0.01179358,
      "auxiliary_loss_mlp": 0.01026847,
      "balance_loss_clip": 1.05036509,
      "balance_loss_mlp": 1.0188657,
      "epoch": 0.5733181025671857,
      "flos": 26359509749760.0,
      "grad_norm": 2.2365607861852745,
      "language_loss": 0.69929868,
      "learning_rate": 1.6244416207208956e-06,
      "loss": 0.72136074,
      "num_input_tokens_seen": 102933765,
      "step": 4768,
      "time_per_iteration": 2.638136863708496
    },
    {
      "auxiliary_loss_clip": 0.01185081,
      "auxiliary_loss_mlp": 0.01030453,
      "balance_loss_clip": 1.04857016,
      "balance_loss_mlp": 1.02265096,
      "epoch": 0.5734383454578248,
      "flos": 29423874833280.0,
      "grad_norm": 1.980166977603498,
      "language_loss": 0.73874354,
      "learning_rate": 1.6236765344963787e-06,
      "loss": 0.76089895,
      "num_input_tokens_seen": 102955025,
      "step": 4769,
      "time_per_iteration": 2.6965484619140625
    },
    {
      "auxiliary_loss_clip": 0.01181632,
      "auxiliary_loss_mlp": 0.01025673,
      "balance_loss_clip": 1.04804802,
      "balance_loss_mlp": 1.01769793,
      "epoch": 0.5735585883484638,
      "flos": 34969954487040.0,
      "grad_norm": 2.998437821566959,
      "language_loss": 0.69422168,
      "learning_rate": 1.6229115053598322e-06,
      "loss": 0.71629477,
      "num_input_tokens_seen": 102976780,
      "step": 4770,
      "time_per_iteration": 2.7647480964660645
    },
    {
      "auxiliary_loss_clip": 0.01181634,
      "auxiliary_loss_mlp": 0.01030379,
      "balance_loss_clip": 1.05207741,
      "balance_loss_mlp": 1.02202857,
      "epoch": 0.573678831239103,
      "flos": 18770759464320.0,
      "grad_norm": 2.0162902927398836,
      "language_loss": 0.72038013,
      "learning_rate": 1.6221465334273108e-06,
      "loss": 0.74250025,
      "num_input_tokens_seen": 102995990,
      "step": 4771,
      "time_per_iteration": 2.570596218109131
    },
    {
      "auxiliary_loss_clip": 0.01189863,
      "auxiliary_loss_mlp": 0.01024868,
      "balance_loss_clip": 1.04794812,
      "balance_loss_mlp": 1.01646364,
      "epoch": 0.5737990741297421,
      "flos": 25702883176320.0,
      "grad_norm": 2.1887095272884296,
      "language_loss": 0.6165697,
      "learning_rate": 1.6213816188148593e-06,
      "loss": 0.63871706,
      "num_input_tokens_seen": 103014695,
      "step": 4772,
      "time_per_iteration": 3.6169111728668213
    },
    {
      "auxiliary_loss_clip": 0.01172961,
      "auxiliary_loss_mlp": 0.01024544,
      "balance_loss_clip": 1.05019021,
      "balance_loss_mlp": 1.01653337,
      "epoch": 0.5739193170203811,
      "flos": 27269234530560.0,
      "grad_norm": 1.7250885230222246,
      "language_loss": 0.7742936,
      "learning_rate": 1.6206167616385162e-06,
      "loss": 0.79626864,
      "num_input_tokens_seen": 103035760,
      "step": 4773,
      "time_per_iteration": 2.657231569290161
    },
    {
      "auxiliary_loss_clip": 0.01189414,
      "auxiliary_loss_mlp": 0.01025946,
      "balance_loss_clip": 1.05050325,
      "balance_loss_mlp": 1.01727891,
      "epoch": 0.5740395599110203,
      "flos": 12239721993600.0,
      "grad_norm": 2.140403521461081,
      "language_loss": 0.73557603,
      "learning_rate": 1.6198519620143078e-06,
      "loss": 0.75772965,
      "num_input_tokens_seen": 103052915,
      "step": 4774,
      "time_per_iteration": 2.7390646934509277
    },
    {
      "auxiliary_loss_clip": 0.01190389,
      "auxiliary_loss_mlp": 0.01030775,
      "balance_loss_clip": 1.0507344,
      "balance_loss_mlp": 1.02307439,
      "epoch": 0.5741598028016593,
      "flos": 25921399564800.0,
      "grad_norm": 2.9698244882061786,
      "language_loss": 0.78269112,
      "learning_rate": 1.6190872200582546e-06,
      "loss": 0.80490279,
      "num_input_tokens_seen": 103074655,
      "step": 4775,
      "time_per_iteration": 2.724738836288452
    },
    {
      "auxiliary_loss_clip": 0.01176541,
      "auxiliary_loss_mlp": 0.01208071,
      "balance_loss_clip": 1.04756832,
      "balance_loss_mlp": 1.0005517,
      "epoch": 0.5742800456922984,
      "flos": 19244133826560.0,
      "grad_norm": 2.535948195110204,
      "language_loss": 0.78117883,
      "learning_rate": 1.6183225358863676e-06,
      "loss": 0.80502486,
      "num_input_tokens_seen": 103091550,
      "step": 4776,
      "time_per_iteration": 2.5933539867401123
    },
    {
      "auxiliary_loss_clip": 0.01172851,
      "auxiliary_loss_mlp": 0.01026721,
      "balance_loss_clip": 1.04630792,
      "balance_loss_mlp": 1.01816821,
      "epoch": 0.5744002885829376,
      "flos": 30920487932160.0,
      "grad_norm": 2.6496046784187817,
      "language_loss": 0.72015333,
      "learning_rate": 1.617557909614648e-06,
      "loss": 0.74214911,
      "num_input_tokens_seen": 103110985,
      "step": 4777,
      "time_per_iteration": 2.7571475505828857
    },
    {
      "auxiliary_loss_clip": 0.0118151,
      "auxiliary_loss_mlp": 0.01023151,
      "balance_loss_clip": 1.04520059,
      "balance_loss_mlp": 1.01553917,
      "epoch": 0.5745205314735766,
      "flos": 23840017050240.0,
      "grad_norm": 4.204506753696148,
      "language_loss": 0.85922229,
      "learning_rate": 1.6167933413590899e-06,
      "loss": 0.88126886,
      "num_input_tokens_seen": 103129890,
      "step": 4778,
      "time_per_iteration": 2.683833599090576
    },
    {
      "auxiliary_loss_clip": 0.01178299,
      "auxiliary_loss_mlp": 0.01028463,
      "balance_loss_clip": 1.04980624,
      "balance_loss_mlp": 1.02024913,
      "epoch": 0.5746407743642157,
      "flos": 12311902373760.0,
      "grad_norm": 3.7219678079231215,
      "language_loss": 0.90603572,
      "learning_rate": 1.6160288312356773e-06,
      "loss": 0.92810333,
      "num_input_tokens_seen": 103147020,
      "step": 4779,
      "time_per_iteration": 2.6379446983337402
    },
    {
      "auxiliary_loss_clip": 0.01182859,
      "auxiliary_loss_mlp": 0.01028126,
      "balance_loss_clip": 1.04991126,
      "balance_loss_mlp": 1.01975203,
      "epoch": 0.5747610172548548,
      "flos": 24133658734080.0,
      "grad_norm": 1.8149516528184353,
      "language_loss": 0.81410658,
      "learning_rate": 1.6152643793603857e-06,
      "loss": 0.83621645,
      "num_input_tokens_seen": 103167370,
      "step": 4780,
      "time_per_iteration": 2.6385138034820557
    },
    {
      "auxiliary_loss_clip": 0.01177623,
      "auxiliary_loss_mlp": 0.0102584,
      "balance_loss_clip": 1.05296612,
      "balance_loss_mlp": 1.01771021,
      "epoch": 0.5748812601454939,
      "flos": 25408451393280.0,
      "grad_norm": 1.7558419449365914,
      "language_loss": 0.87716198,
      "learning_rate": 1.6144999858491815e-06,
      "loss": 0.89919662,
      "num_input_tokens_seen": 103186000,
      "step": 4781,
      "time_per_iteration": 2.6503565311431885
    },
    {
      "auxiliary_loss_clip": 0.01187202,
      "auxiliary_loss_mlp": 0.01025718,
      "balance_loss_clip": 1.04845285,
      "balance_loss_mlp": 1.01706386,
      "epoch": 0.575001503036133,
      "flos": 30624942827520.0,
      "grad_norm": 1.7179146661217743,
      "language_loss": 0.85718358,
      "learning_rate": 1.6137356508180232e-06,
      "loss": 0.87931275,
      "num_input_tokens_seen": 103207710,
      "step": 4782,
      "time_per_iteration": 2.6824657917022705
    },
    {
      "auxiliary_loss_clip": 0.01177855,
      "auxiliary_loss_mlp": 0.01207809,
      "balance_loss_clip": 1.0513978,
      "balance_loss_mlp": 1.00059772,
      "epoch": 0.5751217459267721,
      "flos": 21726566668800.0,
      "grad_norm": 1.7401674181499032,
      "language_loss": 0.81092948,
      "learning_rate": 1.6129713743828593e-06,
      "loss": 0.83478612,
      "num_input_tokens_seen": 103226720,
      "step": 4783,
      "time_per_iteration": 2.61895751953125
    },
    {
      "auxiliary_loss_clip": 0.01181808,
      "auxiliary_loss_mlp": 0.01023573,
      "balance_loss_clip": 1.04640996,
      "balance_loss_mlp": 1.01583302,
      "epoch": 0.5752419888174112,
      "flos": 21651620941440.0,
      "grad_norm": 1.4835985641117717,
      "language_loss": 0.75452232,
      "learning_rate": 1.6122071566596306e-06,
      "loss": 0.77657616,
      "num_input_tokens_seen": 103246995,
      "step": 4784,
      "time_per_iteration": 2.607685089111328
    },
    {
      "auxiliary_loss_clip": 0.01182123,
      "auxiliary_loss_mlp": 0.01028565,
      "balance_loss_clip": 1.05072522,
      "balance_loss_mlp": 1.01995254,
      "epoch": 0.5753622317080502,
      "flos": 17775997234560.0,
      "grad_norm": 3.294775349334902,
      "language_loss": 0.83210504,
      "learning_rate": 1.6114429977642674e-06,
      "loss": 0.85421193,
      "num_input_tokens_seen": 103261500,
      "step": 4785,
      "time_per_iteration": 2.6404945850372314
    },
    {
      "auxiliary_loss_clip": 0.01182627,
      "auxiliary_loss_mlp": 0.01026565,
      "balance_loss_clip": 1.05401659,
      "balance_loss_mlp": 1.01902521,
      "epoch": 0.5754824745986894,
      "flos": 19789616741760.0,
      "grad_norm": 1.8910966605039867,
      "language_loss": 0.7379418,
      "learning_rate": 1.6106788978126926e-06,
      "loss": 0.76003373,
      "num_input_tokens_seen": 103280475,
      "step": 4786,
      "time_per_iteration": 3.5320651531219482
    },
    {
      "auxiliary_loss_clip": 0.01182274,
      "auxiliary_loss_mlp": 0.01028218,
      "balance_loss_clip": 1.04238534,
      "balance_loss_mlp": 1.01954579,
      "epoch": 0.5756027174893285,
      "flos": 30985665160320.0,
      "grad_norm": 2.7866086461115556,
      "language_loss": 0.79174799,
      "learning_rate": 1.6099148569208196e-06,
      "loss": 0.81385291,
      "num_input_tokens_seen": 103297695,
      "step": 4787,
      "time_per_iteration": 2.7568423748016357
    },
    {
      "auxiliary_loss_clip": 0.01181561,
      "auxiliary_loss_mlp": 0.01029362,
      "balance_loss_clip": 1.05004048,
      "balance_loss_mlp": 1.020576,
      "epoch": 0.5757229603799675,
      "flos": 28546864364160.0,
      "grad_norm": 1.7365350157889647,
      "language_loss": 0.63060778,
      "learning_rate": 1.6091508752045523e-06,
      "loss": 0.65271699,
      "num_input_tokens_seen": 103318575,
      "step": 4788,
      "time_per_iteration": 2.685131311416626
    },
    {
      "auxiliary_loss_clip": 0.01175206,
      "auxiliary_loss_mlp": 0.01026813,
      "balance_loss_clip": 1.04337168,
      "balance_loss_mlp": 1.01902294,
      "epoch": 0.5758432032706067,
      "flos": 22999024944000.0,
      "grad_norm": 1.6691633497644822,
      "language_loss": 0.86630869,
      "learning_rate": 1.608386952779787e-06,
      "loss": 0.88832885,
      "num_input_tokens_seen": 103337945,
      "step": 4789,
      "time_per_iteration": 2.6545486450195312
    },
    {
      "auxiliary_loss_clip": 0.01187773,
      "auxiliary_loss_mlp": 0.01026378,
      "balance_loss_clip": 1.0504303,
      "balance_loss_mlp": 1.01881421,
      "epoch": 0.5759634461612457,
      "flos": 25739727552000.0,
      "grad_norm": 1.6712457212020437,
      "language_loss": 0.74553216,
      "learning_rate": 1.6076230897624098e-06,
      "loss": 0.76767373,
      "num_input_tokens_seen": 103360150,
      "step": 4790,
      "time_per_iteration": 3.7273473739624023
    },
    {
      "auxiliary_loss_clip": 0.01181227,
      "auxiliary_loss_mlp": 0.01026755,
      "balance_loss_clip": 1.04870248,
      "balance_loss_mlp": 1.01813006,
      "epoch": 0.5760836890518848,
      "flos": 30591761639040.0,
      "grad_norm": 2.1813443489397653,
      "language_loss": 0.77527374,
      "learning_rate": 1.6068592862682974e-06,
      "loss": 0.79735363,
      "num_input_tokens_seen": 103378305,
      "step": 4791,
      "time_per_iteration": 3.6101417541503906
    },
    {
      "auxiliary_loss_clip": 0.01184105,
      "auxiliary_loss_mlp": 0.01027455,
      "balance_loss_clip": 1.04996181,
      "balance_loss_mlp": 1.01956367,
      "epoch": 0.576203931942524,
      "flos": 36538963447680.0,
      "grad_norm": 6.75689753162339,
      "language_loss": 0.73682272,
      "learning_rate": 1.6060955424133187e-06,
      "loss": 0.75893837,
      "num_input_tokens_seen": 103399230,
      "step": 4792,
      "time_per_iteration": 2.725728988647461
    },
    {
      "auxiliary_loss_clip": 0.01181303,
      "auxiliary_loss_mlp": 0.01026978,
      "balance_loss_clip": 1.05311728,
      "balance_loss_mlp": 1.01832342,
      "epoch": 0.576324174833163,
      "flos": 25516937445120.0,
      "grad_norm": 2.343242352709758,
      "language_loss": 0.89504218,
      "learning_rate": 1.6053318583133332e-06,
      "loss": 0.91712505,
      "num_input_tokens_seen": 103420100,
      "step": 4793,
      "time_per_iteration": 2.700263023376465
    },
    {
      "auxiliary_loss_clip": 0.01179601,
      "auxiliary_loss_mlp": 0.01028828,
      "balance_loss_clip": 1.05109441,
      "balance_loss_mlp": 1.02066827,
      "epoch": 0.5764444177238021,
      "flos": 25119262995840.0,
      "grad_norm": 2.185810724256701,
      "language_loss": 0.75488961,
      "learning_rate": 1.6045682340841907e-06,
      "loss": 0.7769739,
      "num_input_tokens_seen": 103439025,
      "step": 4794,
      "time_per_iteration": 2.637204647064209
    },
    {
      "auxiliary_loss_clip": 0.01094031,
      "auxiliary_loss_mlp": 0.01199336,
      "balance_loss_clip": 1.01609719,
      "balance_loss_mlp": 0.99994451,
      "epoch": 0.5765646606144411,
      "flos": 62212687758720.0,
      "grad_norm": 0.7561491912988677,
      "language_loss": 0.57970345,
      "learning_rate": 1.6038046698417336e-06,
      "loss": 0.60263717,
      "num_input_tokens_seen": 103499920,
      "step": 4795,
      "time_per_iteration": 3.1754722595214844
    },
    {
      "auxiliary_loss_clip": 0.01178434,
      "auxiliary_loss_mlp": 0.01025334,
      "balance_loss_clip": 1.04893172,
      "balance_loss_mlp": 1.01732361,
      "epoch": 0.5766849035050803,
      "flos": 25118760205440.0,
      "grad_norm": 1.9453133567089231,
      "language_loss": 0.68957973,
      "learning_rate": 1.6030411657017919e-06,
      "loss": 0.71161741,
      "num_input_tokens_seen": 103519575,
      "step": 4796,
      "time_per_iteration": 2.587448835372925
    },
    {
      "auxiliary_loss_clip": 0.0117084,
      "auxiliary_loss_mlp": 0.01023311,
      "balance_loss_clip": 1.04851532,
      "balance_loss_mlp": 1.01560962,
      "epoch": 0.5768051463957193,
      "flos": 15991093578240.0,
      "grad_norm": 1.9094986924770871,
      "language_loss": 0.8470639,
      "learning_rate": 1.6022777217801903e-06,
      "loss": 0.86900544,
      "num_input_tokens_seen": 103536530,
      "step": 4797,
      "time_per_iteration": 2.571092367172241
    },
    {
      "auxiliary_loss_clip": 0.01184491,
      "auxiliary_loss_mlp": 0.01024372,
      "balance_loss_clip": 1.04960895,
      "balance_loss_mlp": 1.01620579,
      "epoch": 0.5769253892863584,
      "flos": 22163635359360.0,
      "grad_norm": 2.1292511538741397,
      "language_loss": 0.73834622,
      "learning_rate": 1.601514338192742e-06,
      "loss": 0.76043487,
      "num_input_tokens_seen": 103556460,
      "step": 4798,
      "time_per_iteration": 2.6258292198181152
    },
    {
      "auxiliary_loss_clip": 0.01172183,
      "auxiliary_loss_mlp": 0.01021447,
      "balance_loss_clip": 1.04989898,
      "balance_loss_mlp": 1.01408243,
      "epoch": 0.5770456321769976,
      "flos": 22856388036480.0,
      "grad_norm": 3.3395829221723896,
      "language_loss": 0.71428138,
      "learning_rate": 1.6007510150552514e-06,
      "loss": 0.73621762,
      "num_input_tokens_seen": 103574520,
      "step": 4799,
      "time_per_iteration": 3.459456443786621
    },
    {
      "auxiliary_loss_clip": 0.01184106,
      "auxiliary_loss_mlp": 0.010297,
      "balance_loss_clip": 1.04978323,
      "balance_loss_mlp": 1.02086031,
      "epoch": 0.5771658750676366,
      "flos": 46353672489600.0,
      "grad_norm": 1.7002593076343586,
      "language_loss": 0.6240446,
      "learning_rate": 1.599987752483515e-06,
      "loss": 0.64618266,
      "num_input_tokens_seen": 103598965,
      "step": 4800,
      "time_per_iteration": 2.819923162460327
    },
    {
      "auxiliary_loss_clip": 0.01178464,
      "auxiliary_loss_mlp": 0.0102873,
      "balance_loss_clip": 1.04450226,
      "balance_loss_mlp": 1.02077889,
      "epoch": 0.5772861179582757,
      "flos": 22159972172160.0,
      "grad_norm": 1.7709178062991695,
      "language_loss": 0.68185937,
      "learning_rate": 1.5992245505933184e-06,
      "loss": 0.70393133,
      "num_input_tokens_seen": 103618665,
      "step": 4801,
      "time_per_iteration": 2.6176719665527344
    },
    {
      "auxiliary_loss_clip": 0.01179712,
      "auxiliary_loss_mlp": 0.01027161,
      "balance_loss_clip": 1.05302882,
      "balance_loss_mlp": 1.01947761,
      "epoch": 0.5774063608489148,
      "flos": 31248926916480.0,
      "grad_norm": 3.161464799471374,
      "language_loss": 0.7158584,
      "learning_rate": 1.5984614095004388e-06,
      "loss": 0.73792714,
      "num_input_tokens_seen": 103639800,
      "step": 4802,
      "time_per_iteration": 2.648470163345337
    },
    {
      "auxiliary_loss_clip": 0.0117359,
      "auxiliary_loss_mlp": 0.01029213,
      "balance_loss_clip": 1.0491426,
      "balance_loss_mlp": 1.02133286,
      "epoch": 0.5775266037395539,
      "flos": 22527123039360.0,
      "grad_norm": 2.002257716939322,
      "language_loss": 0.81164336,
      "learning_rate": 1.5976983293206438e-06,
      "loss": 0.83367133,
      "num_input_tokens_seen": 103655605,
      "step": 4803,
      "time_per_iteration": 2.589474678039551
    },
    {
      "auxiliary_loss_clip": 0.01176971,
      "auxiliary_loss_mlp": 0.01022437,
      "balance_loss_clip": 1.04495883,
      "balance_loss_mlp": 1.01450992,
      "epoch": 0.577646846630193,
      "flos": 21068790860160.0,
      "grad_norm": 2.3741151538295515,
      "language_loss": 0.7152189,
      "learning_rate": 1.5969353101696928e-06,
      "loss": 0.73721296,
      "num_input_tokens_seen": 103674045,
      "step": 4804,
      "time_per_iteration": 2.627686023712158
    },
    {
      "auxiliary_loss_clip": 0.01177564,
      "auxiliary_loss_mlp": 0.01027937,
      "balance_loss_clip": 1.04847193,
      "balance_loss_mlp": 1.02043271,
      "epoch": 0.5777670895208321,
      "flos": 29714284293120.0,
      "grad_norm": 1.9062766785684264,
      "language_loss": 0.79777992,
      "learning_rate": 1.5961723521633341e-06,
      "loss": 0.81983495,
      "num_input_tokens_seen": 103695285,
      "step": 4805,
      "time_per_iteration": 2.6286489963531494
    },
    {
      "auxiliary_loss_clip": 0.01178486,
      "auxiliary_loss_mlp": 0.0102911,
      "balance_loss_clip": 1.04696298,
      "balance_loss_mlp": 1.02077138,
      "epoch": 0.5778873324114712,
      "flos": 19500428344320.0,
      "grad_norm": 2.298976299626059,
      "language_loss": 0.90825719,
      "learning_rate": 1.5954094554173097e-06,
      "loss": 0.93033314,
      "num_input_tokens_seen": 103713275,
      "step": 4806,
      "time_per_iteration": 2.6528890132904053
    },
    {
      "auxiliary_loss_clip": 0.01187004,
      "auxiliary_loss_mlp": 0.0102785,
      "balance_loss_clip": 1.05059671,
      "balance_loss_mlp": 1.02021456,
      "epoch": 0.5780075753021102,
      "flos": 14136846716160.0,
      "grad_norm": 2.0095513708743513,
      "language_loss": 0.79102206,
      "learning_rate": 1.5946466200473482e-06,
      "loss": 0.81317067,
      "num_input_tokens_seen": 103731185,
      "step": 4807,
      "time_per_iteration": 2.5882794857025146
    },
    {
      "auxiliary_loss_clip": 0.01185709,
      "auxiliary_loss_mlp": 0.01029696,
      "balance_loss_clip": 1.04817033,
      "balance_loss_mlp": 1.02188754,
      "epoch": 0.5781278181927494,
      "flos": 15262178883840.0,
      "grad_norm": 4.721808120203469,
      "language_loss": 0.83619833,
      "learning_rate": 1.5938838461691723e-06,
      "loss": 0.8583523,
      "num_input_tokens_seen": 103748095,
      "step": 4808,
      "time_per_iteration": 2.639188051223755
    },
    {
      "auxiliary_loss_clip": 0.01179887,
      "auxiliary_loss_mlp": 0.01028925,
      "balance_loss_clip": 1.05434382,
      "balance_loss_mlp": 1.02104509,
      "epoch": 0.5782480610833884,
      "flos": 16726831856640.0,
      "grad_norm": 2.797406503113993,
      "language_loss": 0.8308531,
      "learning_rate": 1.593121133898494e-06,
      "loss": 0.85294116,
      "num_input_tokens_seen": 103765300,
      "step": 4809,
      "time_per_iteration": 2.544121503829956
    },
    {
      "auxiliary_loss_clip": 0.01186623,
      "auxiliary_loss_mlp": 0.01027596,
      "balance_loss_clip": 1.05210221,
      "balance_loss_mlp": 1.0196389,
      "epoch": 0.5783683039740275,
      "flos": 25482140144640.0,
      "grad_norm": 2.25159674629732,
      "language_loss": 0.79311907,
      "learning_rate": 1.592358483351016e-06,
      "loss": 0.81526124,
      "num_input_tokens_seen": 103785475,
      "step": 4810,
      "time_per_iteration": 2.6589949131011963
    },
    {
      "auxiliary_loss_clip": 0.01174898,
      "auxiliary_loss_mlp": 0.01024595,
      "balance_loss_clip": 1.04819798,
      "balance_loss_mlp": 1.01695967,
      "epoch": 0.5784885468646667,
      "flos": 18405835240320.0,
      "grad_norm": 2.2607526306220564,
      "language_loss": 0.72137368,
      "learning_rate": 1.5915958946424326e-06,
      "loss": 0.74336863,
      "num_input_tokens_seen": 103804160,
      "step": 4811,
      "time_per_iteration": 2.5636963844299316
    },
    {
      "auxiliary_loss_clip": 0.01183692,
      "auxiliary_loss_mlp": 0.01208456,
      "balance_loss_clip": 1.04661191,
      "balance_loss_mlp": 1.00063181,
      "epoch": 0.5786087897553057,
      "flos": 46100717936640.0,
      "grad_norm": 2.0460550719473427,
      "language_loss": 0.7424705,
      "learning_rate": 1.5908333678884271e-06,
      "loss": 0.76639199,
      "num_input_tokens_seen": 103830580,
      "step": 4812,
      "time_per_iteration": 2.8412435054779053
    },
    {
      "auxiliary_loss_clip": 0.01177495,
      "auxiliary_loss_mlp": 0.01030163,
      "balance_loss_clip": 1.05016923,
      "balance_loss_mlp": 1.02236652,
      "epoch": 0.5787290326459448,
      "flos": 12385950261120.0,
      "grad_norm": 2.0016919537594893,
      "language_loss": 0.73651516,
      "learning_rate": 1.5900709032046743e-06,
      "loss": 0.75859177,
      "num_input_tokens_seen": 103848655,
      "step": 4813,
      "time_per_iteration": 3.473574638366699
    },
    {
      "auxiliary_loss_clip": 0.01178898,
      "auxiliary_loss_mlp": 0.01026327,
      "balance_loss_clip": 1.05086851,
      "balance_loss_mlp": 1.01825953,
      "epoch": 0.5788492755365839,
      "flos": 23290332243840.0,
      "grad_norm": 2.218169694620096,
      "language_loss": 0.782076,
      "learning_rate": 1.5893085007068391e-06,
      "loss": 0.80412817,
      "num_input_tokens_seen": 103866215,
      "step": 4814,
      "time_per_iteration": 2.573410749435425
    },
    {
      "auxiliary_loss_clip": 0.01169532,
      "auxiliary_loss_mlp": 0.01029481,
      "balance_loss_clip": 1.0449326,
      "balance_loss_mlp": 1.02092814,
      "epoch": 0.578969518427223,
      "flos": 24061047390720.0,
      "grad_norm": 2.3625819530513863,
      "language_loss": 0.71193779,
      "learning_rate": 1.5885461605105786e-06,
      "loss": 0.73392797,
      "num_input_tokens_seen": 103887815,
      "step": 4815,
      "time_per_iteration": 2.6260623931884766
    },
    {
      "auxiliary_loss_clip": 0.01181787,
      "auxiliary_loss_mlp": 0.0102555,
      "balance_loss_clip": 1.04855871,
      "balance_loss_mlp": 1.01750934,
      "epoch": 0.579089761317862,
      "flos": 21871825269120.0,
      "grad_norm": 2.1205081822459437,
      "language_loss": 0.76790863,
      "learning_rate": 1.5877838827315375e-06,
      "loss": 0.78998196,
      "num_input_tokens_seen": 103906360,
      "step": 4816,
      "time_per_iteration": 3.490710973739624
    },
    {
      "auxiliary_loss_clip": 0.01178375,
      "auxiliary_loss_mlp": 0.01029472,
      "balance_loss_clip": 1.05408907,
      "balance_loss_mlp": 1.02152348,
      "epoch": 0.5792100042085012,
      "flos": 22929681738240.0,
      "grad_norm": 2.224119996992015,
      "language_loss": 0.701285,
      "learning_rate": 1.587021667485355e-06,
      "loss": 0.7233634,
      "num_input_tokens_seen": 103925730,
      "step": 4817,
      "time_per_iteration": 2.6291863918304443
    },
    {
      "auxiliary_loss_clip": 0.01183049,
      "auxiliary_loss_mlp": 0.01019795,
      "balance_loss_clip": 1.04674292,
      "balance_loss_mlp": 1.0118978,
      "epoch": 0.5793302470991403,
      "flos": 21470056669440.0,
      "grad_norm": 1.6820011467100382,
      "language_loss": 0.78386283,
      "learning_rate": 1.5862595148876559e-06,
      "loss": 0.80589128,
      "num_input_tokens_seen": 103945835,
      "step": 4818,
      "time_per_iteration": 3.6492421627044678
    },
    {
      "auxiliary_loss_clip": 0.01187806,
      "auxiliary_loss_mlp": 0.01026983,
      "balance_loss_clip": 1.04504824,
      "balance_loss_mlp": 1.01904941,
      "epoch": 0.5794504899897793,
      "flos": 12711013367040.0,
      "grad_norm": 2.3018125313239155,
      "language_loss": 0.76763773,
      "learning_rate": 1.58549742505406e-06,
      "loss": 0.78978562,
      "num_input_tokens_seen": 103960580,
      "step": 4819,
      "time_per_iteration": 2.57560658454895
    },
    {
      "auxiliary_loss_clip": 0.01177076,
      "auxiliary_loss_mlp": 0.01027238,
      "balance_loss_clip": 1.05129814,
      "balance_loss_mlp": 1.01958489,
      "epoch": 0.5795707328804185,
      "flos": 14867054300160.0,
      "grad_norm": 2.274408715228284,
      "language_loss": 0.75962037,
      "learning_rate": 1.5847353981001747e-06,
      "loss": 0.78166348,
      "num_input_tokens_seen": 103977760,
      "step": 4820,
      "time_per_iteration": 2.6210124492645264
    },
    {
      "auxiliary_loss_clip": 0.01174601,
      "auxiliary_loss_mlp": 0.01030659,
      "balance_loss_clip": 1.04545581,
      "balance_loss_mlp": 1.02243364,
      "epoch": 0.5796909757710575,
      "flos": 36430046432640.0,
      "grad_norm": 1.729887821714172,
      "language_loss": 0.70193058,
      "learning_rate": 1.5839734341415993e-06,
      "loss": 0.72398317,
      "num_input_tokens_seen": 103999960,
      "step": 4821,
      "time_per_iteration": 2.7386534214019775
    },
    {
      "auxiliary_loss_clip": 0.01173287,
      "auxiliary_loss_mlp": 0.01024402,
      "balance_loss_clip": 1.05295849,
      "balance_loss_mlp": 1.01682007,
      "epoch": 0.5798112186616966,
      "flos": 23039891642880.0,
      "grad_norm": 1.9353193230033048,
      "language_loss": 0.76480418,
      "learning_rate": 1.5832115332939238e-06,
      "loss": 0.78678107,
      "num_input_tokens_seen": 104018400,
      "step": 4822,
      "time_per_iteration": 2.621365547180176
    },
    {
      "auxiliary_loss_clip": 0.01180901,
      "auxiliary_loss_mlp": 0.01027701,
      "balance_loss_clip": 1.05168056,
      "balance_loss_mlp": 1.01940382,
      "epoch": 0.5799314615523358,
      "flos": 16652604401280.0,
      "grad_norm": 1.9444162159382836,
      "language_loss": 0.74984562,
      "learning_rate": 1.5824496956727272e-06,
      "loss": 0.77193165,
      "num_input_tokens_seen": 104035605,
      "step": 4823,
      "time_per_iteration": 2.594928026199341
    },
    {
      "auxiliary_loss_clip": 0.01182664,
      "auxiliary_loss_mlp": 0.01025995,
      "balance_loss_clip": 1.04953611,
      "balance_loss_mlp": 1.01848531,
      "epoch": 0.5800517044429748,
      "flos": 20485673470080.0,
      "grad_norm": 1.8739107826597623,
      "language_loss": 0.73422295,
      "learning_rate": 1.5816879213935797e-06,
      "loss": 0.75630957,
      "num_input_tokens_seen": 104054415,
      "step": 4824,
      "time_per_iteration": 2.671555995941162
    },
    {
      "auxiliary_loss_clip": 0.0117582,
      "auxiliary_loss_mlp": 0.01026851,
      "balance_loss_clip": 1.0506618,
      "balance_loss_mlp": 1.01956427,
      "epoch": 0.5801719473336139,
      "flos": 31538258968320.0,
      "grad_norm": 2.6118464118687648,
      "language_loss": 0.79983503,
      "learning_rate": 1.5809262105720416e-06,
      "loss": 0.82186174,
      "num_input_tokens_seen": 104075455,
      "step": 4825,
      "time_per_iteration": 3.681777000427246
    },
    {
      "auxiliary_loss_clip": 0.01173577,
      "auxiliary_loss_mlp": 0.01027816,
      "balance_loss_clip": 1.05154324,
      "balance_loss_mlp": 1.02018702,
      "epoch": 0.580292190224253,
      "flos": 20375966355840.0,
      "grad_norm": 1.5939906843198153,
      "language_loss": 0.79648292,
      "learning_rate": 1.5801645633236644e-06,
      "loss": 0.81849682,
      "num_input_tokens_seen": 104096440,
      "step": 4826,
      "time_per_iteration": 2.6013803482055664
    },
    {
      "auxiliary_loss_clip": 0.01174073,
      "auxiliary_loss_mlp": 0.0102634,
      "balance_loss_clip": 1.04546523,
      "balance_loss_mlp": 1.01843929,
      "epoch": 0.5804124331148921,
      "flos": 26615373304320.0,
      "grad_norm": 2.094888337848483,
      "language_loss": 0.77121127,
      "learning_rate": 1.579402979763989e-06,
      "loss": 0.79321539,
      "num_input_tokens_seen": 104116775,
      "step": 4827,
      "time_per_iteration": 2.6775527000427246
    },
    {
      "auxiliary_loss_clip": 0.01201712,
      "auxiliary_loss_mlp": 0.01024048,
      "balance_loss_clip": 1.04898691,
      "balance_loss_mlp": 1.01631713,
      "epoch": 0.5805326760055312,
      "flos": 13478496289920.0,
      "grad_norm": 3.0973883630151375,
      "language_loss": 0.81413984,
      "learning_rate": 1.578641460008548e-06,
      "loss": 0.83639747,
      "num_input_tokens_seen": 104134510,
      "step": 4828,
      "time_per_iteration": 2.636963367462158
    },
    {
      "auxiliary_loss_clip": 0.01177551,
      "auxiliary_loss_mlp": 0.01026511,
      "balance_loss_clip": 1.05157113,
      "balance_loss_mlp": 1.01790369,
      "epoch": 0.5806529188961702,
      "flos": 12091374823680.0,
      "grad_norm": 2.0114747140411136,
      "language_loss": 0.68221116,
      "learning_rate": 1.5778800041728613e-06,
      "loss": 0.70425177,
      "num_input_tokens_seen": 104150800,
      "step": 4829,
      "time_per_iteration": 2.5788562297821045
    },
    {
      "auxiliary_loss_clip": 0.01173213,
      "auxiliary_loss_mlp": 0.0102324,
      "balance_loss_clip": 1.05020559,
      "balance_loss_mlp": 1.0153904,
      "epoch": 0.5807731617868094,
      "flos": 26214107495040.0,
      "grad_norm": 1.5394613380243811,
      "language_loss": 0.66112041,
      "learning_rate": 1.577118612372443e-06,
      "loss": 0.68308496,
      "num_input_tokens_seen": 104172640,
      "step": 4830,
      "time_per_iteration": 2.6508326530456543
    },
    {
      "auxiliary_loss_clip": 0.01175576,
      "auxiliary_loss_mlp": 0.0120841,
      "balance_loss_clip": 1.04561722,
      "balance_loss_mlp": 1.00068688,
      "epoch": 0.5808934046774484,
      "flos": 37962139190400.0,
      "grad_norm": 1.6331856494888028,
      "language_loss": 0.70190108,
      "learning_rate": 1.5763572847227943e-06,
      "loss": 0.72574091,
      "num_input_tokens_seen": 104193525,
      "step": 4831,
      "time_per_iteration": 2.755197048187256
    },
    {
      "auxiliary_loss_clip": 0.01175475,
      "auxiliary_loss_mlp": 0.01027157,
      "balance_loss_clip": 1.04794955,
      "balance_loss_mlp": 1.01951015,
      "epoch": 0.5810136475680875,
      "flos": 20485853038080.0,
      "grad_norm": 4.318109848890445,
      "language_loss": 0.81638861,
      "learning_rate": 1.5755960213394091e-06,
      "loss": 0.83841497,
      "num_input_tokens_seen": 104210625,
      "step": 4832,
      "time_per_iteration": 2.5854921340942383
    },
    {
      "auxiliary_loss_clip": 0.01185536,
      "auxiliary_loss_mlp": 0.01026821,
      "balance_loss_clip": 1.04645288,
      "balance_loss_mlp": 1.01911378,
      "epoch": 0.5811338904587267,
      "flos": 17530153574400.0,
      "grad_norm": 2.182712348024263,
      "language_loss": 0.78232193,
      "learning_rate": 1.5748348223377703e-06,
      "loss": 0.80444551,
      "num_input_tokens_seen": 104228180,
      "step": 4833,
      "time_per_iteration": 2.654042959213257
    },
    {
      "auxiliary_loss_clip": 0.01176331,
      "auxiliary_loss_mlp": 0.01026467,
      "balance_loss_clip": 1.04871392,
      "balance_loss_mlp": 1.01915598,
      "epoch": 0.5812541333493657,
      "flos": 19458017360640.0,
      "grad_norm": 1.5850632171355719,
      "language_loss": 0.78036457,
      "learning_rate": 1.5740736878333507e-06,
      "loss": 0.80239248,
      "num_input_tokens_seen": 104246020,
      "step": 4834,
      "time_per_iteration": 2.5682778358459473
    },
    {
      "auxiliary_loss_clip": 0.01185889,
      "auxiliary_loss_mlp": 0.0102344,
      "balance_loss_clip": 1.04941869,
      "balance_loss_mlp": 1.01524472,
      "epoch": 0.5813743762400048,
      "flos": 20594949621120.0,
      "grad_norm": 2.3826544562937872,
      "language_loss": 0.77866101,
      "learning_rate": 1.5733126179416143e-06,
      "loss": 0.80075431,
      "num_input_tokens_seen": 104260505,
      "step": 4835,
      "time_per_iteration": 2.6658549308776855
    },
    {
      "auxiliary_loss_clip": 0.0117726,
      "auxiliary_loss_mlp": 0.01026055,
      "balance_loss_clip": 1.04969883,
      "balance_loss_mlp": 1.01822329,
      "epoch": 0.5814946191306439,
      "flos": 33178227246720.0,
      "grad_norm": 1.9575951895582449,
      "language_loss": 0.72711027,
      "learning_rate": 1.5725516127780137e-06,
      "loss": 0.74914348,
      "num_input_tokens_seen": 104282640,
      "step": 4836,
      "time_per_iteration": 2.6821486949920654
    },
    {
      "auxiliary_loss_clip": 0.01185079,
      "auxiliary_loss_mlp": 0.01031822,
      "balance_loss_clip": 1.05012441,
      "balance_loss_mlp": 1.02311969,
      "epoch": 0.581614862021283,
      "flos": 16143283503360.0,
      "grad_norm": 2.346906984904711,
      "language_loss": 0.88463974,
      "learning_rate": 1.5717906724579943e-06,
      "loss": 0.90680873,
      "num_input_tokens_seen": 104299700,
      "step": 4837,
      "time_per_iteration": 2.6779885292053223
    },
    {
      "auxiliary_loss_clip": 0.01192837,
      "auxiliary_loss_mlp": 0.01023027,
      "balance_loss_clip": 1.04962015,
      "balance_loss_mlp": 1.015481,
      "epoch": 0.581735104911922,
      "flos": 33802642298880.0,
      "grad_norm": 1.8783111525479796,
      "language_loss": 0.68506789,
      "learning_rate": 1.571029797096989e-06,
      "loss": 0.70722651,
      "num_input_tokens_seen": 104320805,
      "step": 4838,
      "time_per_iteration": 2.8071446418762207
    },
    {
      "auxiliary_loss_clip": 0.01173247,
      "auxiliary_loss_mlp": 0.0102902,
      "balance_loss_clip": 1.05044615,
      "balance_loss_mlp": 1.02115214,
      "epoch": 0.5818553478025612,
      "flos": 23331163029120.0,
      "grad_norm": 1.831795842234832,
      "language_loss": 0.79005706,
      "learning_rate": 1.570268986810423e-06,
      "loss": 0.81207967,
      "num_input_tokens_seen": 104340700,
      "step": 4839,
      "time_per_iteration": 3.5380709171295166
    },
    {
      "auxiliary_loss_clip": 0.0117749,
      "auxiliary_loss_mlp": 0.01025185,
      "balance_loss_clip": 1.04796755,
      "balance_loss_mlp": 1.01797628,
      "epoch": 0.5819755906932003,
      "flos": 20996143603200.0,
      "grad_norm": 7.353049613713329,
      "language_loss": 0.7489652,
      "learning_rate": 1.5695082417137096e-06,
      "loss": 0.77099192,
      "num_input_tokens_seen": 104358575,
      "step": 4840,
      "time_per_iteration": 2.7321951389312744
    },
    {
      "auxiliary_loss_clip": 0.01176012,
      "auxiliary_loss_mlp": 0.01025881,
      "balance_loss_clip": 1.04437387,
      "balance_loss_mlp": 1.01819205,
      "epoch": 0.5820958335838393,
      "flos": 21431668008960.0,
      "grad_norm": 1.6936767292675372,
      "language_loss": 0.75534749,
      "learning_rate": 1.5687475619222539e-06,
      "loss": 0.77736634,
      "num_input_tokens_seen": 104378530,
      "step": 4841,
      "time_per_iteration": 2.70565128326416
    },
    {
      "auxiliary_loss_clip": 0.01176943,
      "auxiliary_loss_mlp": 0.01027409,
      "balance_loss_clip": 1.04703367,
      "balance_loss_mlp": 1.01936221,
      "epoch": 0.5822160764744785,
      "flos": 17967473660160.0,
      "grad_norm": 2.1321268109786238,
      "language_loss": 0.73306739,
      "learning_rate": 1.5679869475514496e-06,
      "loss": 0.75511092,
      "num_input_tokens_seen": 104395465,
      "step": 4842,
      "time_per_iteration": 2.650973320007324
    },
    {
      "auxiliary_loss_clip": 0.011798,
      "auxiliary_loss_mlp": 0.01028468,
      "balance_loss_clip": 1.04975963,
      "balance_loss_mlp": 1.01993918,
      "epoch": 0.5823363193651175,
      "flos": 23033858158080.0,
      "grad_norm": 2.546088346063947,
      "language_loss": 0.81180358,
      "learning_rate": 1.567226398716682e-06,
      "loss": 0.83388627,
      "num_input_tokens_seen": 104415380,
      "step": 4843,
      "time_per_iteration": 3.4881606101989746
    },
    {
      "auxiliary_loss_clip": 0.01188853,
      "auxiliary_loss_mlp": 0.01023355,
      "balance_loss_clip": 1.04987693,
      "balance_loss_mlp": 1.01475394,
      "epoch": 0.5824565622557566,
      "flos": 32891840110080.0,
      "grad_norm": 1.810838674852733,
      "language_loss": 0.61845136,
      "learning_rate": 1.566465915533326e-06,
      "loss": 0.64057338,
      "num_input_tokens_seen": 104437410,
      "step": 4844,
      "time_per_iteration": 2.7303009033203125
    },
    {
      "auxiliary_loss_clip": 0.01176979,
      "auxiliary_loss_mlp": 0.01025387,
      "balance_loss_clip": 1.05094934,
      "balance_loss_mlp": 1.01740026,
      "epoch": 0.5825768051463958,
      "flos": 22229674513920.0,
      "grad_norm": 1.855395973782592,
      "language_loss": 0.88236529,
      "learning_rate": 1.5657054981167458e-06,
      "loss": 0.9043889,
      "num_input_tokens_seen": 104456305,
      "step": 4845,
      "time_per_iteration": 3.543869972229004
    },
    {
      "auxiliary_loss_clip": 0.01173412,
      "auxiliary_loss_mlp": 0.01025524,
      "balance_loss_clip": 1.04890633,
      "balance_loss_mlp": 1.01838326,
      "epoch": 0.5826970480370348,
      "flos": 28001561016960.0,
      "grad_norm": 1.9156598760569536,
      "language_loss": 0.6780439,
      "learning_rate": 1.5649451465822965e-06,
      "loss": 0.70003325,
      "num_input_tokens_seen": 104477695,
      "step": 4846,
      "time_per_iteration": 2.636981964111328
    },
    {
      "auxiliary_loss_clip": 0.01182585,
      "auxiliary_loss_mlp": 0.01026883,
      "balance_loss_clip": 1.04666638,
      "balance_loss_mlp": 1.0190866,
      "epoch": 0.5828172909276739,
      "flos": 17858053854720.0,
      "grad_norm": 1.6904952412925058,
      "language_loss": 0.8391161,
      "learning_rate": 1.5641848610453218e-06,
      "loss": 0.86121082,
      "num_input_tokens_seen": 104496355,
      "step": 4847,
      "time_per_iteration": 2.6844828128814697
    },
    {
      "auxiliary_loss_clip": 0.01174447,
      "auxiliary_loss_mlp": 0.01024188,
      "balance_loss_clip": 1.05074453,
      "balance_loss_mlp": 1.01639223,
      "epoch": 0.582937533818313,
      "flos": 19865244827520.0,
      "grad_norm": 1.9998774681334979,
      "language_loss": 0.85943198,
      "learning_rate": 1.563424641621158e-06,
      "loss": 0.88141829,
      "num_input_tokens_seen": 104515535,
      "step": 4848,
      "time_per_iteration": 2.582242727279663
    },
    {
      "auxiliary_loss_clip": 0.01185817,
      "auxiliary_loss_mlp": 0.01025828,
      "balance_loss_clip": 1.04999137,
      "balance_loss_mlp": 1.01765633,
      "epoch": 0.5830577767089521,
      "flos": 26870734068480.0,
      "grad_norm": 2.271052254588184,
      "language_loss": 0.70165849,
      "learning_rate": 1.5626644884251282e-06,
      "loss": 0.72377497,
      "num_input_tokens_seen": 104535055,
      "step": 4849,
      "time_per_iteration": 2.625943899154663
    },
    {
      "auxiliary_loss_clip": 0.01174753,
      "auxiliary_loss_mlp": 0.01023082,
      "balance_loss_clip": 1.05145431,
      "balance_loss_mlp": 1.01583457,
      "epoch": 0.5831780195995911,
      "flos": 25298205575040.0,
      "grad_norm": 1.8942436937564229,
      "language_loss": 0.88305724,
      "learning_rate": 1.5619044015725488e-06,
      "loss": 0.90503561,
      "num_input_tokens_seen": 104554745,
      "step": 4850,
      "time_per_iteration": 2.5683743953704834
    },
    {
      "auxiliary_loss_clip": 0.01183126,
      "auxiliary_loss_mlp": 0.01030913,
      "balance_loss_clip": 1.05499506,
      "balance_loss_mlp": 1.0222044,
      "epoch": 0.5832982624902303,
      "flos": 14756988049920.0,
      "grad_norm": 2.368762467835275,
      "language_loss": 0.87114596,
      "learning_rate": 1.5611443811787224e-06,
      "loss": 0.89328635,
      "num_input_tokens_seen": 104568870,
      "step": 4851,
      "time_per_iteration": 2.566469192504883
    },
    {
      "auxiliary_loss_clip": 0.01175958,
      "auxiliary_loss_mlp": 0.01023219,
      "balance_loss_clip": 1.05017805,
      "balance_loss_mlp": 1.01563096,
      "epoch": 0.5834185053808694,
      "flos": 20444555376000.0,
      "grad_norm": 2.6940040808596883,
      "language_loss": 0.69596559,
      "learning_rate": 1.560384427358945e-06,
      "loss": 0.71795738,
      "num_input_tokens_seen": 104588415,
      "step": 4852,
      "time_per_iteration": 3.5505130290985107
    },
    {
      "auxiliary_loss_clip": 0.01174945,
      "auxiliary_loss_mlp": 0.01027263,
      "balance_loss_clip": 1.044554,
      "balance_loss_mlp": 1.01950288,
      "epoch": 0.5835387482715084,
      "flos": 27200394115200.0,
      "grad_norm": 1.3967716620561765,
      "language_loss": 0.73240781,
      "learning_rate": 1.5596245402284998e-06,
      "loss": 0.75442988,
      "num_input_tokens_seen": 104611940,
      "step": 4853,
      "time_per_iteration": 2.6926960945129395
    },
    {
      "auxiliary_loss_clip": 0.0118296,
      "auxiliary_loss_mlp": 0.01026416,
      "balance_loss_clip": 1.05388653,
      "balance_loss_mlp": 1.0184294,
      "epoch": 0.5836589911621476,
      "flos": 16654615562880.0,
      "grad_norm": 1.9347518907861763,
      "language_loss": 0.82175601,
      "learning_rate": 1.5588647199026619e-06,
      "loss": 0.84384972,
      "num_input_tokens_seen": 104629675,
      "step": 4854,
      "time_per_iteration": 2.5985474586486816
    },
    {
      "auxiliary_loss_clip": 0.01181791,
      "auxiliary_loss_mlp": 0.01028505,
      "balance_loss_clip": 1.05588436,
      "balance_loss_mlp": 1.02036309,
      "epoch": 0.5837792340527866,
      "flos": 20446817932800.0,
      "grad_norm": 2.0337312862628276,
      "language_loss": 0.87277591,
      "learning_rate": 1.5581049664966956e-06,
      "loss": 0.89487892,
      "num_input_tokens_seen": 104647435,
      "step": 4855,
      "time_per_iteration": 2.5368945598602295
    },
    {
      "auxiliary_loss_clip": 0.01107482,
      "auxiliary_loss_mlp": 0.0100275,
      "balance_loss_clip": 1.0135231,
      "balance_loss_mlp": 1.00190949,
      "epoch": 0.5838994769434257,
      "flos": 65995480765440.0,
      "grad_norm": 0.9926449275305432,
      "language_loss": 0.65051222,
      "learning_rate": 1.5573452801258545e-06,
      "loss": 0.67161453,
      "num_input_tokens_seen": 104694605,
      "step": 4856,
      "time_per_iteration": 3.060004234313965
    },
    {
      "auxiliary_loss_clip": 0.01183766,
      "auxiliary_loss_mlp": 0.0103181,
      "balance_loss_clip": 1.05093646,
      "balance_loss_mlp": 1.02405524,
      "epoch": 0.5840197198340649,
      "flos": 21470523546240.0,
      "grad_norm": 2.6006334768805117,
      "language_loss": 0.63738799,
      "learning_rate": 1.5565856609053824e-06,
      "loss": 0.65954375,
      "num_input_tokens_seen": 104713400,
      "step": 4857,
      "time_per_iteration": 2.5916316509246826
    },
    {
      "auxiliary_loss_clip": 0.01177074,
      "auxiliary_loss_mlp": 0.01023239,
      "balance_loss_clip": 1.05315757,
      "balance_loss_mlp": 1.01520419,
      "epoch": 0.5841399627247039,
      "flos": 19135144984320.0,
      "grad_norm": 2.2440921156067324,
      "language_loss": 0.80166101,
      "learning_rate": 1.5558261089505127e-06,
      "loss": 0.82366407,
      "num_input_tokens_seen": 104732130,
      "step": 4858,
      "time_per_iteration": 2.585061550140381
    },
    {
      "auxiliary_loss_clip": 0.0118076,
      "auxiliary_loss_mlp": 0.0102514,
      "balance_loss_clip": 1.05295038,
      "balance_loss_mlp": 1.01729572,
      "epoch": 0.584260205615343,
      "flos": 26425692558720.0,
      "grad_norm": 1.9355364060266538,
      "language_loss": 0.79923368,
      "learning_rate": 1.5550666243764697e-06,
      "loss": 0.82129276,
      "num_input_tokens_seen": 104750290,
      "step": 4859,
      "time_per_iteration": 2.6004128456115723
    },
    {
      "auxiliary_loss_clip": 0.01178453,
      "auxiliary_loss_mlp": 0.01027733,
      "balance_loss_clip": 1.04979718,
      "balance_loss_mlp": 1.01978207,
      "epoch": 0.584380448505982,
      "flos": 13881809174400.0,
      "grad_norm": 8.328182375055766,
      "language_loss": 0.77382755,
      "learning_rate": 1.554307207298465e-06,
      "loss": 0.79588938,
      "num_input_tokens_seen": 104768550,
      "step": 4860,
      "time_per_iteration": 2.683076858520508
    },
    {
      "auxiliary_loss_clip": 0.01180932,
      "auxiliary_loss_mlp": 0.01030104,
      "balance_loss_clip": 1.05493999,
      "balance_loss_mlp": 1.02202749,
      "epoch": 0.5845006913966212,
      "flos": 21543709507200.0,
      "grad_norm": 2.062234135423953,
      "language_loss": 0.78764683,
      "learning_rate": 1.553547857831704e-06,
      "loss": 0.80975717,
      "num_input_tokens_seen": 104785060,
      "step": 4861,
      "time_per_iteration": 2.5672149658203125
    },
    {
      "auxiliary_loss_clip": 0.01073776,
      "auxiliary_loss_mlp": 0.010043,
      "balance_loss_clip": 1.01541948,
      "balance_loss_mlp": 1.00338829,
      "epoch": 0.5846209342872603,
      "flos": 58375452712320.0,
      "grad_norm": 0.8885879105022757,
      "language_loss": 0.64107388,
      "learning_rate": 1.5527885760913771e-06,
      "loss": 0.66185468,
      "num_input_tokens_seen": 104834950,
      "step": 4862,
      "time_per_iteration": 2.9729039669036865
    },
    {
      "auxiliary_loss_clip": 0.01179648,
      "auxiliary_loss_mlp": 0.01027452,
      "balance_loss_clip": 1.04985642,
      "balance_loss_mlp": 1.01990604,
      "epoch": 0.5847411771778993,
      "flos": 18588045957120.0,
      "grad_norm": 1.6389713751644843,
      "language_loss": 0.76485562,
      "learning_rate": 1.552029362192668e-06,
      "loss": 0.78692669,
      "num_input_tokens_seen": 104854210,
      "step": 4863,
      "time_per_iteration": 2.5863001346588135
    },
    {
      "auxiliary_loss_clip": 0.01179824,
      "auxiliary_loss_mlp": 0.01026694,
      "balance_loss_clip": 1.04682958,
      "balance_loss_mlp": 1.01902592,
      "epoch": 0.5848614200685385,
      "flos": 24240780069120.0,
      "grad_norm": 7.027680098707396,
      "language_loss": 0.72844899,
      "learning_rate": 1.5512702162507478e-06,
      "loss": 0.75051415,
      "num_input_tokens_seen": 104874525,
      "step": 4864,
      "time_per_iteration": 2.66605544090271
    },
    {
      "auxiliary_loss_clip": 0.0108531,
      "auxiliary_loss_mlp": 0.01002543,
      "balance_loss_clip": 1.01417649,
      "balance_loss_mlp": 1.00163674,
      "epoch": 0.5849816629591775,
      "flos": 71660245933440.0,
      "grad_norm": 1.1712984856837487,
      "language_loss": 0.55788243,
      "learning_rate": 1.5505111383807792e-06,
      "loss": 0.57876098,
      "num_input_tokens_seen": 104937195,
      "step": 4865,
      "time_per_iteration": 3.379500389099121
    },
    {
      "auxiliary_loss_clip": 0.01189684,
      "auxiliary_loss_mlp": 0.01023366,
      "balance_loss_clip": 1.0453676,
      "balance_loss_mlp": 1.01596928,
      "epoch": 0.5851019058498166,
      "flos": 23802095266560.0,
      "grad_norm": 2.057388217018185,
      "language_loss": 0.80931973,
      "learning_rate": 1.5497521286979138e-06,
      "loss": 0.83145022,
      "num_input_tokens_seen": 104957435,
      "step": 4866,
      "time_per_iteration": 3.722374200820923
    },
    {
      "auxiliary_loss_clip": 0.01187605,
      "auxiliary_loss_mlp": 0.01026444,
      "balance_loss_clip": 1.04782236,
      "balance_loss_mlp": 1.01812363,
      "epoch": 0.5852221487404557,
      "flos": 24388516707840.0,
      "grad_norm": 2.134625681278378,
      "language_loss": 0.74654067,
      "learning_rate": 1.5489931873172927e-06,
      "loss": 0.76868123,
      "num_input_tokens_seen": 104978755,
      "step": 4867,
      "time_per_iteration": 2.703490734100342
    },
    {
      "auxiliary_loss_clip": 0.01169345,
      "auxiliary_loss_mlp": 0.01025177,
      "balance_loss_clip": 1.03843677,
      "balance_loss_mlp": 1.01758885,
      "epoch": 0.5853423916310948,
      "flos": 27271425260160.0,
      "grad_norm": 1.6392028309668616,
      "language_loss": 0.79512548,
      "learning_rate": 1.5482343143540467e-06,
      "loss": 0.81707072,
      "num_input_tokens_seen": 105000020,
      "step": 4868,
      "time_per_iteration": 2.7880067825317383
    },
    {
      "auxiliary_loss_clip": 0.01182317,
      "auxiliary_loss_mlp": 0.01207575,
      "balance_loss_clip": 1.04592395,
      "balance_loss_mlp": 1.00074291,
      "epoch": 0.5854626345217339,
      "flos": 11983786611840.0,
      "grad_norm": 2.2759841050783796,
      "language_loss": 0.82731551,
      "learning_rate": 1.547475509923295e-06,
      "loss": 0.85121447,
      "num_input_tokens_seen": 105017060,
      "step": 4869,
      "time_per_iteration": 3.582547426223755
    },
    {
      "auxiliary_loss_clip": 0.01100684,
      "auxiliary_loss_mlp": 0.0100508,
      "balance_loss_clip": 1.01471269,
      "balance_loss_mlp": 1.00422716,
      "epoch": 0.585582877412373,
      "flos": 64342335173760.0,
      "grad_norm": 0.7296818097349824,
      "language_loss": 0.56109798,
      "learning_rate": 1.5467167741401495e-06,
      "loss": 0.58215559,
      "num_input_tokens_seen": 105078540,
      "step": 4870,
      "time_per_iteration": 3.209188461303711
    },
    {
      "auxiliary_loss_clip": 0.0117823,
      "auxiliary_loss_mlp": 0.01026715,
      "balance_loss_clip": 1.04618907,
      "balance_loss_mlp": 1.01830482,
      "epoch": 0.5857031203030121,
      "flos": 17011926103680.0,
      "grad_norm": 2.2539610373524104,
      "language_loss": 0.71059501,
      "learning_rate": 1.5459581071197083e-06,
      "loss": 0.7326445,
      "num_input_tokens_seen": 105094200,
      "step": 4871,
      "time_per_iteration": 3.5734174251556396
    },
    {
      "auxiliary_loss_clip": 0.01184668,
      "auxiliary_loss_mlp": 0.01024218,
      "balance_loss_clip": 1.05492878,
      "balance_loss_mlp": 1.0164094,
      "epoch": 0.5858233631936511,
      "flos": 20885682303360.0,
      "grad_norm": 2.5309672744428275,
      "language_loss": 0.8336513,
      "learning_rate": 1.5451995089770624e-06,
      "loss": 0.85574019,
      "num_input_tokens_seen": 105113985,
      "step": 4872,
      "time_per_iteration": 2.6032564640045166
    },
    {
      "auxiliary_loss_clip": 0.0117353,
      "auxiliary_loss_mlp": 0.01024386,
      "balance_loss_clip": 1.05103254,
      "balance_loss_mlp": 1.01722193,
      "epoch": 0.5859436060842903,
      "flos": 23191902000000.0,
      "grad_norm": 1.473732080097969,
      "language_loss": 0.71833158,
      "learning_rate": 1.5444409798272885e-06,
      "loss": 0.74031073,
      "num_input_tokens_seen": 105138075,
      "step": 4873,
      "time_per_iteration": 2.7111001014709473
    },
    {
      "auxiliary_loss_clip": 0.01184412,
      "auxiliary_loss_mlp": 0.01029499,
      "balance_loss_clip": 1.04547834,
      "balance_loss_mlp": 1.02151155,
      "epoch": 0.5860638489749294,
      "flos": 22492648961280.0,
      "grad_norm": 2.0415740654648893,
      "language_loss": 0.80836761,
      "learning_rate": 1.543682519785456e-06,
      "loss": 0.83050674,
      "num_input_tokens_seen": 105156555,
      "step": 4874,
      "time_per_iteration": 2.682802677154541
    },
    {
      "auxiliary_loss_clip": 0.01182599,
      "auxiliary_loss_mlp": 0.01029405,
      "balance_loss_clip": 1.04963326,
      "balance_loss_mlp": 1.02193093,
      "epoch": 0.5861840918655684,
      "flos": 17566243764480.0,
      "grad_norm": 3.0274946860375693,
      "language_loss": 0.80214995,
      "learning_rate": 1.5429241289666219e-06,
      "loss": 0.82427001,
      "num_input_tokens_seen": 105174055,
      "step": 4875,
      "time_per_iteration": 2.6397883892059326
    },
    {
      "auxiliary_loss_clip": 0.01173049,
      "auxiliary_loss_mlp": 0.01024831,
      "balance_loss_clip": 1.04832816,
      "balance_loss_mlp": 1.01738095,
      "epoch": 0.5863043347562076,
      "flos": 25556152118400.0,
      "grad_norm": 2.034235324491874,
      "language_loss": 0.70302725,
      "learning_rate": 1.5421658074858342e-06,
      "loss": 0.72500598,
      "num_input_tokens_seen": 105192160,
      "step": 4876,
      "time_per_iteration": 2.616340160369873
    },
    {
      "auxiliary_loss_clip": 0.01178426,
      "auxiliary_loss_mlp": 0.01027421,
      "balance_loss_clip": 1.04940939,
      "balance_loss_mlp": 1.01915359,
      "epoch": 0.5864245776468466,
      "flos": 20667525050880.0,
      "grad_norm": 2.3798546124773647,
      "language_loss": 0.66164452,
      "learning_rate": 1.5414075554581298e-06,
      "loss": 0.68370306,
      "num_input_tokens_seen": 105210205,
      "step": 4877,
      "time_per_iteration": 2.633892774581909
    },
    {
      "auxiliary_loss_clip": 0.01177563,
      "auxiliary_loss_mlp": 0.01024058,
      "balance_loss_clip": 1.05131531,
      "balance_loss_mlp": 1.01657176,
      "epoch": 0.5865448205374857,
      "flos": 28913907490560.0,
      "grad_norm": 3.0465169502980776,
      "language_loss": 0.78568602,
      "learning_rate": 1.5406493729985348e-06,
      "loss": 0.8077023,
      "num_input_tokens_seen": 105229400,
      "step": 4878,
      "time_per_iteration": 2.6130406856536865
    },
    {
      "auxiliary_loss_clip": 0.01188921,
      "auxiliary_loss_mlp": 0.0120767,
      "balance_loss_clip": 1.0477984,
      "balance_loss_mlp": 1.00046062,
      "epoch": 0.5866650634281249,
      "flos": 25842575168640.0,
      "grad_norm": 2.0503180354937274,
      "language_loss": 0.71957427,
      "learning_rate": 1.5398912602220644e-06,
      "loss": 0.74354017,
      "num_input_tokens_seen": 105248675,
      "step": 4879,
      "time_per_iteration": 3.578547716140747
    },
    {
      "auxiliary_loss_clip": 0.01199927,
      "auxiliary_loss_mlp": 0.01026285,
      "balance_loss_clip": 1.04824615,
      "balance_loss_mlp": 1.01833355,
      "epoch": 0.5867853063187639,
      "flos": 17052325925760.0,
      "grad_norm": 1.9795459025238722,
      "language_loss": 0.78852737,
      "learning_rate": 1.539133217243724e-06,
      "loss": 0.81078947,
      "num_input_tokens_seen": 105265695,
      "step": 4880,
      "time_per_iteration": 2.696668863296509
    },
    {
      "auxiliary_loss_clip": 0.01191317,
      "auxiliary_loss_mlp": 0.01027111,
      "balance_loss_clip": 1.04842687,
      "balance_loss_mlp": 1.01839066,
      "epoch": 0.586905549209403,
      "flos": 24645026707200.0,
      "grad_norm": 2.1037043879476753,
      "language_loss": 0.76255965,
      "learning_rate": 1.5383752441785081e-06,
      "loss": 0.78474396,
      "num_input_tokens_seen": 105284920,
      "step": 4881,
      "time_per_iteration": 2.7361693382263184
    },
    {
      "auxiliary_loss_clip": 0.01184642,
      "auxiliary_loss_mlp": 0.01033958,
      "balance_loss_clip": 1.05191267,
      "balance_loss_mlp": 1.02603698,
      "epoch": 0.5870257921000421,
      "flos": 14720538723840.0,
      "grad_norm": 2.616195280533264,
      "language_loss": 0.86216676,
      "learning_rate": 1.5376173411414003e-06,
      "loss": 0.8843528,
      "num_input_tokens_seen": 105302960,
      "step": 4882,
      "time_per_iteration": 2.686936140060425
    },
    {
      "auxiliary_loss_clip": 0.01180646,
      "auxiliary_loss_mlp": 0.01030135,
      "balance_loss_clip": 1.04531074,
      "balance_loss_mlp": 1.02175438,
      "epoch": 0.5871460349906812,
      "flos": 23914998691200.0,
      "grad_norm": 1.9712013995111257,
      "language_loss": 0.79137629,
      "learning_rate": 1.5368595082473753e-06,
      "loss": 0.81348407,
      "num_input_tokens_seen": 105321260,
      "step": 4883,
      "time_per_iteration": 2.734874963760376
    },
    {
      "auxiliary_loss_clip": 0.01179814,
      "auxiliary_loss_mlp": 0.01021156,
      "balance_loss_clip": 1.0489912,
      "balance_loss_mlp": 1.01360965,
      "epoch": 0.5872662778813202,
      "flos": 22164174063360.0,
      "grad_norm": 1.6740186808080755,
      "language_loss": 0.77900171,
      "learning_rate": 1.5361017456113935e-06,
      "loss": 0.80101138,
      "num_input_tokens_seen": 105341610,
      "step": 4884,
      "time_per_iteration": 2.7206249237060547
    },
    {
      "auxiliary_loss_clip": 0.01181112,
      "auxiliary_loss_mlp": 0.01028717,
      "balance_loss_clip": 1.0504247,
      "balance_loss_mlp": 1.02023554,
      "epoch": 0.5873865207719594,
      "flos": 18441925430400.0,
      "grad_norm": 2.289607121838394,
      "language_loss": 0.85729283,
      "learning_rate": 1.5353440533484085e-06,
      "loss": 0.87939113,
      "num_input_tokens_seen": 105360465,
      "step": 4885,
      "time_per_iteration": 2.6060187816619873
    },
    {
      "auxiliary_loss_clip": 0.01185124,
      "auxiliary_loss_mlp": 0.01030075,
      "balance_loss_clip": 1.05022717,
      "balance_loss_mlp": 1.02146244,
      "epoch": 0.5875067636625985,
      "flos": 54015321427200.0,
      "grad_norm": 3.8262459692499395,
      "language_loss": 0.66050398,
      "learning_rate": 1.534586431573361e-06,
      "loss": 0.68265593,
      "num_input_tokens_seen": 105385405,
      "step": 4886,
      "time_per_iteration": 2.9933197498321533
    },
    {
      "auxiliary_loss_clip": 0.01189964,
      "auxiliary_loss_mlp": 0.01024665,
      "balance_loss_clip": 1.041538,
      "balance_loss_mlp": 1.0154624,
      "epoch": 0.5876270065532375,
      "flos": 27995707100160.0,
      "grad_norm": 2.2666995685801328,
      "language_loss": 0.79289639,
      "learning_rate": 1.5338288804011817e-06,
      "loss": 0.81504273,
      "num_input_tokens_seen": 105404905,
      "step": 4887,
      "time_per_iteration": 2.741253137588501
    },
    {
      "auxiliary_loss_clip": 0.01175822,
      "auxiliary_loss_mlp": 0.010286,
      "balance_loss_clip": 1.04522681,
      "balance_loss_mlp": 1.02028525,
      "epoch": 0.5877472494438767,
      "flos": 21361462876800.0,
      "grad_norm": 3.7781408586463865,
      "language_loss": 0.71621031,
      "learning_rate": 1.533071399946791e-06,
      "loss": 0.73825455,
      "num_input_tokens_seen": 105423650,
      "step": 4888,
      "time_per_iteration": 2.635658025741577
    },
    {
      "auxiliary_loss_clip": 0.0118403,
      "auxiliary_loss_mlp": 0.0102785,
      "balance_loss_clip": 1.0479418,
      "balance_loss_mlp": 1.0201726,
      "epoch": 0.5878674923345157,
      "flos": 22383013674240.0,
      "grad_norm": 3.2060290583608864,
      "language_loss": 0.57417649,
      "learning_rate": 1.5323139903250977e-06,
      "loss": 0.59629536,
      "num_input_tokens_seen": 105444255,
      "step": 4889,
      "time_per_iteration": 2.6394143104553223
    },
    {
      "auxiliary_loss_clip": 0.01187262,
      "auxiliary_loss_mlp": 0.01022838,
      "balance_loss_clip": 1.05367982,
      "balance_loss_mlp": 1.01511586,
      "epoch": 0.5879877352251548,
      "flos": 21868664872320.0,
      "grad_norm": 1.47148331296307,
      "language_loss": 0.77138072,
      "learning_rate": 1.5315566516510002e-06,
      "loss": 0.79348171,
      "num_input_tokens_seen": 105462425,
      "step": 4890,
      "time_per_iteration": 2.6855969429016113
    },
    {
      "auxiliary_loss_clip": 0.01177148,
      "auxiliary_loss_mlp": 0.01027706,
      "balance_loss_clip": 1.05251408,
      "balance_loss_mlp": 1.01926637,
      "epoch": 0.5881079781157939,
      "flos": 17493811989120.0,
      "grad_norm": 1.7535980493010037,
      "language_loss": 0.67501551,
      "learning_rate": 1.5307993840393857e-06,
      "loss": 0.69706404,
      "num_input_tokens_seen": 105480505,
      "step": 4891,
      "time_per_iteration": 2.575056314468384
    },
    {
      "auxiliary_loss_clip": 0.01174521,
      "auxiliary_loss_mlp": 0.01023236,
      "balance_loss_clip": 1.05056643,
      "balance_loss_mlp": 1.01557028,
      "epoch": 0.588228221006433,
      "flos": 22601853285120.0,
      "grad_norm": 1.9081278319065038,
      "language_loss": 0.80330604,
      "learning_rate": 1.530042187605132e-06,
      "loss": 0.82528365,
      "num_input_tokens_seen": 105499760,
      "step": 4892,
      "time_per_iteration": 2.553555965423584
    },
    {
      "auxiliary_loss_clip": 0.0118197,
      "auxiliary_loss_mlp": 0.01207749,
      "balance_loss_clip": 1.05269945,
      "balance_loss_mlp": 1.0004909,
      "epoch": 0.5883484638970721,
      "flos": 26176939896960.0,
      "grad_norm": 1.466944773047568,
      "language_loss": 0.84261,
      "learning_rate": 1.5292850624631044e-06,
      "loss": 0.86650717,
      "num_input_tokens_seen": 105521955,
      "step": 4893,
      "time_per_iteration": 3.614023208618164
    },
    {
      "auxiliary_loss_clip": 0.01176723,
      "auxiliary_loss_mlp": 0.01030512,
      "balance_loss_clip": 1.05205083,
      "balance_loss_mlp": 1.02209544,
      "epoch": 0.5884687067877111,
      "flos": 30443737691520.0,
      "grad_norm": 2.2708214564622553,
      "language_loss": 0.8029381,
      "learning_rate": 1.5285280087281593e-06,
      "loss": 0.82501042,
      "num_input_tokens_seen": 105542685,
      "step": 4894,
      "time_per_iteration": 2.629199981689453
    },
    {
      "auxiliary_loss_clip": 0.01087255,
      "auxiliary_loss_mlp": 0.01001553,
      "balance_loss_clip": 1.01540279,
      "balance_loss_mlp": 1.00060558,
      "epoch": 0.5885889496783503,
      "flos": 70507550580480.0,
      "grad_norm": 0.6379352726712537,
      "language_loss": 0.56602633,
      "learning_rate": 1.5277710265151398e-06,
      "loss": 0.58691442,
      "num_input_tokens_seen": 105612165,
      "step": 4895,
      "time_per_iteration": 3.3986165523529053
    },
    {
      "auxiliary_loss_clip": 0.01180349,
      "auxiliary_loss_mlp": 0.0102237,
      "balance_loss_clip": 1.05173433,
      "balance_loss_mlp": 1.01380444,
      "epoch": 0.5887091925689893,
      "flos": 19098767485440.0,
      "grad_norm": 2.807942276364715,
      "language_loss": 0.77060586,
      "learning_rate": 1.5270141159388803e-06,
      "loss": 0.79263306,
      "num_input_tokens_seen": 105629185,
      "step": 4896,
      "time_per_iteration": 3.5994765758514404
    },
    {
      "auxiliary_loss_clip": 0.01175913,
      "auxiliary_loss_mlp": 0.01026567,
      "balance_loss_clip": 1.05049014,
      "balance_loss_mlp": 1.01824629,
      "epoch": 0.5888294354596284,
      "flos": 23294282739840.0,
      "grad_norm": 1.6761763079474563,
      "language_loss": 0.80740786,
      "learning_rate": 1.526257277114203e-06,
      "loss": 0.82943267,
      "num_input_tokens_seen": 105650260,
      "step": 4897,
      "time_per_iteration": 2.648047685623169
    },
    {
      "auxiliary_loss_clip": 0.01178193,
      "auxiliary_loss_mlp": 0.01023814,
      "balance_loss_clip": 1.04955924,
      "balance_loss_mlp": 1.01591611,
      "epoch": 0.5889496783502676,
      "flos": 21981532383360.0,
      "grad_norm": 3.876413526427402,
      "language_loss": 0.79982144,
      "learning_rate": 1.5255005101559201e-06,
      "loss": 0.82184148,
      "num_input_tokens_seen": 105667870,
      "step": 4898,
      "time_per_iteration": 3.506685256958008
    },
    {
      "auxiliary_loss_clip": 0.01182725,
      "auxiliary_loss_mlp": 0.01025939,
      "balance_loss_clip": 1.05110359,
      "balance_loss_mlp": 1.01837802,
      "epoch": 0.5890699212409066,
      "flos": 21685233093120.0,
      "grad_norm": 2.0607199023710554,
      "language_loss": 0.77147263,
      "learning_rate": 1.524743815178833e-06,
      "loss": 0.79355931,
      "num_input_tokens_seen": 105685830,
      "step": 4899,
      "time_per_iteration": 2.5936882495880127
    },
    {
      "auxiliary_loss_clip": 0.01183692,
      "auxiliary_loss_mlp": 0.01025433,
      "balance_loss_clip": 1.04802179,
      "balance_loss_mlp": 1.01735687,
      "epoch": 0.5891901641315457,
      "flos": 19464553635840.0,
      "grad_norm": 1.8443071637626203,
      "language_loss": 0.81027591,
      "learning_rate": 1.5239871922977315e-06,
      "loss": 0.83236712,
      "num_input_tokens_seen": 105705745,
      "step": 4900,
      "time_per_iteration": 2.674625873565674
    },
    {
      "auxiliary_loss_clip": 0.01179971,
      "auxiliary_loss_mlp": 0.01029666,
      "balance_loss_clip": 1.04646301,
      "balance_loss_mlp": 1.02123761,
      "epoch": 0.5893104070221848,
      "flos": 19609884063360.0,
      "grad_norm": 5.219307032651852,
      "language_loss": 0.89795744,
      "learning_rate": 1.523230641627394e-06,
      "loss": 0.92005384,
      "num_input_tokens_seen": 105724730,
      "step": 4901,
      "time_per_iteration": 2.650041341781616
    },
    {
      "auxiliary_loss_clip": 0.01187319,
      "auxiliary_loss_mlp": 0.01024293,
      "balance_loss_clip": 1.04139376,
      "balance_loss_mlp": 1.01666379,
      "epoch": 0.5894306499128239,
      "flos": 29060063930880.0,
      "grad_norm": 2.12857435700551,
      "language_loss": 0.72908199,
      "learning_rate": 1.5224741632825888e-06,
      "loss": 0.75119811,
      "num_input_tokens_seen": 105744920,
      "step": 4902,
      "time_per_iteration": 2.792224884033203
    },
    {
      "auxiliary_loss_clip": 0.01181396,
      "auxiliary_loss_mlp": 0.01027119,
      "balance_loss_clip": 1.05498362,
      "balance_loss_mlp": 1.0187031,
      "epoch": 0.589550892803463,
      "flos": 42298890721920.0,
      "grad_norm": 1.9903547919849267,
      "language_loss": 0.69293666,
      "learning_rate": 1.521717757378074e-06,
      "loss": 0.71502179,
      "num_input_tokens_seen": 105765465,
      "step": 4903,
      "time_per_iteration": 2.7947468757629395
    },
    {
      "auxiliary_loss_clip": 0.01187334,
      "auxiliary_loss_mlp": 0.01029034,
      "balance_loss_clip": 1.0535748,
      "balance_loss_mlp": 1.02036774,
      "epoch": 0.5896711356941021,
      "flos": 14137062197760.0,
      "grad_norm": 1.8372901539024993,
      "language_loss": 0.68932092,
      "learning_rate": 1.5209614240285943e-06,
      "loss": 0.71148455,
      "num_input_tokens_seen": 105783120,
      "step": 4904,
      "time_per_iteration": 2.589174270629883
    },
    {
      "auxiliary_loss_clip": 0.01175471,
      "auxiliary_loss_mlp": 0.01208334,
      "balance_loss_clip": 1.05041242,
      "balance_loss_mlp": 1.00062108,
      "epoch": 0.5897913785847412,
      "flos": 17201355454080.0,
      "grad_norm": 2.119406146471068,
      "language_loss": 0.84708726,
      "learning_rate": 1.520205163348887e-06,
      "loss": 0.87092531,
      "num_input_tokens_seen": 105801055,
      "step": 4905,
      "time_per_iteration": 3.449571371078491
    },
    {
      "auxiliary_loss_clip": 0.01094698,
      "auxiliary_loss_mlp": 0.01000423,
      "balance_loss_clip": 1.01459432,
      "balance_loss_mlp": 0.99954712,
      "epoch": 0.5899116214753802,
      "flos": 48794164202880.0,
      "grad_norm": 0.8353687913313617,
      "language_loss": 0.56947553,
      "learning_rate": 1.519448975453674e-06,
      "loss": 0.59042674,
      "num_input_tokens_seen": 105856155,
      "step": 4906,
      "time_per_iteration": 3.034999370574951
    },
    {
      "auxiliary_loss_clip": 0.01180886,
      "auxiliary_loss_mlp": 0.01208417,
      "balance_loss_clip": 1.05196381,
      "balance_loss_mlp": 1.00056469,
      "epoch": 0.5900318643660194,
      "flos": 21103659987840.0,
      "grad_norm": 2.5920875081977197,
      "language_loss": 0.7613191,
      "learning_rate": 1.5186928604576696e-06,
      "loss": 0.78521216,
      "num_input_tokens_seen": 105873350,
      "step": 4907,
      "time_per_iteration": 2.6685900688171387
    },
    {
      "auxiliary_loss_clip": 0.01181963,
      "auxiliary_loss_mlp": 0.0102747,
      "balance_loss_clip": 1.04704833,
      "balance_loss_mlp": 1.01957262,
      "epoch": 0.5901521072566585,
      "flos": 21178390233600.0,
      "grad_norm": 1.953223602522409,
      "language_loss": 0.77272862,
      "learning_rate": 1.5179368184755752e-06,
      "loss": 0.79482305,
      "num_input_tokens_seen": 105891435,
      "step": 4908,
      "time_per_iteration": 2.6682090759277344
    },
    {
      "auxiliary_loss_clip": 0.01181159,
      "auxiliary_loss_mlp": 0.01021199,
      "balance_loss_clip": 1.05015004,
      "balance_loss_mlp": 1.01327157,
      "epoch": 0.5902723501472975,
      "flos": 20225967160320.0,
      "grad_norm": 1.551561651679436,
      "language_loss": 0.82711053,
      "learning_rate": 1.5171808496220821e-06,
      "loss": 0.84913409,
      "num_input_tokens_seen": 105910190,
      "step": 4909,
      "time_per_iteration": 2.6357421875
    },
    {
      "auxiliary_loss_clip": 0.01187655,
      "auxiliary_loss_mlp": 0.01025648,
      "balance_loss_clip": 1.04942536,
      "balance_loss_mlp": 1.01817918,
      "epoch": 0.5903925930379367,
      "flos": 22964407211520.0,
      "grad_norm": 2.6703683851130267,
      "language_loss": 0.8181076,
      "learning_rate": 1.5164249540118708e-06,
      "loss": 0.8402406,
      "num_input_tokens_seen": 105929315,
      "step": 4910,
      "time_per_iteration": 2.636833429336548
    },
    {
      "auxiliary_loss_clip": 0.01193759,
      "auxiliary_loss_mlp": 0.01027253,
      "balance_loss_clip": 1.04446125,
      "balance_loss_mlp": 1.01886654,
      "epoch": 0.5905128359285757,
      "flos": 23367720096000.0,
      "grad_norm": 1.6106494536463773,
      "language_loss": 0.83419544,
      "learning_rate": 1.5156691317596093e-06,
      "loss": 0.8564055,
      "num_input_tokens_seen": 105950740,
      "step": 4911,
      "time_per_iteration": 2.703150510787964
    },
    {
      "auxiliary_loss_clip": 0.01183988,
      "auxiliary_loss_mlp": 0.01208248,
      "balance_loss_clip": 1.05166101,
      "balance_loss_mlp": 1.00063205,
      "epoch": 0.5906330788192148,
      "flos": 28032335994240.0,
      "grad_norm": 2.8816682173653603,
      "language_loss": 0.66731668,
      "learning_rate": 1.5149133829799556e-06,
      "loss": 0.691239,
      "num_input_tokens_seen": 105968735,
      "step": 4912,
      "time_per_iteration": 2.6765620708465576
    },
    {
      "auxiliary_loss_clip": 0.01189321,
      "auxiliary_loss_mlp": 0.01033975,
      "balance_loss_clip": 1.0487802,
      "balance_loss_mlp": 1.02561212,
      "epoch": 0.590753321709854,
      "flos": 18477943793280.0,
      "grad_norm": 2.0124531461264237,
      "language_loss": 0.80904227,
      "learning_rate": 1.5141577077875556e-06,
      "loss": 0.83127522,
      "num_input_tokens_seen": 105986060,
      "step": 4913,
      "time_per_iteration": 2.6012535095214844
    },
    {
      "auxiliary_loss_clip": 0.01182465,
      "auxiliary_loss_mlp": 0.01026531,
      "balance_loss_clip": 1.05158961,
      "balance_loss_mlp": 1.01839471,
      "epoch": 0.590873564600493,
      "flos": 16873706568960.0,
      "grad_norm": 3.401861095412418,
      "language_loss": 0.72753769,
      "learning_rate": 1.5134021062970451e-06,
      "loss": 0.74962771,
      "num_input_tokens_seen": 106004440,
      "step": 4914,
      "time_per_iteration": 2.6264994144439697
    },
    {
      "auxiliary_loss_clip": 0.01172394,
      "auxiliary_loss_mlp": 0.01026525,
      "balance_loss_clip": 1.04709995,
      "balance_loss_mlp": 1.01891947,
      "epoch": 0.5909938074911321,
      "flos": 13516166678400.0,
      "grad_norm": 2.0059170110330276,
      "language_loss": 0.81276321,
      "learning_rate": 1.5126465786230483e-06,
      "loss": 0.83475244,
      "num_input_tokens_seen": 106021215,
      "step": 4915,
      "time_per_iteration": 2.621782064437866
    },
    {
      "auxiliary_loss_clip": 0.01175356,
      "auxiliary_loss_mlp": 0.01028233,
      "balance_loss_clip": 1.05088627,
      "balance_loss_mlp": 1.02005541,
      "epoch": 0.5911140503817712,
      "flos": 26024067613440.0,
      "grad_norm": 1.877102033018548,
      "language_loss": 0.82146412,
      "learning_rate": 1.5118911248801787e-06,
      "loss": 0.84350002,
      "num_input_tokens_seen": 106039225,
      "step": 4916,
      "time_per_iteration": 2.6533169746398926
    },
    {
      "auxiliary_loss_clip": 0.01176191,
      "auxiliary_loss_mlp": 0.01024304,
      "balance_loss_clip": 1.04993415,
      "balance_loss_mlp": 1.01686549,
      "epoch": 0.5912342932724103,
      "flos": 23258731253760.0,
      "grad_norm": 1.994271732875616,
      "language_loss": 0.79776669,
      "learning_rate": 1.5111357451830364e-06,
      "loss": 0.81977165,
      "num_input_tokens_seen": 106057920,
      "step": 4917,
      "time_per_iteration": 2.6392996311187744
    },
    {
      "auxiliary_loss_clip": 0.01179727,
      "auxiliary_loss_mlp": 0.0102943,
      "balance_loss_clip": 1.05007553,
      "balance_loss_mlp": 1.02149701,
      "epoch": 0.5913545361630493,
      "flos": 19573039687680.0,
      "grad_norm": 1.8554701373741849,
      "language_loss": 0.71158898,
      "learning_rate": 1.5103804396462131e-06,
      "loss": 0.73368049,
      "num_input_tokens_seen": 106077855,
      "step": 4918,
      "time_per_iteration": 2.717782497406006
    },
    {
      "auxiliary_loss_clip": 0.01184203,
      "auxiliary_loss_mlp": 0.01031929,
      "balance_loss_clip": 1.05022717,
      "balance_loss_mlp": 1.02313185,
      "epoch": 0.5914747790536885,
      "flos": 26213532877440.0,
      "grad_norm": 11.858401227931733,
      "language_loss": 0.80239666,
      "learning_rate": 1.5096252083842877e-06,
      "loss": 0.82455802,
      "num_input_tokens_seen": 106097065,
      "step": 4919,
      "time_per_iteration": 3.638101577758789
    },
    {
      "auxiliary_loss_clip": 0.01177,
      "auxiliary_loss_mlp": 0.01026762,
      "balance_loss_clip": 1.0480957,
      "balance_loss_mlp": 1.01864362,
      "epoch": 0.5915950219443276,
      "flos": 27417545786880.0,
      "grad_norm": 2.2536159252265393,
      "language_loss": 0.85667109,
      "learning_rate": 1.5088700515118285e-06,
      "loss": 0.87870866,
      "num_input_tokens_seen": 106116385,
      "step": 4920,
      "time_per_iteration": 2.631978988647461
    },
    {
      "auxiliary_loss_clip": 0.01181069,
      "auxiliary_loss_mlp": 0.01026813,
      "balance_loss_clip": 1.0483042,
      "balance_loss_mlp": 1.01847196,
      "epoch": 0.5917152648349666,
      "flos": 21907879545600.0,
      "grad_norm": 1.8833554757582727,
      "language_loss": 0.66466105,
      "learning_rate": 1.508114969143392e-06,
      "loss": 0.68673992,
      "num_input_tokens_seen": 106136370,
      "step": 4921,
      "time_per_iteration": 2.66365647315979
    },
    {
      "auxiliary_loss_clip": 0.01183259,
      "auxiliary_loss_mlp": 0.01029221,
      "balance_loss_clip": 1.04760146,
      "balance_loss_mlp": 1.02146912,
      "epoch": 0.5918355077256057,
      "flos": 28109185142400.0,
      "grad_norm": 1.5867631421941308,
      "language_loss": 0.77458429,
      "learning_rate": 1.5073599613935238e-06,
      "loss": 0.79670906,
      "num_input_tokens_seen": 106158490,
      "step": 4922,
      "time_per_iteration": 2.680325508117676
    },
    {
      "auxiliary_loss_clip": 0.0118409,
      "auxiliary_loss_mlp": 0.01025976,
      "balance_loss_clip": 1.0490427,
      "balance_loss_mlp": 1.01755416,
      "epoch": 0.5919557506162448,
      "flos": 28183807647360.0,
      "grad_norm": 2.092698081927904,
      "language_loss": 0.57819438,
      "learning_rate": 1.5066050283767574e-06,
      "loss": 0.60029501,
      "num_input_tokens_seen": 106179170,
      "step": 4923,
      "time_per_iteration": 3.5564801692962646
    },
    {
      "auxiliary_loss_clip": 0.01177214,
      "auxiliary_loss_mlp": 0.01026463,
      "balance_loss_clip": 1.0485599,
      "balance_loss_mlp": 1.01869655,
      "epoch": 0.5920759935068839,
      "flos": 12094355652480.0,
      "grad_norm": 2.1235582296190105,
      "language_loss": 0.83121574,
      "learning_rate": 1.505850170207616e-06,
      "loss": 0.85325253,
      "num_input_tokens_seen": 106196035,
      "step": 4924,
      "time_per_iteration": 3.500957489013672
    },
    {
      "auxiliary_loss_clip": 0.01180157,
      "auxiliary_loss_mlp": 0.01026839,
      "balance_loss_clip": 1.0471344,
      "balance_loss_mlp": 1.01901317,
      "epoch": 0.592196236397523,
      "flos": 29424772673280.0,
      "grad_norm": 2.5291507092596275,
      "language_loss": 0.78154123,
      "learning_rate": 1.505095387000611e-06,
      "loss": 0.80361116,
      "num_input_tokens_seen": 106218335,
      "step": 4925,
      "time_per_iteration": 2.675934076309204
    },
    {
      "auxiliary_loss_clip": 0.01173733,
      "auxiliary_loss_mlp": 0.01025709,
      "balance_loss_clip": 1.04911387,
      "balance_loss_mlp": 1.01797795,
      "epoch": 0.5923164792881621,
      "flos": 24384709866240.0,
      "grad_norm": 3.9239432824861122,
      "language_loss": 0.74712992,
      "learning_rate": 1.504340678870242e-06,
      "loss": 0.76912427,
      "num_input_tokens_seen": 106236550,
      "step": 4926,
      "time_per_iteration": 2.6223347187042236
    },
    {
      "auxiliary_loss_clip": 0.01178061,
      "auxiliary_loss_mlp": 0.01031527,
      "balance_loss_clip": 1.0500803,
      "balance_loss_mlp": 1.02412415,
      "epoch": 0.5924367221788012,
      "flos": 24024238928640.0,
      "grad_norm": 2.049388733924979,
      "language_loss": 0.89807308,
      "learning_rate": 1.5035860459309989e-06,
      "loss": 0.920169,
      "num_input_tokens_seen": 106254265,
      "step": 4927,
      "time_per_iteration": 2.6118178367614746
    },
    {
      "auxiliary_loss_clip": 0.01176395,
      "auxiliary_loss_mlp": 0.01027022,
      "balance_loss_clip": 1.04736042,
      "balance_loss_mlp": 1.01841569,
      "epoch": 0.5925569650694402,
      "flos": 26870590414080.0,
      "grad_norm": 2.0854191580768866,
      "language_loss": 0.63865203,
      "learning_rate": 1.5028314882973568e-06,
      "loss": 0.66068619,
      "num_input_tokens_seen": 106274670,
      "step": 4928,
      "time_per_iteration": 2.6959943771362305
    },
    {
      "auxiliary_loss_clip": 0.01181868,
      "auxiliary_loss_mlp": 0.01031066,
      "balance_loss_clip": 1.04978609,
      "balance_loss_mlp": 1.02272785,
      "epoch": 0.5926772079600794,
      "flos": 22302788647680.0,
      "grad_norm": 1.8455883842269656,
      "language_loss": 0.84748411,
      "learning_rate": 1.502077006083783e-06,
      "loss": 0.86961353,
      "num_input_tokens_seen": 106293330,
      "step": 4929,
      "time_per_iteration": 2.6212308406829834
    },
    {
      "auxiliary_loss_clip": 0.01183696,
      "auxiliary_loss_mlp": 0.01207212,
      "balance_loss_clip": 1.05133963,
      "balance_loss_mlp": 1.00056815,
      "epoch": 0.5927974508507184,
      "flos": 19865244827520.0,
      "grad_norm": 1.802500216493641,
      "language_loss": 0.76476908,
      "learning_rate": 1.5013225994047315e-06,
      "loss": 0.78867811,
      "num_input_tokens_seen": 106310960,
      "step": 4930,
      "time_per_iteration": 2.9664652347564697
    },
    {
      "auxiliary_loss_clip": 0.01181214,
      "auxiliary_loss_mlp": 0.01207477,
      "balance_loss_clip": 1.05182981,
      "balance_loss_mlp": 1.00055826,
      "epoch": 0.5929176937413575,
      "flos": 15776743167360.0,
      "grad_norm": 1.545247616803266,
      "language_loss": 0.80784649,
      "learning_rate": 1.5005682683746452e-06,
      "loss": 0.83173335,
      "num_input_tokens_seen": 106329475,
      "step": 4931,
      "time_per_iteration": 2.5932412147521973
    },
    {
      "auxiliary_loss_clip": 0.01181417,
      "auxiliary_loss_mlp": 0.0102684,
      "balance_loss_clip": 1.05406106,
      "balance_loss_mlp": 1.01882052,
      "epoch": 0.5930379366319967,
      "flos": 17601472028160.0,
      "grad_norm": 2.464654182752228,
      "language_loss": 0.72816181,
      "learning_rate": 1.4998140131079553e-06,
      "loss": 0.75024438,
      "num_input_tokens_seen": 106345565,
      "step": 4932,
      "time_per_iteration": 3.4788818359375
    },
    {
      "auxiliary_loss_clip": 0.01186043,
      "auxiliary_loss_mlp": 0.0120749,
      "balance_loss_clip": 1.04367805,
      "balance_loss_mlp": 1.00052953,
      "epoch": 0.5931581795226357,
      "flos": 17704283731200.0,
      "grad_norm": 1.9398128817481959,
      "language_loss": 0.73784542,
      "learning_rate": 1.4990598337190821e-06,
      "loss": 0.76178074,
      "num_input_tokens_seen": 106361920,
      "step": 4933,
      "time_per_iteration": 2.6772384643554688
    },
    {
      "auxiliary_loss_clip": 0.01176058,
      "auxiliary_loss_mlp": 0.01208172,
      "balance_loss_clip": 1.05168188,
      "balance_loss_mlp": 1.00046885,
      "epoch": 0.5932784224132748,
      "flos": 24280102483200.0,
      "grad_norm": 1.7503417008804958,
      "language_loss": 0.67943716,
      "learning_rate": 1.4983057303224338e-06,
      "loss": 0.7032795,
      "num_input_tokens_seen": 106381735,
      "step": 4934,
      "time_per_iteration": 2.652303457260132
    },
    {
      "auxiliary_loss_clip": 0.01185955,
      "auxiliary_loss_mlp": 0.01028883,
      "balance_loss_clip": 1.04590762,
      "balance_loss_mlp": 1.02120018,
      "epoch": 0.5933986653039139,
      "flos": 22926700909440.0,
      "grad_norm": 2.4837869575886957,
      "language_loss": 0.87657654,
      "learning_rate": 1.4975517030324072e-06,
      "loss": 0.89872491,
      "num_input_tokens_seen": 106399745,
      "step": 4935,
      "time_per_iteration": 2.751981258392334
    },
    {
      "auxiliary_loss_clip": 0.01072203,
      "auxiliary_loss_mlp": 0.01199056,
      "balance_loss_clip": 1.01524127,
      "balance_loss_mlp": 0.99999136,
      "epoch": 0.593518908194553,
      "flos": 71121730256640.0,
      "grad_norm": 0.7794444077306535,
      "language_loss": 0.61765844,
      "learning_rate": 1.4967977519633882e-06,
      "loss": 0.64037102,
      "num_input_tokens_seen": 106457205,
      "step": 4936,
      "time_per_iteration": 3.2475223541259766
    },
    {
      "auxiliary_loss_clip": 0.01183013,
      "auxiliary_loss_mlp": 0.01023068,
      "balance_loss_clip": 1.04713893,
      "balance_loss_mlp": 1.01490796,
      "epoch": 0.593639151085192,
      "flos": 20448649526400.0,
      "grad_norm": 1.8661114945224706,
      "language_loss": 0.78093863,
      "learning_rate": 1.4960438772297494e-06,
      "loss": 0.80299938,
      "num_input_tokens_seen": 106474250,
      "step": 4937,
      "time_per_iteration": 2.6842212677001953
    },
    {
      "auxiliary_loss_clip": 0.01183454,
      "auxiliary_loss_mlp": 0.01028198,
      "balance_loss_clip": 1.04776525,
      "balance_loss_mlp": 1.02024102,
      "epoch": 0.5937593939758312,
      "flos": 30883428074880.0,
      "grad_norm": 2.563171154936432,
      "language_loss": 0.73748839,
      "learning_rate": 1.495290078945855e-06,
      "loss": 0.75960493,
      "num_input_tokens_seen": 106494015,
      "step": 4938,
      "time_per_iteration": 2.7095792293548584
    },
    {
      "auxiliary_loss_clip": 0.01175811,
      "auxiliary_loss_mlp": 0.0102952,
      "balance_loss_clip": 1.05213547,
      "balance_loss_mlp": 1.02162814,
      "epoch": 0.5938796368664703,
      "flos": 36898069668480.0,
      "grad_norm": 2.1895325495592215,
      "language_loss": 0.74144399,
      "learning_rate": 1.4945363572260529e-06,
      "loss": 0.76349723,
      "num_input_tokens_seen": 106515010,
      "step": 4939,
      "time_per_iteration": 2.6881215572357178
    },
    {
      "auxiliary_loss_clip": 0.01178797,
      "auxiliary_loss_mlp": 0.01025207,
      "balance_loss_clip": 1.04997349,
      "balance_loss_mlp": 1.01733589,
      "epoch": 0.5939998797571093,
      "flos": 23842926051840.0,
      "grad_norm": 2.267807587514518,
      "language_loss": 0.68092442,
      "learning_rate": 1.4937827121846845e-06,
      "loss": 0.70296448,
      "num_input_tokens_seen": 106535265,
      "step": 4940,
      "time_per_iteration": 2.612295150756836
    },
    {
      "auxiliary_loss_clip": 0.01178999,
      "auxiliary_loss_mlp": 0.01032473,
      "balance_loss_clip": 1.0489372,
      "balance_loss_mlp": 1.02474785,
      "epoch": 0.5941201226477485,
      "flos": 25191407462400.0,
      "grad_norm": 2.933284070976646,
      "language_loss": 0.73362219,
      "learning_rate": 1.4930291439360755e-06,
      "loss": 0.75573695,
      "num_input_tokens_seen": 106557830,
      "step": 4941,
      "time_per_iteration": 2.6800875663757324
    },
    {
      "auxiliary_loss_clip": 0.01182233,
      "auxiliary_loss_mlp": 0.01028527,
      "balance_loss_clip": 1.05194962,
      "balance_loss_mlp": 1.01962256,
      "epoch": 0.5942403655383875,
      "flos": 22418996123520.0,
      "grad_norm": 1.930079849737775,
      "language_loss": 0.7951774,
      "learning_rate": 1.4922756525945427e-06,
      "loss": 0.81728506,
      "num_input_tokens_seen": 106577140,
      "step": 4942,
      "time_per_iteration": 2.652614116668701
    },
    {
      "auxiliary_loss_clip": 0.01079643,
      "auxiliary_loss_mlp": 0.01001217,
      "balance_loss_clip": 1.0148983,
      "balance_loss_mlp": 1.00043583,
      "epoch": 0.5943606084290266,
      "flos": 67629310796160.0,
      "grad_norm": 0.9039429857260907,
      "language_loss": 0.59551191,
      "learning_rate": 1.4915222382743894e-06,
      "loss": 0.61632049,
      "num_input_tokens_seen": 106635975,
      "step": 4943,
      "time_per_iteration": 3.2087349891662598
    },
    {
      "auxiliary_loss_clip": 0.01183732,
      "auxiliary_loss_mlp": 0.0102711,
      "balance_loss_clip": 1.05396914,
      "balance_loss_mlp": 1.01868463,
      "epoch": 0.5944808513196658,
      "flos": 18223157646720.0,
      "grad_norm": 2.386716880040873,
      "language_loss": 0.7206285,
      "learning_rate": 1.4907689010899085e-06,
      "loss": 0.74273688,
      "num_input_tokens_seen": 106653555,
      "step": 4944,
      "time_per_iteration": 2.588353157043457
    },
    {
      "auxiliary_loss_clip": 0.01182031,
      "auxiliary_loss_mlp": 0.01022793,
      "balance_loss_clip": 1.04872799,
      "balance_loss_mlp": 1.0146507,
      "epoch": 0.5946010942103048,
      "flos": 24790824011520.0,
      "grad_norm": 4.510623908779433,
      "language_loss": 0.627738,
      "learning_rate": 1.4900156411553804e-06,
      "loss": 0.64978623,
      "num_input_tokens_seen": 106673385,
      "step": 4945,
      "time_per_iteration": 2.6666996479034424
    },
    {
      "auxiliary_loss_clip": 0.01185255,
      "auxiliary_loss_mlp": 0.01025606,
      "balance_loss_clip": 1.04986811,
      "balance_loss_mlp": 1.01739836,
      "epoch": 0.5947213371009439,
      "flos": 15231619388160.0,
      "grad_norm": 2.0636993341333865,
      "language_loss": 0.85954702,
      "learning_rate": 1.4892624585850739e-06,
      "loss": 0.88165557,
      "num_input_tokens_seen": 106691740,
      "step": 4946,
      "time_per_iteration": 3.714491128921509
    },
    {
      "auxiliary_loss_clip": 0.01179987,
      "auxiliary_loss_mlp": 0.01027132,
      "balance_loss_clip": 1.05331492,
      "balance_loss_mlp": 1.01888895,
      "epoch": 0.594841579991583,
      "flos": 25848069949440.0,
      "grad_norm": 2.068380984528916,
      "language_loss": 0.79987949,
      "learning_rate": 1.4885093534932465e-06,
      "loss": 0.82195067,
      "num_input_tokens_seen": 106709705,
      "step": 4947,
      "time_per_iteration": 2.6365702152252197
    },
    {
      "auxiliary_loss_clip": 0.01184117,
      "auxiliary_loss_mlp": 0.01029315,
      "balance_loss_clip": 1.05254626,
      "balance_loss_mlp": 1.02079725,
      "epoch": 0.5949618228822221,
      "flos": 23981109672960.0,
      "grad_norm": 2.4568719607064766,
      "language_loss": 0.71030325,
      "learning_rate": 1.4877563259941433e-06,
      "loss": 0.73243761,
      "num_input_tokens_seen": 106727560,
      "step": 4948,
      "time_per_iteration": 2.6635539531707764
    },
    {
      "auxiliary_loss_clip": 0.0118716,
      "auxiliary_loss_mlp": 0.01029051,
      "balance_loss_clip": 1.05287755,
      "balance_loss_mlp": 1.02058089,
      "epoch": 0.5950820657728612,
      "flos": 40547491476480.0,
      "grad_norm": 1.7523912744564465,
      "language_loss": 0.67738748,
      "learning_rate": 1.4870033762019988e-06,
      "loss": 0.69954956,
      "num_input_tokens_seen": 106747725,
      "step": 4949,
      "time_per_iteration": 2.766244649887085
    },
    {
      "auxiliary_loss_clip": 0.01182536,
      "auxiliary_loss_mlp": 0.01026532,
      "balance_loss_clip": 1.05064178,
      "balance_loss_mlp": 1.01827109,
      "epoch": 0.5952023086635003,
      "flos": 23184467884800.0,
      "grad_norm": 1.7260784769544022,
      "language_loss": 0.73167175,
      "learning_rate": 1.4862505042310334e-06,
      "loss": 0.75376248,
      "num_input_tokens_seen": 106767010,
      "step": 4950,
      "time_per_iteration": 3.495488166809082
    },
    {
      "auxiliary_loss_clip": 0.01174267,
      "auxiliary_loss_mlp": 0.01032505,
      "balance_loss_clip": 1.04858816,
      "balance_loss_mlp": 1.02484596,
      "epoch": 0.5953225515541394,
      "flos": 33653289548160.0,
      "grad_norm": 1.5767153521017074,
      "language_loss": 0.69631457,
      "learning_rate": 1.4854977101954587e-06,
      "loss": 0.7183823,
      "num_input_tokens_seen": 106789230,
      "step": 4951,
      "time_per_iteration": 3.6582138538360596
    },
    {
      "auxiliary_loss_clip": 0.01179842,
      "auxiliary_loss_mlp": 0.01027343,
      "balance_loss_clip": 1.04758143,
      "balance_loss_mlp": 1.01914096,
      "epoch": 0.5954427944447784,
      "flos": 24459619680000.0,
      "grad_norm": 4.443025002421071,
      "language_loss": 0.86432052,
      "learning_rate": 1.4847449942094716e-06,
      "loss": 0.88639235,
      "num_input_tokens_seen": 106808110,
      "step": 4952,
      "time_per_iteration": 2.710343599319458
    },
    {
      "auxiliary_loss_clip": 0.0117798,
      "auxiliary_loss_mlp": 0.01024671,
      "balance_loss_clip": 1.04948115,
      "balance_loss_mlp": 1.01667237,
      "epoch": 0.5955630373354175,
      "flos": 18551848026240.0,
      "grad_norm": 1.8210184453256453,
      "language_loss": 0.8618257,
      "learning_rate": 1.4839923563872598e-06,
      "loss": 0.88385224,
      "num_input_tokens_seen": 106826650,
      "step": 4953,
      "time_per_iteration": 2.649195909500122
    },
    {
      "auxiliary_loss_clip": 0.01183576,
      "auxiliary_loss_mlp": 0.01024079,
      "balance_loss_clip": 1.04823864,
      "balance_loss_mlp": 1.0159018,
      "epoch": 0.5956832802260567,
      "flos": 19791699730560.0,
      "grad_norm": 1.6678865863805044,
      "language_loss": 0.76490921,
      "learning_rate": 1.483239796842997e-06,
      "loss": 0.78698575,
      "num_input_tokens_seen": 106844680,
      "step": 4954,
      "time_per_iteration": 2.629283905029297
    },
    {
      "auxiliary_loss_clip": 0.0118408,
      "auxiliary_loss_mlp": 0.01027322,
      "balance_loss_clip": 1.04798651,
      "balance_loss_mlp": 1.0194602,
      "epoch": 0.5958035231166957,
      "flos": 19750868945280.0,
      "grad_norm": 1.7395493765757972,
      "language_loss": 0.84115696,
      "learning_rate": 1.4824873156908462e-06,
      "loss": 0.86327094,
      "num_input_tokens_seen": 106862605,
      "step": 4955,
      "time_per_iteration": 2.6126105785369873
    },
    {
      "auxiliary_loss_clip": 0.0118053,
      "auxiliary_loss_mlp": 0.0120865,
      "balance_loss_clip": 1.05245793,
      "balance_loss_mlp": 1.00055492,
      "epoch": 0.5959237660073348,
      "flos": 21652806090240.0,
      "grad_norm": 1.5921835998046565,
      "language_loss": 0.75698155,
      "learning_rate": 1.4817349130449584e-06,
      "loss": 0.78087336,
      "num_input_tokens_seen": 106882325,
      "step": 4956,
      "time_per_iteration": 2.6292433738708496
    },
    {
      "auxiliary_loss_clip": 0.01177636,
      "auxiliary_loss_mlp": 0.0102695,
      "balance_loss_clip": 1.05084443,
      "balance_loss_mlp": 1.01911831,
      "epoch": 0.5960440088979739,
      "flos": 21171207513600.0,
      "grad_norm": 1.8274361037024973,
      "language_loss": 0.83302039,
      "learning_rate": 1.4809825890194717e-06,
      "loss": 0.85506618,
      "num_input_tokens_seen": 106900995,
      "step": 4957,
      "time_per_iteration": 2.59464955329895
    },
    {
      "auxiliary_loss_clip": 0.01175556,
      "auxiliary_loss_mlp": 0.01023216,
      "balance_loss_clip": 1.0459249,
      "balance_loss_mlp": 1.01515162,
      "epoch": 0.596164251788613,
      "flos": 14757526753920.0,
      "grad_norm": 1.7383887349795533,
      "language_loss": 0.77700299,
      "learning_rate": 1.4802303437285139e-06,
      "loss": 0.79899073,
      "num_input_tokens_seen": 106918265,
      "step": 4958,
      "time_per_iteration": 3.5308523178100586
    },
    {
      "auxiliary_loss_clip": 0.01180962,
      "auxiliary_loss_mlp": 0.01025719,
      "balance_loss_clip": 1.04782391,
      "balance_loss_mlp": 1.0174222,
      "epoch": 0.596284494679252,
      "flos": 20485924865280.0,
      "grad_norm": 2.680667252722123,
      "language_loss": 0.81091183,
      "learning_rate": 1.4794781772861994e-06,
      "loss": 0.83297861,
      "num_input_tokens_seen": 106934760,
      "step": 4959,
      "time_per_iteration": 2.6025969982147217
    },
    {
      "auxiliary_loss_clip": 0.0117979,
      "auxiliary_loss_mlp": 0.01208227,
      "balance_loss_clip": 1.04812765,
      "balance_loss_mlp": 1.00074172,
      "epoch": 0.5964047375698912,
      "flos": 31212262108800.0,
      "grad_norm": 3.14721683812741,
      "language_loss": 0.67087907,
      "learning_rate": 1.4787260898066324e-06,
      "loss": 0.69475925,
      "num_input_tokens_seen": 106954760,
      "step": 4960,
      "time_per_iteration": 2.7664191722869873
    },
    {
      "auxiliary_loss_clip": 0.01174832,
      "auxiliary_loss_mlp": 0.01024349,
      "balance_loss_clip": 1.05225348,
      "balance_loss_mlp": 1.0159266,
      "epoch": 0.5965249804605303,
      "flos": 27483620855040.0,
      "grad_norm": 2.362992223445766,
      "language_loss": 0.85838747,
      "learning_rate": 1.4779740814039023e-06,
      "loss": 0.88037926,
      "num_input_tokens_seen": 106974845,
      "step": 4961,
      "time_per_iteration": 2.6323788166046143
    },
    {
      "auxiliary_loss_clip": 0.01175216,
      "auxiliary_loss_mlp": 0.01030119,
      "balance_loss_clip": 1.05085993,
      "balance_loss_mlp": 1.02139246,
      "epoch": 0.5966452233511693,
      "flos": 30773936442240.0,
      "grad_norm": 1.9929157255334722,
      "language_loss": 0.69119036,
      "learning_rate": 1.4772221521920894e-06,
      "loss": 0.71324372,
      "num_input_tokens_seen": 106994870,
      "step": 4962,
      "time_per_iteration": 2.6594231128692627
    },
    {
      "auxiliary_loss_clip": 0.01181806,
      "auxiliary_loss_mlp": 0.01023874,
      "balance_loss_clip": 1.05113888,
      "balance_loss_mlp": 1.01604795,
      "epoch": 0.5967654662418085,
      "flos": 25481170477440.0,
      "grad_norm": 2.00284052468434,
      "language_loss": 0.74405932,
      "learning_rate": 1.4764703022852598e-06,
      "loss": 0.76611614,
      "num_input_tokens_seen": 107015390,
      "step": 4963,
      "time_per_iteration": 2.6801397800445557
    },
    {
      "auxiliary_loss_clip": 0.01175695,
      "auxiliary_loss_mlp": 0.01022475,
      "balance_loss_clip": 1.04160047,
      "balance_loss_mlp": 1.01474416,
      "epoch": 0.5968857091324475,
      "flos": 19099126621440.0,
      "grad_norm": 1.9124136713564521,
      "language_loss": 0.77090037,
      "learning_rate": 1.4757185317974696e-06,
      "loss": 0.79288208,
      "num_input_tokens_seen": 107033775,
      "step": 4964,
      "time_per_iteration": 2.619685173034668
    },
    {
      "auxiliary_loss_clip": 0.01182815,
      "auxiliary_loss_mlp": 0.01027858,
      "balance_loss_clip": 1.05218959,
      "balance_loss_mlp": 1.0197283,
      "epoch": 0.5970059520230866,
      "flos": 23692711374720.0,
      "grad_norm": 2.6550304285824535,
      "language_loss": 0.70875919,
      "learning_rate": 1.474966840842761e-06,
      "loss": 0.73086596,
      "num_input_tokens_seen": 107053355,
      "step": 4965,
      "time_per_iteration": 2.715122699737549
    },
    {
      "auxiliary_loss_clip": 0.01183148,
      "auxiliary_loss_mlp": 0.01024583,
      "balance_loss_clip": 1.05127513,
      "balance_loss_mlp": 1.01689386,
      "epoch": 0.5971261949137258,
      "flos": 23185545292800.0,
      "grad_norm": 1.8464619386883268,
      "language_loss": 0.86925626,
      "learning_rate": 1.4742152295351655e-06,
      "loss": 0.89133358,
      "num_input_tokens_seen": 107072510,
      "step": 4966,
      "time_per_iteration": 2.605334758758545
    },
    {
      "auxiliary_loss_clip": 0.01181795,
      "auxiliary_loss_mlp": 0.01209037,
      "balance_loss_clip": 1.05137968,
      "balance_loss_mlp": 1.00077248,
      "epoch": 0.5972464378043648,
      "flos": 20557710195840.0,
      "grad_norm": 6.103286931298781,
      "language_loss": 0.63990641,
      "learning_rate": 1.4734636979887016e-06,
      "loss": 0.66381472,
      "num_input_tokens_seen": 107089970,
      "step": 4967,
      "time_per_iteration": 2.6574513912200928
    },
    {
      "auxiliary_loss_clip": 0.01187673,
      "auxiliary_loss_mlp": 0.01029347,
      "balance_loss_clip": 1.04712987,
      "balance_loss_mlp": 1.02122855,
      "epoch": 0.5973666806950039,
      "flos": 29387030457600.0,
      "grad_norm": 2.4550983859291566,
      "language_loss": 0.90470099,
      "learning_rate": 1.4727122463173755e-06,
      "loss": 0.92687118,
      "num_input_tokens_seen": 107108500,
      "step": 4968,
      "time_per_iteration": 2.6969754695892334
    },
    {
      "auxiliary_loss_clip": 0.01183736,
      "auxiliary_loss_mlp": 0.01026812,
      "balance_loss_clip": 1.05106652,
      "balance_loss_mlp": 1.01858079,
      "epoch": 0.597486923585643,
      "flos": 22273522041600.0,
      "grad_norm": 3.226962866108451,
      "language_loss": 0.64441609,
      "learning_rate": 1.471960874635183e-06,
      "loss": 0.66652155,
      "num_input_tokens_seen": 107128060,
      "step": 4969,
      "time_per_iteration": 2.603088855743408
    },
    {
      "auxiliary_loss_clip": 0.01178843,
      "auxiliary_loss_mlp": 0.01025753,
      "balance_loss_clip": 1.047351,
      "balance_loss_mlp": 1.01715755,
      "epoch": 0.5976071664762821,
      "flos": 13772461196160.0,
      "grad_norm": 3.5063545566282337,
      "language_loss": 0.71212566,
      "learning_rate": 1.4712095830561055e-06,
      "loss": 0.73417163,
      "num_input_tokens_seen": 107146550,
      "step": 4970,
      "time_per_iteration": 2.560307741165161
    },
    {
      "auxiliary_loss_clip": 0.01181252,
      "auxiliary_loss_mlp": 0.01026863,
      "balance_loss_clip": 1.04718399,
      "balance_loss_mlp": 1.01904345,
      "epoch": 0.5977274093669211,
      "flos": 19098623831040.0,
      "grad_norm": 1.915003049208897,
      "language_loss": 0.80878884,
      "learning_rate": 1.4704583716941147e-06,
      "loss": 0.83086991,
      "num_input_tokens_seen": 107165415,
      "step": 4971,
      "time_per_iteration": 2.6494758129119873
    },
    {
      "auxiliary_loss_clip": 0.01174009,
      "auxiliary_loss_mlp": 0.01032875,
      "balance_loss_clip": 1.05200922,
      "balance_loss_mlp": 1.02500772,
      "epoch": 0.5978476522575603,
      "flos": 20376002269440.0,
      "grad_norm": 1.7440016527137634,
      "language_loss": 0.72436041,
      "learning_rate": 1.4697072406631672e-06,
      "loss": 0.74642932,
      "num_input_tokens_seen": 107185320,
      "step": 4972,
      "time_per_iteration": 2.612783908843994
    },
    {
      "auxiliary_loss_clip": 0.01190748,
      "auxiliary_loss_mlp": 0.01028227,
      "balance_loss_clip": 1.04922771,
      "balance_loss_mlp": 1.01914954,
      "epoch": 0.5979678951481994,
      "flos": 29023147728000.0,
      "grad_norm": 1.7160359276205832,
      "language_loss": 0.72660834,
      "learning_rate": 1.4689561900772097e-06,
      "loss": 0.74879813,
      "num_input_tokens_seen": 107205380,
      "step": 4973,
      "time_per_iteration": 3.747405529022217
    },
    {
      "auxiliary_loss_clip": 0.01181038,
      "auxiliary_loss_mlp": 0.01028432,
      "balance_loss_clip": 1.04769564,
      "balance_loss_mlp": 1.02080894,
      "epoch": 0.5980881380388384,
      "flos": 17967689141760.0,
      "grad_norm": 2.2923663248831416,
      "language_loss": 0.72142822,
      "learning_rate": 1.4682052200501758e-06,
      "loss": 0.74352288,
      "num_input_tokens_seen": 107222585,
      "step": 4974,
      "time_per_iteration": 2.623854398727417
    },
    {
      "auxiliary_loss_clip": 0.01175664,
      "auxiliary_loss_mlp": 0.01028587,
      "balance_loss_clip": 1.05133045,
      "balance_loss_mlp": 1.01981378,
      "epoch": 0.5982083809294776,
      "flos": 22962827013120.0,
      "grad_norm": 1.8744640743258638,
      "language_loss": 0.80160511,
      "learning_rate": 1.4674543306959876e-06,
      "loss": 0.82364762,
      "num_input_tokens_seen": 107242055,
      "step": 4975,
      "time_per_iteration": 2.6170923709869385
    },
    {
      "auxiliary_loss_clip": 0.01188448,
      "auxiliary_loss_mlp": 0.01028069,
      "balance_loss_clip": 1.05177498,
      "balance_loss_mlp": 1.01967049,
      "epoch": 0.5983286238201166,
      "flos": 20991941712000.0,
      "grad_norm": 3.537816380731132,
      "language_loss": 0.84730923,
      "learning_rate": 1.4667035221285535e-06,
      "loss": 0.86947441,
      "num_input_tokens_seen": 107259695,
      "step": 4976,
      "time_per_iteration": 2.797485113143921
    },
    {
      "auxiliary_loss_clip": 0.01177304,
      "auxiliary_loss_mlp": 0.01024037,
      "balance_loss_clip": 1.05199587,
      "balance_loss_mlp": 1.01632452,
      "epoch": 0.5984488667107557,
      "flos": 28183448511360.0,
      "grad_norm": 1.8479653079535316,
      "language_loss": 0.7419908,
      "learning_rate": 1.4659527944617715e-06,
      "loss": 0.76400429,
      "num_input_tokens_seen": 107279640,
      "step": 4977,
      "time_per_iteration": 3.697561502456665
    },
    {
      "auxiliary_loss_clip": 0.01185236,
      "auxiliary_loss_mlp": 0.01027101,
      "balance_loss_clip": 1.04220974,
      "balance_loss_mlp": 1.01881576,
      "epoch": 0.5985691096013949,
      "flos": 16471794314880.0,
      "grad_norm": 2.846710238204654,
      "language_loss": 0.76013172,
      "learning_rate": 1.465202147809526e-06,
      "loss": 0.78225505,
      "num_input_tokens_seen": 107298135,
      "step": 4978,
      "time_per_iteration": 3.6110286712646484
    },
    {
      "auxiliary_loss_clip": 0.01177867,
      "auxiliary_loss_mlp": 0.01025671,
      "balance_loss_clip": 1.05353808,
      "balance_loss_mlp": 1.01810694,
      "epoch": 0.5986893524920339,
      "flos": 26719046933760.0,
      "grad_norm": 1.915184394924195,
      "language_loss": 0.76361692,
      "learning_rate": 1.4644515822856888e-06,
      "loss": 0.78565234,
      "num_input_tokens_seen": 107316570,
      "step": 4979,
      "time_per_iteration": 2.664240598678589
    },
    {
      "auxiliary_loss_clip": 0.01092116,
      "auxiliary_loss_mlp": 0.01002056,
      "balance_loss_clip": 1.01552963,
      "balance_loss_mlp": 1.0012455,
      "epoch": 0.598809595382673,
      "flos": 61608061100160.0,
      "grad_norm": 0.7636775425101704,
      "language_loss": 0.56481183,
      "learning_rate": 1.4637010980041215e-06,
      "loss": 0.58575356,
      "num_input_tokens_seen": 107378680,
      "step": 4980,
      "time_per_iteration": 3.220845937728882
    },
    {
      "auxiliary_loss_clip": 0.01179741,
      "auxiliary_loss_mlp": 0.01026303,
      "balance_loss_clip": 1.05288434,
      "balance_loss_mlp": 1.01800013,
      "epoch": 0.5989298382733121,
      "flos": 11801719549440.0,
      "grad_norm": 2.215373371419951,
      "language_loss": 0.89617658,
      "learning_rate": 1.4629506950786707e-06,
      "loss": 0.91823709,
      "num_input_tokens_seen": 107394860,
      "step": 4981,
      "time_per_iteration": 2.5496089458465576
    },
    {
      "auxiliary_loss_clip": 0.01073118,
      "auxiliary_loss_mlp": 0.01000479,
      "balance_loss_clip": 1.01582456,
      "balance_loss_mlp": 0.99969178,
      "epoch": 0.5990500811639512,
      "flos": 60025800021120.0,
      "grad_norm": 0.8137203596885144,
      "language_loss": 0.56004578,
      "learning_rate": 1.4622003736231733e-06,
      "loss": 0.5807817,
      "num_input_tokens_seen": 107453850,
      "step": 4982,
      "time_per_iteration": 3.2064766883850098
    },
    {
      "auxiliary_loss_clip": 0.01175934,
      "auxiliary_loss_mlp": 0.01026568,
      "balance_loss_clip": 1.04965639,
      "balance_loss_mlp": 1.01808596,
      "epoch": 0.5991703240545903,
      "flos": 18222726683520.0,
      "grad_norm": 1.9000331329982632,
      "language_loss": 0.805076,
      "learning_rate": 1.461450133751451e-06,
      "loss": 0.82710099,
      "num_input_tokens_seen": 107471920,
      "step": 4983,
      "time_per_iteration": 2.62302303314209
    },
    {
      "auxiliary_loss_clip": 0.01184169,
      "auxiliary_loss_mlp": 0.01025171,
      "balance_loss_clip": 1.05271769,
      "balance_loss_mlp": 1.01681435,
      "epoch": 0.5992905669452293,
      "flos": 27709894581120.0,
      "grad_norm": 1.863910271875842,
      "language_loss": 0.76093233,
      "learning_rate": 1.4606999755773153e-06,
      "loss": 0.78302574,
      "num_input_tokens_seen": 107493125,
      "step": 4984,
      "time_per_iteration": 2.686176300048828
    },
    {
      "auxiliary_loss_clip": 0.01175625,
      "auxiliary_loss_mlp": 0.01023942,
      "balance_loss_clip": 1.05196881,
      "balance_loss_mlp": 1.01581764,
      "epoch": 0.5994108098358685,
      "flos": 20449008662400.0,
      "grad_norm": 2.065572822817169,
      "language_loss": 0.82279456,
      "learning_rate": 1.4599498992145643e-06,
      "loss": 0.84479022,
      "num_input_tokens_seen": 107513150,
      "step": 4985,
      "time_per_iteration": 3.56836199760437
    },
    {
      "auxiliary_loss_clip": 0.01188655,
      "auxiliary_loss_mlp": 0.01208115,
      "balance_loss_clip": 1.05143166,
      "balance_loss_mlp": 1.00075245,
      "epoch": 0.5995310527265075,
      "flos": 22269966595200.0,
      "grad_norm": 2.157695658485529,
      "language_loss": 0.71100998,
      "learning_rate": 1.4591999047769846e-06,
      "loss": 0.73497766,
      "num_input_tokens_seen": 107532005,
      "step": 4986,
      "time_per_iteration": 2.6897735595703125
    },
    {
      "auxiliary_loss_clip": 0.01185589,
      "auxiliary_loss_mlp": 0.01029805,
      "balance_loss_clip": 1.04151535,
      "balance_loss_mlp": 1.02131104,
      "epoch": 0.5996512956171466,
      "flos": 18916951818240.0,
      "grad_norm": 1.6800574785646019,
      "language_loss": 0.75275028,
      "learning_rate": 1.4584499923783486e-06,
      "loss": 0.77490419,
      "num_input_tokens_seen": 107550585,
      "step": 4987,
      "time_per_iteration": 2.6709868907928467
    },
    {
      "auxiliary_loss_clip": 0.01180416,
      "auxiliary_loss_mlp": 0.01022391,
      "balance_loss_clip": 1.04911911,
      "balance_loss_mlp": 1.01476717,
      "epoch": 0.5997715385077858,
      "flos": 15370916330880.0,
      "grad_norm": 2.062815622343247,
      "language_loss": 0.7594589,
      "learning_rate": 1.457700162132419e-06,
      "loss": 0.78148699,
      "num_input_tokens_seen": 107567575,
      "step": 4988,
      "time_per_iteration": 2.604248046875
    },
    {
      "auxiliary_loss_clip": 0.01183935,
      "auxiliary_loss_mlp": 0.01023534,
      "balance_loss_clip": 1.04680681,
      "balance_loss_mlp": 1.01558924,
      "epoch": 0.5998917813984248,
      "flos": 25264844818560.0,
      "grad_norm": 2.3469767451166907,
      "language_loss": 0.72162402,
      "learning_rate": 1.4569504141529433e-06,
      "loss": 0.74369872,
      "num_input_tokens_seen": 107585410,
      "step": 4989,
      "time_per_iteration": 2.646576166152954
    },
    {
      "auxiliary_loss_clip": 0.01178418,
      "auxiliary_loss_mlp": 0.01028966,
      "balance_loss_clip": 1.05238068,
      "balance_loss_mlp": 1.02047777,
      "epoch": 0.6000120242890639,
      "flos": 22054502862720.0,
      "grad_norm": 2.097822743083971,
      "language_loss": 0.72397822,
      "learning_rate": 1.456200748553658e-06,
      "loss": 0.74605203,
      "num_input_tokens_seen": 107603405,
      "step": 4990,
      "time_per_iteration": 2.6309194564819336
    },
    {
      "auxiliary_loss_clip": 0.01179332,
      "auxiliary_loss_mlp": 0.01030214,
      "balance_loss_clip": 1.05317998,
      "balance_loss_mlp": 1.0216372,
      "epoch": 0.600132267179703,
      "flos": 29863421562240.0,
      "grad_norm": 1.5314099792273388,
      "language_loss": 0.78607881,
      "learning_rate": 1.455451165448287e-06,
      "loss": 0.80817425,
      "num_input_tokens_seen": 107626060,
      "step": 4991,
      "time_per_iteration": 2.6421101093292236
    },
    {
      "auxiliary_loss_clip": 0.01178966,
      "auxiliary_loss_mlp": 0.01025511,
      "balance_loss_clip": 1.05037498,
      "balance_loss_mlp": 1.01723218,
      "epoch": 0.6002525100703421,
      "flos": 25045358762880.0,
      "grad_norm": 2.416096131402006,
      "language_loss": 0.73963308,
      "learning_rate": 1.4547016649505407e-06,
      "loss": 0.76167786,
      "num_input_tokens_seen": 107644070,
      "step": 4992,
      "time_per_iteration": 2.713528633117676
    },
    {
      "auxiliary_loss_clip": 0.01183297,
      "auxiliary_loss_mlp": 0.01024622,
      "balance_loss_clip": 1.04644084,
      "balance_loss_mlp": 1.01620579,
      "epoch": 0.6003727529609811,
      "flos": 20849592113280.0,
      "grad_norm": 2.2172705780822284,
      "language_loss": 0.84938139,
      "learning_rate": 1.4539522471741193e-06,
      "loss": 0.87146056,
      "num_input_tokens_seen": 107661495,
      "step": 4993,
      "time_per_iteration": 2.6574769020080566
    },
    {
      "auxiliary_loss_clip": 0.01183149,
      "auxiliary_loss_mlp": 0.0102897,
      "balance_loss_clip": 1.05000842,
      "balance_loss_mlp": 1.02062511,
      "epoch": 0.6004929958516203,
      "flos": 15594604277760.0,
      "grad_norm": 2.196456484967823,
      "language_loss": 0.70760179,
      "learning_rate": 1.4532029122327067e-06,
      "loss": 0.72972298,
      "num_input_tokens_seen": 107678280,
      "step": 4994,
      "time_per_iteration": 2.7086181640625
    },
    {
      "auxiliary_loss_clip": 0.01176492,
      "auxiliary_loss_mlp": 0.01026604,
      "balance_loss_clip": 1.04865909,
      "balance_loss_mlp": 1.01884365,
      "epoch": 0.6006132387422594,
      "flos": 21763267390080.0,
      "grad_norm": 13.147811502165993,
      "language_loss": 0.75425601,
      "learning_rate": 1.4524536602399783e-06,
      "loss": 0.77628702,
      "num_input_tokens_seen": 107697370,
      "step": 4995,
      "time_per_iteration": 2.81860089302063
    },
    {
      "auxiliary_loss_clip": 0.01177485,
      "auxiliary_loss_mlp": 0.01032485,
      "balance_loss_clip": 1.05036879,
      "balance_loss_mlp": 1.02477801,
      "epoch": 0.6007334816328984,
      "flos": 22858542852480.0,
      "grad_norm": 1.6440523422170061,
      "language_loss": 0.77377999,
      "learning_rate": 1.4517044913095938e-06,
      "loss": 0.79587972,
      "num_input_tokens_seen": 107717790,
      "step": 4996,
      "time_per_iteration": 2.7014880180358887
    },
    {
      "auxiliary_loss_clip": 0.01180846,
      "auxiliary_loss_mlp": 0.01025708,
      "balance_loss_clip": 1.05128634,
      "balance_loss_mlp": 1.01713657,
      "epoch": 0.6008537245235376,
      "flos": 28324577047680.0,
      "grad_norm": 1.6991083774340525,
      "language_loss": 0.81491864,
      "learning_rate": 1.4509554055552022e-06,
      "loss": 0.83698416,
      "num_input_tokens_seen": 107738020,
      "step": 4997,
      "time_per_iteration": 2.6088316440582275
    },
    {
      "auxiliary_loss_clip": 0.01178835,
      "auxiliary_loss_mlp": 0.01028404,
      "balance_loss_clip": 1.04829931,
      "balance_loss_mlp": 1.02026165,
      "epoch": 0.6009739674141766,
      "flos": 20886113266560.0,
      "grad_norm": 2.782053732109489,
      "language_loss": 0.84199822,
      "learning_rate": 1.450206403090439e-06,
      "loss": 0.86407059,
      "num_input_tokens_seen": 107756215,
      "step": 4998,
      "time_per_iteration": 2.627336263656616
    },
    {
      "auxiliary_loss_clip": 0.01178596,
      "auxiliary_loss_mlp": 0.01022903,
      "balance_loss_clip": 1.05122793,
      "balance_loss_mlp": 1.01492214,
      "epoch": 0.6010942103048157,
      "flos": 20481004702080.0,
      "grad_norm": 1.8950136482179551,
      "language_loss": 0.86602104,
      "learning_rate": 1.4494574840289274e-06,
      "loss": 0.88803601,
      "num_input_tokens_seen": 107773330,
      "step": 4999,
      "time_per_iteration": 3.59110426902771
    },
    {
      "auxiliary_loss_clip": 0.01183814,
      "auxiliary_loss_mlp": 0.01032163,
      "balance_loss_clip": 1.05012536,
      "balance_loss_mlp": 1.02337778,
      "epoch": 0.6012144531954549,
      "flos": 23805973935360.0,
      "grad_norm": 1.6830961706058958,
      "language_loss": 0.74309361,
      "learning_rate": 1.4487086484842782e-06,
      "loss": 0.76525342,
      "num_input_tokens_seen": 107791975,
      "step": 5000,
      "time_per_iteration": 2.586566209793091
    },
    {
      "auxiliary_loss_clip": 0.01172975,
      "auxiliary_loss_mlp": 0.01025881,
      "balance_loss_clip": 1.04990411,
      "balance_loss_mlp": 1.01814127,
      "epoch": 0.6013346960860939,
      "flos": 18988378012800.0,
      "grad_norm": 1.9661597069882308,
      "language_loss": 0.60028279,
      "learning_rate": 1.4479598965700878e-06,
      "loss": 0.62227136,
      "num_input_tokens_seen": 107809240,
      "step": 5001,
      "time_per_iteration": 2.5288796424865723
    },
    {
      "auxiliary_loss_clip": 0.01182095,
      "auxiliary_loss_mlp": 0.01028331,
      "balance_loss_clip": 1.04525232,
      "balance_loss_mlp": 1.01983762,
      "epoch": 0.601454938976733,
      "flos": 24025316336640.0,
      "grad_norm": 2.704283381746668,
      "language_loss": 0.68695444,
      "learning_rate": 1.4472112283999427e-06,
      "loss": 0.7090587,
      "num_input_tokens_seen": 107827895,
      "step": 5002,
      "time_per_iteration": 2.6620850563049316
    },
    {
      "auxiliary_loss_clip": 0.01172783,
      "auxiliary_loss_mlp": 0.01024889,
      "balance_loss_clip": 1.05023503,
      "balance_loss_mlp": 1.01734924,
      "epoch": 0.6015751818673721,
      "flos": 26427129102720.0,
      "grad_norm": 2.0825878886080926,
      "language_loss": 0.69297099,
      "learning_rate": 1.4464626440874143e-06,
      "loss": 0.7149477,
      "num_input_tokens_seen": 107847010,
      "step": 5003,
      "time_per_iteration": 3.5781378746032715
    },
    {
      "auxiliary_loss_clip": 0.01194363,
      "auxiliary_loss_mlp": 0.01027335,
      "balance_loss_clip": 1.0437634,
      "balance_loss_mlp": 1.01870418,
      "epoch": 0.6016954247580112,
      "flos": 13115260005120.0,
      "grad_norm": 2.366119422401922,
      "language_loss": 0.73778832,
      "learning_rate": 1.4457141437460636e-06,
      "loss": 0.7600053,
      "num_input_tokens_seen": 107864235,
      "step": 5004,
      "time_per_iteration": 3.6008071899414062
    },
    {
      "auxiliary_loss_clip": 0.01184677,
      "auxiliary_loss_mlp": 0.01027318,
      "balance_loss_clip": 1.04913282,
      "balance_loss_mlp": 1.01889622,
      "epoch": 0.6018156676486502,
      "flos": 23768447201280.0,
      "grad_norm": 2.0760026005522674,
      "language_loss": 0.73013157,
      "learning_rate": 1.444965727489436e-06,
      "loss": 0.75225151,
      "num_input_tokens_seen": 107883680,
      "step": 5005,
      "time_per_iteration": 2.6136794090270996
    },
    {
      "auxiliary_loss_clip": 0.01180719,
      "auxiliary_loss_mlp": 0.0102766,
      "balance_loss_clip": 1.04430842,
      "balance_loss_mlp": 1.0195837,
      "epoch": 0.6019359105392894,
      "flos": 26469360518400.0,
      "grad_norm": 1.7664653002075508,
      "language_loss": 0.63392627,
      "learning_rate": 1.444217395431066e-06,
      "loss": 0.65601003,
      "num_input_tokens_seen": 107906220,
      "step": 5006,
      "time_per_iteration": 2.740549325942993
    },
    {
      "auxiliary_loss_clip": 0.01093022,
      "auxiliary_loss_mlp": 0.01001182,
      "balance_loss_clip": 1.02117109,
      "balance_loss_mlp": 1.00037706,
      "epoch": 0.6020561534299285,
      "flos": 69190849728000.0,
      "grad_norm": 0.8161926651953684,
      "language_loss": 0.55799496,
      "learning_rate": 1.4434691476844755e-06,
      "loss": 0.57893699,
      "num_input_tokens_seen": 107967195,
      "step": 5007,
      "time_per_iteration": 3.119879961013794
    },
    {
      "auxiliary_loss_clip": 0.01176167,
      "auxiliary_loss_mlp": 0.01023315,
      "balance_loss_clip": 1.05002284,
      "balance_loss_mlp": 1.01627588,
      "epoch": 0.6021763963205675,
      "flos": 21835304115840.0,
      "grad_norm": 3.5100871756693066,
      "language_loss": 0.67267179,
      "learning_rate": 1.4427209843631729e-06,
      "loss": 0.69466662,
      "num_input_tokens_seen": 107984245,
      "step": 5008,
      "time_per_iteration": 2.6483426094055176
    },
    {
      "auxiliary_loss_clip": 0.01174811,
      "auxiliary_loss_mlp": 0.01208042,
      "balance_loss_clip": 1.05261445,
      "balance_loss_mlp": 1.0007993,
      "epoch": 0.6022966392112067,
      "flos": 26578636669440.0,
      "grad_norm": 1.7592523975633256,
      "language_loss": 0.80852038,
      "learning_rate": 1.4419729055806534e-06,
      "loss": 0.83234894,
      "num_input_tokens_seen": 108003680,
      "step": 5009,
      "time_per_iteration": 2.6400060653686523
    },
    {
      "auxiliary_loss_clip": 0.0117908,
      "auxiliary_loss_mlp": 0.01207509,
      "balance_loss_clip": 1.05202413,
      "balance_loss_mlp": 1.00062418,
      "epoch": 0.6024168821018457,
      "flos": 20703722981760.0,
      "grad_norm": 1.7009427851901657,
      "language_loss": 0.82601261,
      "learning_rate": 1.441224911450401e-06,
      "loss": 0.84987855,
      "num_input_tokens_seen": 108019635,
      "step": 5010,
      "time_per_iteration": 2.6097874641418457
    },
    {
      "auxiliary_loss_clip": 0.01183443,
      "auxiliary_loss_mlp": 0.01032714,
      "balance_loss_clip": 1.05070162,
      "balance_loss_mlp": 1.02435124,
      "epoch": 0.6025371249924848,
      "flos": 24680973242880.0,
      "grad_norm": 1.7907219565065395,
      "language_loss": 0.82313997,
      "learning_rate": 1.4404770020858851e-06,
      "loss": 0.84530151,
      "num_input_tokens_seen": 108039120,
      "step": 5011,
      "time_per_iteration": 3.496891498565674
    },
    {
      "auxiliary_loss_clip": 0.01170931,
      "auxiliary_loss_mlp": 0.01026658,
      "balance_loss_clip": 1.04819775,
      "balance_loss_mlp": 1.01898694,
      "epoch": 0.602657367883124,
      "flos": 25955801815680.0,
      "grad_norm": 1.6832422724638818,
      "language_loss": 0.86008191,
      "learning_rate": 1.439729177600563e-06,
      "loss": 0.88205779,
      "num_input_tokens_seen": 108059615,
      "step": 5012,
      "time_per_iteration": 2.626208543777466
    },
    {
      "auxiliary_loss_clip": 0.01177653,
      "auxiliary_loss_mlp": 0.01025341,
      "balance_loss_clip": 1.05170012,
      "balance_loss_mlp": 1.01740742,
      "epoch": 0.602777610773763,
      "flos": 16690633925760.0,
      "grad_norm": 6.831702838780706,
      "language_loss": 0.73699087,
      "learning_rate": 1.4389814381078793e-06,
      "loss": 0.75902081,
      "num_input_tokens_seen": 108078855,
      "step": 5013,
      "time_per_iteration": 2.6833560466766357
    },
    {
      "auxiliary_loss_clip": 0.01195076,
      "auxiliary_loss_mlp": 0.01034928,
      "balance_loss_clip": 1.0404222,
      "balance_loss_mlp": 1.02764487,
      "epoch": 0.6028978536644021,
      "flos": 13334243270400.0,
      "grad_norm": 3.2955625806069926,
      "language_loss": 0.80023128,
      "learning_rate": 1.438233783721265e-06,
      "loss": 0.82253128,
      "num_input_tokens_seen": 108095020,
      "step": 5014,
      "time_per_iteration": 2.686105728149414
    },
    {
      "auxiliary_loss_clip": 0.01176653,
      "auxiliary_loss_mlp": 0.01027752,
      "balance_loss_clip": 1.05115366,
      "balance_loss_mlp": 1.02006936,
      "epoch": 0.6030180965550412,
      "flos": 19644825018240.0,
      "grad_norm": 2.389511526304552,
      "language_loss": 0.78349602,
      "learning_rate": 1.43748621455414e-06,
      "loss": 0.80554008,
      "num_input_tokens_seen": 108111455,
      "step": 5015,
      "time_per_iteration": 2.606887102127075
    },
    {
      "auxiliary_loss_clip": 0.0117585,
      "auxiliary_loss_mlp": 0.01025864,
      "balance_loss_clip": 1.04796243,
      "balance_loss_mlp": 1.01770449,
      "epoch": 0.6031383394456803,
      "flos": 14458390289280.0,
      "grad_norm": 2.426741825576268,
      "language_loss": 0.81123364,
      "learning_rate": 1.4367387307199082e-06,
      "loss": 0.83325076,
      "num_input_tokens_seen": 108128305,
      "step": 5016,
      "time_per_iteration": 2.5751423835754395
    },
    {
      "auxiliary_loss_clip": 0.01174526,
      "auxiliary_loss_mlp": 0.01029052,
      "balance_loss_clip": 1.04898095,
      "balance_loss_mlp": 1.02127337,
      "epoch": 0.6032585823363193,
      "flos": 13917791623680.0,
      "grad_norm": 2.1892232531976887,
      "language_loss": 0.8233285,
      "learning_rate": 1.4359913323319632e-06,
      "loss": 0.84536433,
      "num_input_tokens_seen": 108145475,
      "step": 5017,
      "time_per_iteration": 2.6281275749206543
    },
    {
      "auxiliary_loss_clip": 0.01177836,
      "auxiliary_loss_mlp": 0.01027144,
      "balance_loss_clip": 1.0425663,
      "balance_loss_mlp": 1.01907396,
      "epoch": 0.6033788252269584,
      "flos": 24353252530560.0,
      "grad_norm": 1.8550407762929602,
      "language_loss": 0.77459228,
      "learning_rate": 1.4352440195036847e-06,
      "loss": 0.79664212,
      "num_input_tokens_seen": 108165650,
      "step": 5018,
      "time_per_iteration": 2.782928466796875
    },
    {
      "auxiliary_loss_clip": 0.0119686,
      "auxiliary_loss_mlp": 0.01024112,
      "balance_loss_clip": 1.04077089,
      "balance_loss_mlp": 1.01608324,
      "epoch": 0.6034990681175976,
      "flos": 25521247077120.0,
      "grad_norm": 1.9483611157708531,
      "language_loss": 0.7994442,
      "learning_rate": 1.4344967923484395e-06,
      "loss": 0.8216539,
      "num_input_tokens_seen": 108187620,
      "step": 5019,
      "time_per_iteration": 2.787440299987793
    },
    {
      "auxiliary_loss_clip": 0.01176985,
      "auxiliary_loss_mlp": 0.0102996,
      "balance_loss_clip": 1.050565,
      "balance_loss_mlp": 1.0221045,
      "epoch": 0.6036193110082366,
      "flos": 25958387594880.0,
      "grad_norm": 2.088780046146332,
      "language_loss": 0.72374308,
      "learning_rate": 1.433749650979581e-06,
      "loss": 0.74581254,
      "num_input_tokens_seen": 108207605,
      "step": 5020,
      "time_per_iteration": 2.6041297912597656
    },
    {
      "auxiliary_loss_clip": 0.01188463,
      "auxiliary_loss_mlp": 0.01024399,
      "balance_loss_clip": 1.04646492,
      "balance_loss_mlp": 1.01665342,
      "epoch": 0.6037395538988757,
      "flos": 25593427457280.0,
      "grad_norm": 2.900641692513554,
      "language_loss": 0.68494016,
      "learning_rate": 1.433002595510451e-06,
      "loss": 0.7070688,
      "num_input_tokens_seen": 108226385,
      "step": 5021,
      "time_per_iteration": 2.7125911712646484
    },
    {
      "auxiliary_loss_clip": 0.01177135,
      "auxiliary_loss_mlp": 0.01208457,
      "balance_loss_clip": 1.04681611,
      "balance_loss_mlp": 1.0007441,
      "epoch": 0.6038597967895148,
      "flos": 17816253402240.0,
      "grad_norm": 2.2141844430250477,
      "language_loss": 0.71998405,
      "learning_rate": 1.4322556260543757e-06,
      "loss": 0.74383998,
      "num_input_tokens_seen": 108242960,
      "step": 5022,
      "time_per_iteration": 2.6049344539642334
    },
    {
      "auxiliary_loss_clip": 0.01096808,
      "auxiliary_loss_mlp": 0.01004251,
      "balance_loss_clip": 1.01902652,
      "balance_loss_mlp": 1.0034523,
      "epoch": 0.6039800396801539,
      "flos": 65169213235200.0,
      "grad_norm": 0.8943770554771915,
      "language_loss": 0.62717152,
      "learning_rate": 1.4315087427246703e-06,
      "loss": 0.64818203,
      "num_input_tokens_seen": 108296785,
      "step": 5023,
      "time_per_iteration": 3.089792490005493
    },
    {
      "auxiliary_loss_clip": 0.01074817,
      "auxiliary_loss_mlp": 0.01003252,
      "balance_loss_clip": 1.01759028,
      "balance_loss_mlp": 1.00248933,
      "epoch": 0.604100282570793,
      "flos": 67386409073280.0,
      "grad_norm": 0.8751392350107458,
      "language_loss": 0.5849545,
      "learning_rate": 1.4307619456346372e-06,
      "loss": 0.60573518,
      "num_input_tokens_seen": 108341090,
      "step": 5024,
      "time_per_iteration": 2.8619325160980225
    },
    {
      "auxiliary_loss_clip": 0.01181799,
      "auxiliary_loss_mlp": 0.01024562,
      "balance_loss_clip": 1.0489918,
      "balance_loss_mlp": 1.01611054,
      "epoch": 0.6042205254614321,
      "flos": 35297495631360.0,
      "grad_norm": 2.793275606776947,
      "language_loss": 0.74709678,
      "learning_rate": 1.430015234897564e-06,
      "loss": 0.76916039,
      "num_input_tokens_seen": 108364370,
      "step": 5025,
      "time_per_iteration": 2.689180850982666
    },
    {
      "auxiliary_loss_clip": 0.01176569,
      "auxiliary_loss_mlp": 0.01208437,
      "balance_loss_clip": 1.05210948,
      "balance_loss_mlp": 1.00073862,
      "epoch": 0.6043407683520712,
      "flos": 45658262206080.0,
      "grad_norm": 1.6322867404160877,
      "language_loss": 0.66704667,
      "learning_rate": 1.4292686106267274e-06,
      "loss": 0.69089663,
      "num_input_tokens_seen": 108387220,
      "step": 5026,
      "time_per_iteration": 3.7222025394439697
    },
    {
      "auxiliary_loss_clip": 0.01183877,
      "auxiliary_loss_mlp": 0.01032763,
      "balance_loss_clip": 1.05152059,
      "balance_loss_mlp": 1.02423358,
      "epoch": 0.6044610112427102,
      "flos": 16180020138240.0,
      "grad_norm": 3.916768323075943,
      "language_loss": 0.77141345,
      "learning_rate": 1.4285220729353876e-06,
      "loss": 0.79357982,
      "num_input_tokens_seen": 108405760,
      "step": 5027,
      "time_per_iteration": 2.570478916168213
    },
    {
      "auxiliary_loss_clip": 0.0117775,
      "auxiliary_loss_mlp": 0.01027348,
      "balance_loss_clip": 1.04554927,
      "balance_loss_mlp": 1.01945662,
      "epoch": 0.6045812541333494,
      "flos": 13804062186240.0,
      "grad_norm": 1.8452001326400866,
      "language_loss": 0.77900535,
      "learning_rate": 1.4277756219367957e-06,
      "loss": 0.80105639,
      "num_input_tokens_seen": 108422785,
      "step": 5028,
      "time_per_iteration": 2.6257829666137695
    },
    {
      "auxiliary_loss_clip": 0.01193059,
      "auxiliary_loss_mlp": 0.01026357,
      "balance_loss_clip": 1.04934609,
      "balance_loss_mlp": 1.01798844,
      "epoch": 0.6047014970239885,
      "flos": 19975059682560.0,
      "grad_norm": 2.561008600025775,
      "language_loss": 0.79925919,
      "learning_rate": 1.4270292577441864e-06,
      "loss": 0.82145333,
      "num_input_tokens_seen": 108442290,
      "step": 5029,
      "time_per_iteration": 2.7162516117095947
    },
    {
      "auxiliary_loss_clip": 0.01181238,
      "auxiliary_loss_mlp": 0.01026544,
      "balance_loss_clip": 1.04814935,
      "balance_loss_mlp": 1.01818144,
      "epoch": 0.6048217399146275,
      "flos": 25337097025920.0,
      "grad_norm": 2.1299555703400497,
      "language_loss": 0.71834356,
      "learning_rate": 1.4262829804707836e-06,
      "loss": 0.74042135,
      "num_input_tokens_seen": 108464280,
      "step": 5030,
      "time_per_iteration": 3.6451117992401123
    },
    {
      "auxiliary_loss_clip": 0.0117935,
      "auxiliary_loss_mlp": 0.01030028,
      "balance_loss_clip": 1.0475657,
      "balance_loss_mlp": 1.02183819,
      "epoch": 0.6049419828052667,
      "flos": 26030819370240.0,
      "grad_norm": 1.5982998797806172,
      "language_loss": 0.70050704,
      "learning_rate": 1.4255367902297958e-06,
      "loss": 0.72260082,
      "num_input_tokens_seen": 108485610,
      "step": 5031,
      "time_per_iteration": 3.580777883529663
    },
    {
      "auxiliary_loss_clip": 0.01172987,
      "auxiliary_loss_mlp": 0.01025158,
      "balance_loss_clip": 1.05085516,
      "balance_loss_mlp": 1.01769578,
      "epoch": 0.6050622256959057,
      "flos": 14648106948480.0,
      "grad_norm": 2.7713670652720634,
      "language_loss": 0.79227233,
      "learning_rate": 1.4247906871344215e-06,
      "loss": 0.81425393,
      "num_input_tokens_seen": 108501005,
      "step": 5032,
      "time_per_iteration": 2.553950309753418
    },
    {
      "auxiliary_loss_clip": 0.01172913,
      "auxiliary_loss_mlp": 0.01023296,
      "balance_loss_clip": 1.04407072,
      "balance_loss_mlp": 1.01545227,
      "epoch": 0.6051824685865448,
      "flos": 23331450337920.0,
      "grad_norm": 2.3548184473681664,
      "language_loss": 0.75832593,
      "learning_rate": 1.4240446712978415e-06,
      "loss": 0.7802881,
      "num_input_tokens_seen": 108519990,
      "step": 5033,
      "time_per_iteration": 2.6379687786102295
    },
    {
      "auxiliary_loss_clip": 0.01183206,
      "auxiliary_loss_mlp": 0.01023906,
      "balance_loss_clip": 1.05186605,
      "balance_loss_mlp": 1.01549602,
      "epoch": 0.605302711477184,
      "flos": 27563307177600.0,
      "grad_norm": 2.173885137805229,
      "language_loss": 0.74618816,
      "learning_rate": 1.423298742833227e-06,
      "loss": 0.76825929,
      "num_input_tokens_seen": 108538650,
      "step": 5034,
      "time_per_iteration": 2.646556854248047
    },
    {
      "auxiliary_loss_clip": 0.01187477,
      "auxiliary_loss_mlp": 0.01027469,
      "balance_loss_clip": 1.04498959,
      "balance_loss_mlp": 1.01952958,
      "epoch": 0.605422954367823,
      "flos": 15154698412800.0,
      "grad_norm": 2.0834873239183547,
      "language_loss": 0.7163105,
      "learning_rate": 1.4225529018537352e-06,
      "loss": 0.73846,
      "num_input_tokens_seen": 108554155,
      "step": 5035,
      "time_per_iteration": 2.618210554122925
    },
    {
      "auxiliary_loss_clip": 0.01175154,
      "auxiliary_loss_mlp": 0.01027091,
      "balance_loss_clip": 1.05205536,
      "balance_loss_mlp": 1.01918137,
      "epoch": 0.6055431972584621,
      "flos": 27673912131840.0,
      "grad_norm": 2.565720883085662,
      "language_loss": 0.78022432,
      "learning_rate": 1.4218071484725082e-06,
      "loss": 0.80224681,
      "num_input_tokens_seen": 108576275,
      "step": 5036,
      "time_per_iteration": 2.6119887828826904
    },
    {
      "auxiliary_loss_clip": 0.01178163,
      "auxiliary_loss_mlp": 0.01030707,
      "balance_loss_clip": 1.05043077,
      "balance_loss_mlp": 1.02311659,
      "epoch": 0.6056634401491012,
      "flos": 19387489006080.0,
      "grad_norm": 2.2902250916733053,
      "language_loss": 0.76614302,
      "learning_rate": 1.4210614828026786e-06,
      "loss": 0.78823173,
      "num_input_tokens_seen": 108594125,
      "step": 5037,
      "time_per_iteration": 3.5032567977905273
    },
    {
      "auxiliary_loss_clip": 0.01173997,
      "auxiliary_loss_mlp": 0.01022913,
      "balance_loss_clip": 1.05012941,
      "balance_loss_mlp": 1.01499772,
      "epoch": 0.6057836830397403,
      "flos": 24789459294720.0,
      "grad_norm": 3.443004995919777,
      "language_loss": 0.74584115,
      "learning_rate": 1.4203159049573605e-06,
      "loss": 0.76781023,
      "num_input_tokens_seen": 108615360,
      "step": 5038,
      "time_per_iteration": 2.6491329669952393
    },
    {
      "auxiliary_loss_clip": 0.01188803,
      "auxiliary_loss_mlp": 0.01028933,
      "balance_loss_clip": 1.04893112,
      "balance_loss_mlp": 1.02091014,
      "epoch": 0.6059039259303793,
      "flos": 20558248899840.0,
      "grad_norm": 2.1387424183932815,
      "language_loss": 0.87204307,
      "learning_rate": 1.4195704150496593e-06,
      "loss": 0.89422041,
      "num_input_tokens_seen": 108633075,
      "step": 5039,
      "time_per_iteration": 2.5720555782318115
    },
    {
      "auxiliary_loss_clip": 0.01180558,
      "auxiliary_loss_mlp": 0.01025896,
      "balance_loss_clip": 1.04997468,
      "balance_loss_mlp": 1.01788557,
      "epoch": 0.6060241688210185,
      "flos": 21069724613760.0,
      "grad_norm": 7.062919077395469,
      "language_loss": 0.73850918,
      "learning_rate": 1.4188250131926639e-06,
      "loss": 0.76057374,
      "num_input_tokens_seen": 108651875,
      "step": 5040,
      "time_per_iteration": 2.6150922775268555
    },
    {
      "auxiliary_loss_clip": 0.01181176,
      "auxiliary_loss_mlp": 0.01029753,
      "balance_loss_clip": 1.04814029,
      "balance_loss_mlp": 1.02142584,
      "epoch": 0.6061444117116576,
      "flos": 16361081619840.0,
      "grad_norm": 3.510038960637255,
      "language_loss": 0.80597758,
      "learning_rate": 1.4180796994994525e-06,
      "loss": 0.82808685,
      "num_input_tokens_seen": 108669290,
      "step": 5041,
      "time_per_iteration": 2.5717761516571045
    },
    {
      "auxiliary_loss_clip": 0.01178979,
      "auxiliary_loss_mlp": 0.01020614,
      "balance_loss_clip": 1.04740834,
      "balance_loss_mlp": 1.01259136,
      "epoch": 0.6062646546022966,
      "flos": 21507296094720.0,
      "grad_norm": 2.21495807580048,
      "language_loss": 0.72357035,
      "learning_rate": 1.4173344740830877e-06,
      "loss": 0.74556625,
      "num_input_tokens_seen": 108688420,
      "step": 5042,
      "time_per_iteration": 2.633653163909912
    },
    {
      "auxiliary_loss_clip": 0.01174496,
      "auxiliary_loss_mlp": 0.01031649,
      "balance_loss_clip": 1.050228,
      "balance_loss_mlp": 1.02361417,
      "epoch": 0.6063848974929358,
      "flos": 38983151283840.0,
      "grad_norm": 1.8794445828893294,
      "language_loss": 0.7080003,
      "learning_rate": 1.4165893370566206e-06,
      "loss": 0.73006177,
      "num_input_tokens_seen": 108712175,
      "step": 5043,
      "time_per_iteration": 0.16152715682983398
    },
    {
      "auxiliary_loss_clip": 0.01173254,
      "auxiliary_loss_mlp": 0.01028576,
      "balance_loss_clip": 1.04732394,
      "balance_loss_mlp": 1.02017212,
      "epoch": 0.6065051403835748,
      "flos": 19646584784640.0,
      "grad_norm": 2.1193765204144674,
      "language_loss": 0.77821457,
      "learning_rate": 1.4158442885330865e-06,
      "loss": 0.80023289,
      "num_input_tokens_seen": 108730745,
      "step": 5044,
      "time_per_iteration": 2.6316590309143066
    },
    {
      "auxiliary_loss_clip": 0.01172298,
      "auxiliary_loss_mlp": 0.01028344,
      "balance_loss_clip": 1.04792714,
      "balance_loss_mlp": 1.02006507,
      "epoch": 0.6066253832742139,
      "flos": 23513086437120.0,
      "grad_norm": 4.31762734549136,
      "language_loss": 0.78710723,
      "learning_rate": 1.4150993286255094e-06,
      "loss": 0.80911362,
      "num_input_tokens_seen": 108749995,
      "step": 5045,
      "time_per_iteration": 2.8026137351989746
    },
    {
      "auxiliary_loss_clip": 0.01172911,
      "auxiliary_loss_mlp": 0.01024448,
      "balance_loss_clip": 1.04942703,
      "balance_loss_mlp": 1.01690507,
      "epoch": 0.6067456261648531,
      "flos": 19133708440320.0,
      "grad_norm": 2.351345328780664,
      "language_loss": 0.80223757,
      "learning_rate": 1.4143544574468993e-06,
      "loss": 0.82421112,
      "num_input_tokens_seen": 108768355,
      "step": 5046,
      "time_per_iteration": 2.5328972339630127
    },
    {
      "auxiliary_loss_clip": 0.01177238,
      "auxiliary_loss_mlp": 0.01024039,
      "balance_loss_clip": 1.05100822,
      "balance_loss_mlp": 1.01567101,
      "epoch": 0.6068658690554921,
      "flos": 20520614424960.0,
      "grad_norm": 2.3209785460331367,
      "language_loss": 0.82462043,
      "learning_rate": 1.4136096751102523e-06,
      "loss": 0.84663326,
      "num_input_tokens_seen": 108786685,
      "step": 5047,
      "time_per_iteration": 2.644817352294922
    },
    {
      "auxiliary_loss_clip": 0.01185765,
      "auxiliary_loss_mlp": 0.01027839,
      "balance_loss_clip": 1.05248618,
      "balance_loss_mlp": 1.01989412,
      "epoch": 0.6069861119461312,
      "flos": 27374560185600.0,
      "grad_norm": 3.230582509067918,
      "language_loss": 0.827564,
      "learning_rate": 1.4128649817285516e-06,
      "loss": 0.84970009,
      "num_input_tokens_seen": 108804820,
      "step": 5048,
      "time_per_iteration": 2.674861431121826
    },
    {
      "auxiliary_loss_clip": 0.01182394,
      "auxiliary_loss_mlp": 0.01036413,
      "balance_loss_clip": 1.04690182,
      "balance_loss_mlp": 1.0277288,
      "epoch": 0.6071063548367702,
      "flos": 25626500904960.0,
      "grad_norm": 1.8290046441514365,
      "language_loss": 0.63503176,
      "learning_rate": 1.412120377414766e-06,
      "loss": 0.65721977,
      "num_input_tokens_seen": 108825010,
      "step": 5049,
      "time_per_iteration": 2.7172346115112305
    },
    {
      "auxiliary_loss_clip": 0.01176044,
      "auxiliary_loss_mlp": 0.01028488,
      "balance_loss_clip": 1.05338001,
      "balance_loss_mlp": 1.02060282,
      "epoch": 0.6072265977274094,
      "flos": 24460517520000.0,
      "grad_norm": 1.7352252068372476,
      "language_loss": 0.711371,
      "learning_rate": 1.4113758622818522e-06,
      "loss": 0.73341632,
      "num_input_tokens_seen": 108845075,
      "step": 5050,
      "time_per_iteration": 2.611215829849243
    },
    {
      "auxiliary_loss_clip": 0.01185709,
      "auxiliary_loss_mlp": 0.01207743,
      "balance_loss_clip": 1.05092382,
      "balance_loss_mlp": 1.0007031,
      "epoch": 0.6073468406180484,
      "flos": 18149253413760.0,
      "grad_norm": 2.160389338184562,
      "language_loss": 0.82792222,
      "learning_rate": 1.410631436442751e-06,
      "loss": 0.85185671,
      "num_input_tokens_seen": 108863870,
      "step": 5051,
      "time_per_iteration": 2.660306930541992
    },
    {
      "auxiliary_loss_clip": 0.01182767,
      "auxiliary_loss_mlp": 0.01023565,
      "balance_loss_clip": 1.05163956,
      "balance_loss_mlp": 1.01550102,
      "epoch": 0.6074670835086875,
      "flos": 20697617669760.0,
      "grad_norm": 3.1079776954220977,
      "language_loss": 0.86849928,
      "learning_rate": 1.4098871000103936e-06,
      "loss": 0.89056259,
      "num_input_tokens_seen": 108882470,
      "step": 5052,
      "time_per_iteration": 3.5587103366851807
    },
    {
      "auxiliary_loss_clip": 0.01180602,
      "auxiliary_loss_mlp": 0.01023057,
      "balance_loss_clip": 1.0475049,
      "balance_loss_mlp": 1.01575875,
      "epoch": 0.6075873263993267,
      "flos": 23769955572480.0,
      "grad_norm": 1.7363763385668483,
      "language_loss": 0.82716548,
      "learning_rate": 1.409142853097693e-06,
      "loss": 0.8492021,
      "num_input_tokens_seen": 108902710,
      "step": 5053,
      "time_per_iteration": 2.6449806690216064
    },
    {
      "auxiliary_loss_clip": 0.01183401,
      "auxiliary_loss_mlp": 0.01023924,
      "balance_loss_clip": 1.05008698,
      "balance_loss_mlp": 1.01636612,
      "epoch": 0.6077075692899657,
      "flos": 24454484035200.0,
      "grad_norm": 2.247412843755293,
      "language_loss": 0.79679465,
      "learning_rate": 1.408398695817553e-06,
      "loss": 0.81886792,
      "num_input_tokens_seen": 108919935,
      "step": 5054,
      "time_per_iteration": 2.6193654537200928
    },
    {
      "auxiliary_loss_clip": 0.01179373,
      "auxiliary_loss_mlp": 0.01033983,
      "balance_loss_clip": 1.04749489,
      "balance_loss_mlp": 1.0245533,
      "epoch": 0.6078278121806048,
      "flos": 27382102041600.0,
      "grad_norm": 1.8822890717609,
      "language_loss": 0.70295423,
      "learning_rate": 1.4076546282828593e-06,
      "loss": 0.72508776,
      "num_input_tokens_seen": 108942790,
      "step": 5055,
      "time_per_iteration": 2.70114803314209
    },
    {
      "auxiliary_loss_clip": 0.0118236,
      "auxiliary_loss_mlp": 0.01028067,
      "balance_loss_clip": 1.04446065,
      "balance_loss_mlp": 1.02080083,
      "epoch": 0.6079480550712439,
      "flos": 38436447306240.0,
      "grad_norm": 2.5767355376044723,
      "language_loss": 0.66338557,
      "learning_rate": 1.4069106506064874e-06,
      "loss": 0.68548983,
      "num_input_tokens_seen": 108964215,
      "step": 5056,
      "time_per_iteration": 3.687307834625244
    },
    {
      "auxiliary_loss_clip": 0.01177705,
      "auxiliary_loss_mlp": 0.01028567,
      "balance_loss_clip": 1.04925823,
      "balance_loss_mlp": 1.02086043,
      "epoch": 0.608068297961883,
      "flos": 25336271013120.0,
      "grad_norm": 1.8651875924128096,
      "language_loss": 0.78320497,
      "learning_rate": 1.4061667629012989e-06,
      "loss": 0.80526769,
      "num_input_tokens_seen": 108984885,
      "step": 5057,
      "time_per_iteration": 2.6230480670928955
    },
    {
      "auxiliary_loss_clip": 0.0117343,
      "auxiliary_loss_mlp": 0.01025587,
      "balance_loss_clip": 1.04931283,
      "balance_loss_mlp": 1.01779675,
      "epoch": 0.608188540852522,
      "flos": 24202463235840.0,
      "grad_norm": 1.5411727429296045,
      "language_loss": 0.83195144,
      "learning_rate": 1.40542296528014e-06,
      "loss": 0.85394168,
      "num_input_tokens_seen": 109004545,
      "step": 5058,
      "time_per_iteration": 3.5833475589752197
    },
    {
      "auxiliary_loss_clip": 0.01178124,
      "auxiliary_loss_mlp": 0.01031047,
      "balance_loss_clip": 1.04902017,
      "balance_loss_mlp": 1.02283013,
      "epoch": 0.6083087837431612,
      "flos": 21284146851840.0,
      "grad_norm": 1.99829340727315,
      "language_loss": 0.75947529,
      "learning_rate": 1.4046792578558452e-06,
      "loss": 0.78156704,
      "num_input_tokens_seen": 109022440,
      "step": 5059,
      "time_per_iteration": 2.592033863067627
    },
    {
      "auxiliary_loss_clip": 0.01177579,
      "auxiliary_loss_mlp": 0.01027893,
      "balance_loss_clip": 1.04838204,
      "balance_loss_mlp": 1.0198108,
      "epoch": 0.6084290266338003,
      "flos": 16471435178880.0,
      "grad_norm": 2.505788661116362,
      "language_loss": 0.7593075,
      "learning_rate": 1.4039356407412325e-06,
      "loss": 0.78136218,
      "num_input_tokens_seen": 109035680,
      "step": 5060,
      "time_per_iteration": 2.5802865028381348
    },
    {
      "auxiliary_loss_clip": 0.01082628,
      "auxiliary_loss_mlp": 0.01002043,
      "balance_loss_clip": 1.01852739,
      "balance_loss_mlp": 1.00121462,
      "epoch": 0.6085492695244393,
      "flos": 66443574931200.0,
      "grad_norm": 0.7804379604570135,
      "language_loss": 0.5708093,
      "learning_rate": 1.40319211404911e-06,
      "loss": 0.59165603,
      "num_input_tokens_seen": 109090680,
      "step": 5061,
      "time_per_iteration": 3.1093366146087646
    },
    {
      "auxiliary_loss_clip": 0.01175942,
      "auxiliary_loss_mlp": 0.01026054,
      "balance_loss_clip": 1.05153859,
      "balance_loss_mlp": 1.01803112,
      "epoch": 0.6086695124150785,
      "flos": 23618986709760.0,
      "grad_norm": 3.1266423714482805,
      "language_loss": 0.90753543,
      "learning_rate": 1.4024486778922691e-06,
      "loss": 0.92955542,
      "num_input_tokens_seen": 109108995,
      "step": 5062,
      "time_per_iteration": 2.628171682357788
    },
    {
      "auxiliary_loss_clip": 0.01185403,
      "auxiliary_loss_mlp": 0.01028251,
      "balance_loss_clip": 1.047719,
      "balance_loss_mlp": 1.02001965,
      "epoch": 0.6087897553057176,
      "flos": 20157054917760.0,
      "grad_norm": 1.8003113215400008,
      "language_loss": 0.7769773,
      "learning_rate": 1.4017053323834884e-06,
      "loss": 0.79911381,
      "num_input_tokens_seen": 109128825,
      "step": 5063,
      "time_per_iteration": 2.6883091926574707
    },
    {
      "auxiliary_loss_clip": 0.01182187,
      "auxiliary_loss_mlp": 0.01022923,
      "balance_loss_clip": 1.04677272,
      "balance_loss_mlp": 1.01525736,
      "epoch": 0.6089099981963566,
      "flos": 25482535194240.0,
      "grad_norm": 1.9799081305000532,
      "language_loss": 0.76105893,
      "learning_rate": 1.4009620776355333e-06,
      "loss": 0.78311002,
      "num_input_tokens_seen": 109150425,
      "step": 5064,
      "time_per_iteration": 3.6044368743896484
    },
    {
      "auxiliary_loss_clip": 0.01175235,
      "auxiliary_loss_mlp": 0.01024798,
      "balance_loss_clip": 1.04853439,
      "balance_loss_mlp": 1.01686502,
      "epoch": 0.6090302410869958,
      "flos": 25332895134720.0,
      "grad_norm": 1.9947922848931026,
      "language_loss": 0.79056668,
      "learning_rate": 1.4002189137611553e-06,
      "loss": 0.81256706,
      "num_input_tokens_seen": 109169765,
      "step": 5065,
      "time_per_iteration": 2.649110794067383
    },
    {
      "auxiliary_loss_clip": 0.01177196,
      "auxiliary_loss_mlp": 0.0102332,
      "balance_loss_clip": 1.04913187,
      "balance_loss_mlp": 1.01548529,
      "epoch": 0.6091504839776348,
      "flos": 23987358639360.0,
      "grad_norm": 1.615571030452752,
      "language_loss": 0.6975463,
      "learning_rate": 1.3994758408730901e-06,
      "loss": 0.71955144,
      "num_input_tokens_seen": 109188950,
      "step": 5066,
      "time_per_iteration": 2.610405206680298
    },
    {
      "auxiliary_loss_clip": 0.01184894,
      "auxiliary_loss_mlp": 0.0102592,
      "balance_loss_clip": 1.05129087,
      "balance_loss_mlp": 1.01705122,
      "epoch": 0.6092707268682739,
      "flos": 29643037666560.0,
      "grad_norm": 1.9633369017840825,
      "language_loss": 0.76316214,
      "learning_rate": 1.3987328590840629e-06,
      "loss": 0.78527027,
      "num_input_tokens_seen": 109209895,
      "step": 5067,
      "time_per_iteration": 2.7003724575042725
    },
    {
      "auxiliary_loss_clip": 0.01173786,
      "auxiliary_loss_mlp": 0.01028163,
      "balance_loss_clip": 1.0478363,
      "balance_loss_mlp": 1.02040207,
      "epoch": 0.609390969758913,
      "flos": 24024957200640.0,
      "grad_norm": 1.8659487757222524,
      "language_loss": 0.86115032,
      "learning_rate": 1.397989968506783e-06,
      "loss": 0.88316977,
      "num_input_tokens_seen": 109228905,
      "step": 5068,
      "time_per_iteration": 2.666663408279419
    },
    {
      "auxiliary_loss_clip": 0.01181848,
      "auxiliary_loss_mlp": 0.01032813,
      "balance_loss_clip": 1.05413246,
      "balance_loss_mlp": 1.02450383,
      "epoch": 0.6095112126495521,
      "flos": 11102143288320.0,
      "grad_norm": 2.238993931278287,
      "language_loss": 0.72410131,
      "learning_rate": 1.3972471692539458e-06,
      "loss": 0.74624801,
      "num_input_tokens_seen": 109243620,
      "step": 5069,
      "time_per_iteration": 2.509955406188965
    },
    {
      "auxiliary_loss_clip": 0.01177434,
      "auxiliary_loss_mlp": 0.01024384,
      "balance_loss_clip": 1.04846847,
      "balance_loss_mlp": 1.0164032,
      "epoch": 0.6096314555401912,
      "flos": 17265491187840.0,
      "grad_norm": 2.893376417816524,
      "language_loss": 0.75348091,
      "learning_rate": 1.3965044614382348e-06,
      "loss": 0.77549911,
      "num_input_tokens_seen": 109259070,
      "step": 5070,
      "time_per_iteration": 2.6259515285491943
    },
    {
      "auxiliary_loss_clip": 0.01179629,
      "auxiliary_loss_mlp": 0.01026394,
      "balance_loss_clip": 1.05303144,
      "balance_loss_mlp": 1.01813304,
      "epoch": 0.6097516984308303,
      "flos": 21645910679040.0,
      "grad_norm": 3.2864562525984455,
      "language_loss": 0.75536793,
      "learning_rate": 1.3957618451723162e-06,
      "loss": 0.77742815,
      "num_input_tokens_seen": 109275100,
      "step": 5071,
      "time_per_iteration": 2.5803985595703125
    },
    {
      "auxiliary_loss_clip": 0.01183128,
      "auxiliary_loss_mlp": 0.01027322,
      "balance_loss_clip": 1.04997599,
      "balance_loss_mlp": 1.0195111,
      "epoch": 0.6098719413214694,
      "flos": 27199208966400.0,
      "grad_norm": 2.7111579412857787,
      "language_loss": 0.71055281,
      "learning_rate": 1.3950193205688457e-06,
      "loss": 0.73265731,
      "num_input_tokens_seen": 109294825,
      "step": 5072,
      "time_per_iteration": 2.6897881031036377
    },
    {
      "auxiliary_loss_clip": 0.01177309,
      "auxiliary_loss_mlp": 0.01024152,
      "balance_loss_clip": 1.04891419,
      "balance_loss_mlp": 1.01628435,
      "epoch": 0.6099921842121084,
      "flos": 20412954385920.0,
      "grad_norm": 2.4825566524774887,
      "language_loss": 0.83682263,
      "learning_rate": 1.3942768877404627e-06,
      "loss": 0.85883725,
      "num_input_tokens_seen": 109313790,
      "step": 5073,
      "time_per_iteration": 2.628075361251831
    },
    {
      "auxiliary_loss_clip": 0.01174142,
      "auxiliary_loss_mlp": 0.01028294,
      "balance_loss_clip": 1.05048752,
      "balance_loss_mlp": 1.02065301,
      "epoch": 0.6101124271027476,
      "flos": 23366139897600.0,
      "grad_norm": 1.6061105353903782,
      "language_loss": 0.74111176,
      "learning_rate": 1.393534546799795e-06,
      "loss": 0.76313609,
      "num_input_tokens_seen": 109333490,
      "step": 5074,
      "time_per_iteration": 2.635955333709717
    },
    {
      "auxiliary_loss_clip": 0.01171453,
      "auxiliary_loss_mlp": 0.01032926,
      "balance_loss_clip": 1.04806912,
      "balance_loss_mlp": 1.02434886,
      "epoch": 0.6102326699933867,
      "flos": 26687840993280.0,
      "grad_norm": 1.626576152387059,
      "language_loss": 0.68156779,
      "learning_rate": 1.3927922978594536e-06,
      "loss": 0.70361155,
      "num_input_tokens_seen": 109354575,
      "step": 5075,
      "time_per_iteration": 2.65165638923645
    },
    {
      "auxiliary_loss_clip": 0.01077172,
      "auxiliary_loss_mlp": 0.01001315,
      "balance_loss_clip": 1.01723218,
      "balance_loss_mlp": 1.00048101,
      "epoch": 0.6103529128840257,
      "flos": 60644612551680.0,
      "grad_norm": 0.772099737864492,
      "language_loss": 0.57430315,
      "learning_rate": 1.3920501410320387e-06,
      "loss": 0.59508801,
      "num_input_tokens_seen": 109410690,
      "step": 5076,
      "time_per_iteration": 3.1412718296051025
    },
    {
      "auxiliary_loss_clip": 0.01178737,
      "auxiliary_loss_mlp": 0.01030383,
      "balance_loss_clip": 1.04584563,
      "balance_loss_mlp": 1.02212155,
      "epoch": 0.6104731557746649,
      "flos": 19021307806080.0,
      "grad_norm": 2.510351153144604,
      "language_loss": 0.76228768,
      "learning_rate": 1.3913080764301333e-06,
      "loss": 0.78437889,
      "num_input_tokens_seen": 109427650,
      "step": 5077,
      "time_per_iteration": 2.5976014137268066
    },
    {
      "auxiliary_loss_clip": 0.0119387,
      "auxiliary_loss_mlp": 0.01031543,
      "balance_loss_clip": 1.04367447,
      "balance_loss_mlp": 1.02387142,
      "epoch": 0.6105933986653039,
      "flos": 23366894083200.0,
      "grad_norm": 1.8955406017699297,
      "language_loss": 0.71247256,
      "learning_rate": 1.3905661041663085e-06,
      "loss": 0.73472667,
      "num_input_tokens_seen": 109448835,
      "step": 5078,
      "time_per_iteration": 2.695762872695923
    },
    {
      "auxiliary_loss_clip": 0.01176469,
      "auxiliary_loss_mlp": 0.01029795,
      "balance_loss_clip": 1.04974699,
      "balance_loss_mlp": 1.02143288,
      "epoch": 0.610713641555943,
      "flos": 34637565006720.0,
      "grad_norm": 2.2578640853817444,
      "language_loss": 0.65522736,
      "learning_rate": 1.389824224353122e-06,
      "loss": 0.67729002,
      "num_input_tokens_seen": 109470425,
      "step": 5079,
      "time_per_iteration": 3.696056365966797
    },
    {
      "auxiliary_loss_clip": 0.01177535,
      "auxiliary_loss_mlp": 0.01022625,
      "balance_loss_clip": 1.05167389,
      "balance_loss_mlp": 1.01462317,
      "epoch": 0.610833884446582,
      "flos": 26646471504000.0,
      "grad_norm": 1.4708118395561334,
      "language_loss": 0.7693693,
      "learning_rate": 1.389082437103115e-06,
      "loss": 0.79137093,
      "num_input_tokens_seen": 109489695,
      "step": 5080,
      "time_per_iteration": 2.6110668182373047
    },
    {
      "auxiliary_loss_clip": 0.01180567,
      "auxiliary_loss_mlp": 0.01027631,
      "balance_loss_clip": 1.04388714,
      "balance_loss_mlp": 1.01925063,
      "epoch": 0.6109541273372212,
      "flos": 21215126868480.0,
      "grad_norm": 3.7956594243042576,
      "language_loss": 0.78035128,
      "learning_rate": 1.3883407425288172e-06,
      "loss": 0.80243331,
      "num_input_tokens_seen": 109510030,
      "step": 5081,
      "time_per_iteration": 2.673070192337036
    },
    {
      "auxiliary_loss_clip": 0.01179333,
      "auxiliary_loss_mlp": 0.01023682,
      "balance_loss_clip": 1.04747128,
      "balance_loss_mlp": 1.01559973,
      "epoch": 0.6110743702278603,
      "flos": 20084084438400.0,
      "grad_norm": 2.1210600764850978,
      "language_loss": 0.79455495,
      "learning_rate": 1.3875991407427417e-06,
      "loss": 0.81658512,
      "num_input_tokens_seen": 109528255,
      "step": 5082,
      "time_per_iteration": 3.599820375442505
    },
    {
      "auxiliary_loss_clip": 0.01097633,
      "auxiliary_loss_mlp": 0.01003402,
      "balance_loss_clip": 1.01820552,
      "balance_loss_mlp": 1.00260341,
      "epoch": 0.6111946131184993,
      "flos": 68302957438080.0,
      "grad_norm": 0.7876095164296598,
      "language_loss": 0.58092737,
      "learning_rate": 1.38685763185739e-06,
      "loss": 0.60193777,
      "num_input_tokens_seen": 109581915,
      "step": 5083,
      "time_per_iteration": 3.2628228664398193
    },
    {
      "auxiliary_loss_clip": 0.01173852,
      "auxiliary_loss_mlp": 0.01026176,
      "balance_loss_clip": 1.04989815,
      "balance_loss_mlp": 1.01811755,
      "epoch": 0.6113148560091385,
      "flos": 19937676602880.0,
      "grad_norm": 3.203873230155367,
      "language_loss": 0.675439,
      "learning_rate": 1.3861162159852476e-06,
      "loss": 0.69743931,
      "num_input_tokens_seen": 109600050,
      "step": 5084,
      "time_per_iteration": 3.5828440189361572
    },
    {
      "auxiliary_loss_clip": 0.01185893,
      "auxiliary_loss_mlp": 0.01024836,
      "balance_loss_clip": 1.05011582,
      "balance_loss_mlp": 1.01626492,
      "epoch": 0.6114350988997775,
      "flos": 23731854220800.0,
      "grad_norm": 1.9908404180133539,
      "language_loss": 0.79450369,
      "learning_rate": 1.3853748932387875e-06,
      "loss": 0.81661105,
      "num_input_tokens_seen": 109620690,
      "step": 5085,
      "time_per_iteration": 2.6813294887542725
    },
    {
      "auxiliary_loss_clip": 0.01167824,
      "auxiliary_loss_mlp": 0.01021103,
      "balance_loss_clip": 1.04497218,
      "balance_loss_mlp": 1.01307416,
      "epoch": 0.6115553417904166,
      "flos": 24023700224640.0,
      "grad_norm": 2.862464199657492,
      "language_loss": 0.75127709,
      "learning_rate": 1.3846336637304671e-06,
      "loss": 0.7731663,
      "num_input_tokens_seen": 109638960,
      "step": 5086,
      "time_per_iteration": 2.6040968894958496
    },
    {
      "auxiliary_loss_clip": 0.01173298,
      "auxiliary_loss_mlp": 0.0102119,
      "balance_loss_clip": 1.04815578,
      "balance_loss_mlp": 1.01340568,
      "epoch": 0.6116755846810558,
      "flos": 23733542160000.0,
      "grad_norm": 2.314910949767816,
      "language_loss": 0.83049655,
      "learning_rate": 1.3838925275727316e-06,
      "loss": 0.85244143,
      "num_input_tokens_seen": 109659700,
      "step": 5087,
      "time_per_iteration": 2.6303718090057373
    },
    {
      "auxiliary_loss_clip": 0.01177432,
      "auxiliary_loss_mlp": 0.01022366,
      "balance_loss_clip": 1.05344653,
      "balance_loss_mlp": 1.0148381,
      "epoch": 0.6117958275716948,
      "flos": 18661626967680.0,
      "grad_norm": 3.2293411173891475,
      "language_loss": 0.79235452,
      "learning_rate": 1.3831514848780089e-06,
      "loss": 0.81435251,
      "num_input_tokens_seen": 109679275,
      "step": 5088,
      "time_per_iteration": 2.696194887161255
    },
    {
      "auxiliary_loss_clip": 0.01172841,
      "auxiliary_loss_mlp": 0.0102813,
      "balance_loss_clip": 1.04904413,
      "balance_loss_mlp": 1.02068853,
      "epoch": 0.6119160704623339,
      "flos": 16471183783680.0,
      "grad_norm": 2.418519658602383,
      "language_loss": 0.91997916,
      "learning_rate": 1.3824105357587152e-06,
      "loss": 0.94198883,
      "num_input_tokens_seen": 109696380,
      "step": 5089,
      "time_per_iteration": 2.6485562324523926
    },
    {
      "auxiliary_loss_clip": 0.01175517,
      "auxiliary_loss_mlp": 0.01025791,
      "balance_loss_clip": 1.04652929,
      "balance_loss_mlp": 1.01787543,
      "epoch": 0.612036313352973,
      "flos": 23915465568000.0,
      "grad_norm": 1.5230363849385569,
      "language_loss": 0.82592911,
      "learning_rate": 1.381669680327253e-06,
      "loss": 0.84794223,
      "num_input_tokens_seen": 109718060,
      "step": 5090,
      "time_per_iteration": 2.670034408569336
    },
    {
      "auxiliary_loss_clip": 0.01175313,
      "auxiliary_loss_mlp": 0.01029392,
      "balance_loss_clip": 1.04953635,
      "balance_loss_mlp": 1.0211544,
      "epoch": 0.6121565562436121,
      "flos": 26974766833920.0,
      "grad_norm": 1.936783748205079,
      "language_loss": 0.70665979,
      "learning_rate": 1.380928918696008e-06,
      "loss": 0.72870684,
      "num_input_tokens_seen": 109736830,
      "step": 5091,
      "time_per_iteration": 3.6527633666992188
    },
    {
      "auxiliary_loss_clip": 0.01176472,
      "auxiliary_loss_mlp": 0.01024014,
      "balance_loss_clip": 1.04779077,
      "balance_loss_mlp": 1.01594305,
      "epoch": 0.6122767991342511,
      "flos": 15668867646720.0,
      "grad_norm": 2.6735034088202037,
      "language_loss": 0.71480477,
      "learning_rate": 1.3801882509773548e-06,
      "loss": 0.73680961,
      "num_input_tokens_seen": 109754690,
      "step": 5092,
      "time_per_iteration": 2.6291370391845703
    },
    {
      "auxiliary_loss_clip": 0.01172004,
      "auxiliary_loss_mlp": 0.01026916,
      "balance_loss_clip": 1.04674351,
      "balance_loss_mlp": 1.018363,
      "epoch": 0.6123970420248903,
      "flos": 27964321591680.0,
      "grad_norm": 1.7437594060807529,
      "language_loss": 0.81576681,
      "learning_rate": 1.3794476772836503e-06,
      "loss": 0.83775598,
      "num_input_tokens_seen": 109775790,
      "step": 5093,
      "time_per_iteration": 2.6862785816192627
    },
    {
      "auxiliary_loss_clip": 0.01174005,
      "auxiliary_loss_mlp": 0.01030612,
      "balance_loss_clip": 1.04582834,
      "balance_loss_mlp": 1.02234459,
      "epoch": 0.6125172849155294,
      "flos": 21468727866240.0,
      "grad_norm": 1.7076293528543889,
      "language_loss": 0.84236282,
      "learning_rate": 1.3787071977272402e-06,
      "loss": 0.86440897,
      "num_input_tokens_seen": 109795050,
      "step": 5094,
      "time_per_iteration": 2.670478105545044
    },
    {
      "auxiliary_loss_clip": 0.01178264,
      "auxiliary_loss_mlp": 0.01029128,
      "balance_loss_clip": 1.04598904,
      "balance_loss_mlp": 1.02135563,
      "epoch": 0.6126375278061684,
      "flos": 16248321849600.0,
      "grad_norm": 2.6458999407395365,
      "language_loss": 0.7202065,
      "learning_rate": 1.3779668124204535e-06,
      "loss": 0.74228042,
      "num_input_tokens_seen": 109811465,
      "step": 5095,
      "time_per_iteration": 2.6484639644622803
    },
    {
      "auxiliary_loss_clip": 0.01175529,
      "auxiliary_loss_mlp": 0.01026635,
      "balance_loss_clip": 1.05058634,
      "balance_loss_mlp": 1.01830268,
      "epoch": 0.6127577706968076,
      "flos": 20448865008000.0,
      "grad_norm": 2.0319068146436967,
      "language_loss": 0.80692297,
      "learning_rate": 1.3772265214756074e-06,
      "loss": 0.82894456,
      "num_input_tokens_seen": 109831225,
      "step": 5096,
      "time_per_iteration": 2.7503623962402344
    },
    {
      "auxiliary_loss_clip": 0.01179261,
      "auxiliary_loss_mlp": 0.01027903,
      "balance_loss_clip": 1.04747784,
      "balance_loss_mlp": 1.0201993,
      "epoch": 0.6128780135874466,
      "flos": 18260397072000.0,
      "grad_norm": 2.6550697842569457,
      "language_loss": 0.74911559,
      "learning_rate": 1.3764863250050025e-06,
      "loss": 0.77118719,
      "num_input_tokens_seen": 109849465,
      "step": 5097,
      "time_per_iteration": 2.6563849449157715
    },
    {
      "auxiliary_loss_clip": 0.01183149,
      "auxiliary_loss_mlp": 0.01026381,
      "balance_loss_clip": 1.04635787,
      "balance_loss_mlp": 1.01892471,
      "epoch": 0.6129982564780857,
      "flos": 24937088192640.0,
      "grad_norm": 1.8012961888444112,
      "language_loss": 0.80271059,
      "learning_rate": 1.3757462231209272e-06,
      "loss": 0.82480592,
      "num_input_tokens_seen": 109869770,
      "step": 5098,
      "time_per_iteration": 2.6943418979644775
    },
    {
      "auxiliary_loss_clip": 0.01173351,
      "auxiliary_loss_mlp": 0.01022439,
      "balance_loss_clip": 1.04533815,
      "balance_loss_mlp": 1.01430941,
      "epoch": 0.6131184993687249,
      "flos": 22492038430080.0,
      "grad_norm": 2.2577874851712445,
      "language_loss": 0.88851905,
      "learning_rate": 1.3750062159356525e-06,
      "loss": 0.91047692,
      "num_input_tokens_seen": 109889120,
      "step": 5099,
      "time_per_iteration": 2.6685609817504883
    },
    {
      "auxiliary_loss_clip": 0.01171332,
      "auxiliary_loss_mlp": 0.01025201,
      "balance_loss_clip": 1.04364431,
      "balance_loss_mlp": 1.01747286,
      "epoch": 0.6132387422593639,
      "flos": 15885839750400.0,
      "grad_norm": 2.4785362670954956,
      "language_loss": 0.83180481,
      "learning_rate": 1.3742663035614382e-06,
      "loss": 0.8537702,
      "num_input_tokens_seen": 109906490,
      "step": 5100,
      "time_per_iteration": 2.6382744312286377
    },
    {
      "auxiliary_loss_clip": 0.01175628,
      "auxiliary_loss_mlp": 0.01028393,
      "balance_loss_clip": 1.05099273,
      "balance_loss_mlp": 1.02025104,
      "epoch": 0.613358985150003,
      "flos": 25411539962880.0,
      "grad_norm": 1.7348039668685393,
      "language_loss": 0.80077863,
      "learning_rate": 1.3735264861105283e-06,
      "loss": 0.82281882,
      "num_input_tokens_seen": 109927130,
      "step": 5101,
      "time_per_iteration": 2.623997211456299
    },
    {
      "auxiliary_loss_clip": 0.01182345,
      "auxiliary_loss_mlp": 0.01025312,
      "balance_loss_clip": 1.04458463,
      "balance_loss_mlp": 1.01786721,
      "epoch": 0.6134792280406421,
      "flos": 21361283308800.0,
      "grad_norm": 2.6543457952471257,
      "language_loss": 0.78217554,
      "learning_rate": 1.372786763695152e-06,
      "loss": 0.80425209,
      "num_input_tokens_seen": 109945890,
      "step": 5102,
      "time_per_iteration": 2.68452525138855
    },
    {
      "auxiliary_loss_clip": 0.01176349,
      "auxiliary_loss_mlp": 0.01031913,
      "balance_loss_clip": 1.04709435,
      "balance_loss_mlp": 1.02365136,
      "epoch": 0.6135994709312812,
      "flos": 21211248199680.0,
      "grad_norm": 1.8236879808148188,
      "language_loss": 0.7748614,
      "learning_rate": 1.3720471364275257e-06,
      "loss": 0.79694402,
      "num_input_tokens_seen": 109965535,
      "step": 5103,
      "time_per_iteration": 2.6292030811309814
    },
    {
      "auxiliary_loss_clip": 0.01178766,
      "auxiliary_loss_mlp": 0.01208259,
      "balance_loss_clip": 1.04594827,
      "balance_loss_mlp": 1.00074911,
      "epoch": 0.6137197138219203,
      "flos": 14794047907200.0,
      "grad_norm": 2.4187513028356995,
      "language_loss": 0.78140414,
      "learning_rate": 1.3713076044198486e-06,
      "loss": 0.80527437,
      "num_input_tokens_seen": 109982345,
      "step": 5104,
      "time_per_iteration": 2.7002320289611816
    },
    {
      "auxiliary_loss_clip": 0.0117395,
      "auxiliary_loss_mlp": 0.01028647,
      "balance_loss_clip": 1.04637671,
      "balance_loss_mlp": 1.02036238,
      "epoch": 0.6138399567125594,
      "flos": 20084515401600.0,
      "grad_norm": 2.543820044002376,
      "language_loss": 0.8087523,
      "learning_rate": 1.3705681677843086e-06,
      "loss": 0.83077836,
      "num_input_tokens_seen": 110000940,
      "step": 5105,
      "time_per_iteration": 3.535745859146118
    },
    {
      "auxiliary_loss_clip": 0.01072607,
      "auxiliary_loss_mlp": 0.01003371,
      "balance_loss_clip": 1.01582694,
      "balance_loss_mlp": 1.00253677,
      "epoch": 0.6139601996031985,
      "flos": 60123838193280.0,
      "grad_norm": 0.7776111733599942,
      "language_loss": 0.60553366,
      "learning_rate": 1.3698288266330768e-06,
      "loss": 0.62629342,
      "num_input_tokens_seen": 110061565,
      "step": 5106,
      "time_per_iteration": 3.2298731803894043
    },
    {
      "auxiliary_loss_clip": 0.01176511,
      "auxiliary_loss_mlp": 0.0102348,
      "balance_loss_clip": 1.05139351,
      "balance_loss_mlp": 1.01589572,
      "epoch": 0.6140804424938375,
      "flos": 23586703361280.0,
      "grad_norm": 2.2033696087811037,
      "language_loss": 0.72859883,
      "learning_rate": 1.3690895810783113e-06,
      "loss": 0.75059879,
      "num_input_tokens_seen": 110080360,
      "step": 5107,
      "time_per_iteration": 2.687053680419922
    },
    {
      "auxiliary_loss_clip": 0.01192567,
      "auxiliary_loss_mlp": 0.01208401,
      "balance_loss_clip": 1.04140353,
      "balance_loss_mlp": 1.0007112,
      "epoch": 0.6142006853844767,
      "flos": 21398199511680.0,
      "grad_norm": 2.067780874270299,
      "language_loss": 0.7166903,
      "learning_rate": 1.3683504312321543e-06,
      "loss": 0.74070001,
      "num_input_tokens_seen": 110100695,
      "step": 5108,
      "time_per_iteration": 2.685323715209961
    },
    {
      "auxiliary_loss_clip": 0.01179417,
      "auxiliary_loss_mlp": 0.01025549,
      "balance_loss_clip": 1.0484643,
      "balance_loss_mlp": 1.0173893,
      "epoch": 0.6143209282751158,
      "flos": 12057367622400.0,
      "grad_norm": 1.9702217700559712,
      "language_loss": 0.80302024,
      "learning_rate": 1.3676113772067355e-06,
      "loss": 0.82506984,
      "num_input_tokens_seen": 110117750,
      "step": 5109,
      "time_per_iteration": 3.5832741260528564
    },
    {
      "auxiliary_loss_clip": 0.01190376,
      "auxiliary_loss_mlp": 0.01028499,
      "balance_loss_clip": 1.04525208,
      "balance_loss_mlp": 1.02058387,
      "epoch": 0.6144411711657548,
      "flos": 25082274965760.0,
      "grad_norm": 2.210400036617186,
      "language_loss": 0.72212493,
      "learning_rate": 1.3668724191141671e-06,
      "loss": 0.74431366,
      "num_input_tokens_seen": 110137020,
      "step": 5110,
      "time_per_iteration": 2.6953399181365967
    },
    {
      "auxiliary_loss_clip": 0.01176176,
      "auxiliary_loss_mlp": 0.01032727,
      "balance_loss_clip": 1.05106831,
      "balance_loss_mlp": 1.02470398,
      "epoch": 0.6145614140563939,
      "flos": 20114069316480.0,
      "grad_norm": 2.5816155747159977,
      "language_loss": 0.66276264,
      "learning_rate": 1.3661335570665493e-06,
      "loss": 0.68485171,
      "num_input_tokens_seen": 110154930,
      "step": 5111,
      "time_per_iteration": 3.643441915512085
    },
    {
      "auxiliary_loss_clip": 0.0118328,
      "auxiliary_loss_mlp": 0.01029116,
      "balance_loss_clip": 1.04982924,
      "balance_loss_mlp": 1.02134681,
      "epoch": 0.614681656947033,
      "flos": 16800376953600.0,
      "grad_norm": 3.5184493963824255,
      "language_loss": 0.69456667,
      "learning_rate": 1.3653947911759676e-06,
      "loss": 0.7166906,
      "num_input_tokens_seen": 110172480,
      "step": 5112,
      "time_per_iteration": 2.6442673206329346
    },
    {
      "auxiliary_loss_clip": 0.01179159,
      "auxiliary_loss_mlp": 0.01032242,
      "balance_loss_clip": 1.0451932,
      "balance_loss_mlp": 1.02355206,
      "epoch": 0.6148018998376721,
      "flos": 38801587011840.0,
      "grad_norm": 1.5712723285701726,
      "language_loss": 0.74319583,
      "learning_rate": 1.3646561215544904e-06,
      "loss": 0.76530987,
      "num_input_tokens_seen": 110197120,
      "step": 5113,
      "time_per_iteration": 2.803130865097046
    },
    {
      "auxiliary_loss_clip": 0.01178657,
      "auxiliary_loss_mlp": 0.01027067,
      "balance_loss_clip": 1.05040944,
      "balance_loss_mlp": 1.01900291,
      "epoch": 0.6149221427283111,
      "flos": 23327032965120.0,
      "grad_norm": 2.158500472302463,
      "language_loss": 0.79193383,
      "learning_rate": 1.363917548314176e-06,
      "loss": 0.81399107,
      "num_input_tokens_seen": 110216385,
      "step": 5114,
      "time_per_iteration": 2.7337636947631836
    },
    {
      "auxiliary_loss_clip": 0.01182978,
      "auxiliary_loss_mlp": 0.0102749,
      "balance_loss_clip": 1.04956496,
      "balance_loss_mlp": 1.01946712,
      "epoch": 0.6150423856189503,
      "flos": 22379494141440.0,
      "grad_norm": 2.027629949511071,
      "language_loss": 0.73263627,
      "learning_rate": 1.3631790715670626e-06,
      "loss": 0.75474095,
      "num_input_tokens_seen": 110234790,
      "step": 5115,
      "time_per_iteration": 2.6690595149993896
    },
    {
      "auxiliary_loss_clip": 0.01181634,
      "auxiliary_loss_mlp": 0.01022886,
      "balance_loss_clip": 1.04449344,
      "balance_loss_mlp": 1.01539421,
      "epoch": 0.6151626285095894,
      "flos": 18692078722560.0,
      "grad_norm": 2.1801031699838025,
      "language_loss": 0.86038148,
      "learning_rate": 1.3624406914251783e-06,
      "loss": 0.88242662,
      "num_input_tokens_seen": 110251910,
      "step": 5116,
      "time_per_iteration": 2.701068162918091
    },
    {
      "auxiliary_loss_clip": 0.0117642,
      "auxiliary_loss_mlp": 0.01028267,
      "balance_loss_clip": 1.0475812,
      "balance_loss_mlp": 1.02122819,
      "epoch": 0.6152828714002284,
      "flos": 15851688894720.0,
      "grad_norm": 2.145802651888288,
      "language_loss": 0.88288563,
      "learning_rate": 1.3617024080005335e-06,
      "loss": 0.90493256,
      "num_input_tokens_seen": 110268810,
      "step": 5117,
      "time_per_iteration": 2.5702011585235596
    },
    {
      "auxiliary_loss_clip": 0.01181377,
      "auxiliary_loss_mlp": 0.0120816,
      "balance_loss_clip": 1.04668748,
      "balance_loss_mlp": 1.00082755,
      "epoch": 0.6154031142908676,
      "flos": 24869792062080.0,
      "grad_norm": 1.5692066019159487,
      "language_loss": 0.74428725,
      "learning_rate": 1.3609642214051266e-06,
      "loss": 0.76818258,
      "num_input_tokens_seen": 110293035,
      "step": 5118,
      "time_per_iteration": 3.626207113265991
    },
    {
      "auxiliary_loss_clip": 0.01171608,
      "auxiliary_loss_mlp": 0.01026769,
      "balance_loss_clip": 1.04775405,
      "balance_loss_mlp": 1.018502,
      "epoch": 0.6155233571815066,
      "flos": 19244744357760.0,
      "grad_norm": 3.3751296465293277,
      "language_loss": 0.66111147,
      "learning_rate": 1.3602261317509385e-06,
      "loss": 0.68309528,
      "num_input_tokens_seen": 110309695,
      "step": 5119,
      "time_per_iteration": 2.6120991706848145
    },
    {
      "auxiliary_loss_clip": 0.01178304,
      "auxiliary_loss_mlp": 0.01023832,
      "balance_loss_clip": 1.04912949,
      "balance_loss_mlp": 1.01521921,
      "epoch": 0.6156436000721457,
      "flos": 18770077105920.0,
      "grad_norm": 11.996611789992045,
      "language_loss": 0.83043599,
      "learning_rate": 1.3594881391499387e-06,
      "loss": 0.85245734,
      "num_input_tokens_seen": 110328610,
      "step": 5120,
      "time_per_iteration": 2.5679428577423096
    },
    {
      "auxiliary_loss_clip": 0.01182477,
      "auxiliary_loss_mlp": 0.01027878,
      "balance_loss_clip": 1.04918957,
      "balance_loss_mlp": 1.01973009,
      "epoch": 0.6157638429627849,
      "flos": 18041198325120.0,
      "grad_norm": 2.1576290878672286,
      "language_loss": 0.79102516,
      "learning_rate": 1.3587502437140778e-06,
      "loss": 0.81312871,
      "num_input_tokens_seen": 110346775,
      "step": 5121,
      "time_per_iteration": 2.6684539318084717
    },
    {
      "auxiliary_loss_clip": 0.01180696,
      "auxiliary_loss_mlp": 0.01028226,
      "balance_loss_clip": 1.04625797,
      "balance_loss_mlp": 1.01992297,
      "epoch": 0.6158840858534239,
      "flos": 25556726736000.0,
      "grad_norm": 2.7114882234423403,
      "language_loss": 0.85305023,
      "learning_rate": 1.3580124455552952e-06,
      "loss": 0.87513947,
      "num_input_tokens_seen": 110366140,
      "step": 5122,
      "time_per_iteration": 2.6593947410583496
    },
    {
      "auxiliary_loss_clip": 0.0117802,
      "auxiliary_loss_mlp": 0.01207576,
      "balance_loss_clip": 1.05039954,
      "balance_loss_mlp": 1.00088096,
      "epoch": 0.616004328744063,
      "flos": 24640788902400.0,
      "grad_norm": 2.152612600305351,
      "language_loss": 0.8748821,
      "learning_rate": 1.3572747447855148e-06,
      "loss": 0.89873803,
      "num_input_tokens_seen": 110386550,
      "step": 5123,
      "time_per_iteration": 2.6861391067504883
    },
    {
      "auxiliary_loss_clip": 0.01179713,
      "auxiliary_loss_mlp": 0.01025555,
      "balance_loss_clip": 1.05468297,
      "balance_loss_mlp": 1.01765096,
      "epoch": 0.6161245716347021,
      "flos": 21689686379520.0,
      "grad_norm": 2.96949033692014,
      "language_loss": 0.69435793,
      "learning_rate": 1.356537141516644e-06,
      "loss": 0.71641064,
      "num_input_tokens_seen": 110403970,
      "step": 5124,
      "time_per_iteration": 2.572235107421875
    },
    {
      "auxiliary_loss_clip": 0.01178047,
      "auxiliary_loss_mlp": 0.01025189,
      "balance_loss_clip": 1.05134511,
      "balance_loss_mlp": 1.01703537,
      "epoch": 0.6162448145253412,
      "flos": 35189225061120.0,
      "grad_norm": 2.293649042638785,
      "language_loss": 0.62105203,
      "learning_rate": 1.3557996358605775e-06,
      "loss": 0.64308441,
      "num_input_tokens_seen": 110423890,
      "step": 5125,
      "time_per_iteration": 2.724076986312866
    },
    {
      "auxiliary_loss_clip": 0.01176167,
      "auxiliary_loss_mlp": 0.01027765,
      "balance_loss_clip": 1.04898548,
      "balance_loss_mlp": 1.01990628,
      "epoch": 0.6163650574159802,
      "flos": 21615279356160.0,
      "grad_norm": 3.9406408042894023,
      "language_loss": 0.70411634,
      "learning_rate": 1.3550622279291941e-06,
      "loss": 0.7261557,
      "num_input_tokens_seen": 110442035,
      "step": 5126,
      "time_per_iteration": 2.584393262863159
    },
    {
      "auxiliary_loss_clip": 0.01174062,
      "auxiliary_loss_mlp": 0.01022607,
      "balance_loss_clip": 1.04188061,
      "balance_loss_mlp": 1.01457858,
      "epoch": 0.6164853003066194,
      "flos": 24572163968640.0,
      "grad_norm": 1.3696789050994074,
      "language_loss": 0.8360936,
      "learning_rate": 1.354324917834358e-06,
      "loss": 0.8580603,
      "num_input_tokens_seen": 110463280,
      "step": 5127,
      "time_per_iteration": 2.673964023590088
    },
    {
      "auxiliary_loss_clip": 0.01180101,
      "auxiliary_loss_mlp": 0.01208125,
      "balance_loss_clip": 1.0403651,
      "balance_loss_mlp": 1.0008837,
      "epoch": 0.6166055431972585,
      "flos": 21835986474240.0,
      "grad_norm": 2.177286115600125,
      "language_loss": 0.76884615,
      "learning_rate": 1.353587705687918e-06,
      "loss": 0.79272836,
      "num_input_tokens_seen": 110481455,
      "step": 5128,
      "time_per_iteration": 2.6585693359375
    },
    {
      "auxiliary_loss_clip": 0.0118395,
      "auxiliary_loss_mlp": 0.01027127,
      "balance_loss_clip": 1.05001545,
      "balance_loss_mlp": 1.01909852,
      "epoch": 0.6167257860878975,
      "flos": 17785262943360.0,
      "grad_norm": 2.761737221303858,
      "language_loss": 0.72153252,
      "learning_rate": 1.3528505916017096e-06,
      "loss": 0.74364328,
      "num_input_tokens_seen": 110499155,
      "step": 5129,
      "time_per_iteration": 2.699112892150879
    },
    {
      "auxiliary_loss_clip": 0.01176958,
      "auxiliary_loss_mlp": 0.01029838,
      "balance_loss_clip": 1.04819715,
      "balance_loss_mlp": 1.02158928,
      "epoch": 0.6168460289785367,
      "flos": 23214811898880.0,
      "grad_norm": 3.0031840282177953,
      "language_loss": 0.88764149,
      "learning_rate": 1.3521135756875514e-06,
      "loss": 0.90970945,
      "num_input_tokens_seen": 110515470,
      "step": 5130,
      "time_per_iteration": 2.5623714923858643
    },
    {
      "auxiliary_loss_clip": 0.01177049,
      "auxiliary_loss_mlp": 0.01024649,
      "balance_loss_clip": 1.04208636,
      "balance_loss_mlp": 1.01723146,
      "epoch": 0.6169662718691757,
      "flos": 26213281482240.0,
      "grad_norm": 1.647531538702532,
      "language_loss": 0.86520463,
      "learning_rate": 1.3513766580572496e-06,
      "loss": 0.88722157,
      "num_input_tokens_seen": 110538290,
      "step": 5131,
      "time_per_iteration": 3.6837692260742188
    },
    {
      "auxiliary_loss_clip": 0.01174648,
      "auxiliary_loss_mlp": 0.0102652,
      "balance_loss_clip": 1.04845285,
      "balance_loss_mlp": 1.01930439,
      "epoch": 0.6170865147598148,
      "flos": 19026120228480.0,
      "grad_norm": 8.619505137189442,
      "language_loss": 0.77683699,
      "learning_rate": 1.3506398388225924e-06,
      "loss": 0.79884869,
      "num_input_tokens_seen": 110555610,
      "step": 5132,
      "time_per_iteration": 2.5945889949798584
    },
    {
      "auxiliary_loss_clip": 0.01173005,
      "auxiliary_loss_mlp": 0.0102647,
      "balance_loss_clip": 1.05239463,
      "balance_loss_mlp": 1.01903164,
      "epoch": 0.617206757650454,
      "flos": 18260361158400.0,
      "grad_norm": 1.9996452381548155,
      "language_loss": 0.7204048,
      "learning_rate": 1.349903118095355e-06,
      "loss": 0.74239957,
      "num_input_tokens_seen": 110574745,
      "step": 5133,
      "time_per_iteration": 2.6340582370758057
    },
    {
      "auxiliary_loss_clip": 0.01179498,
      "auxiliary_loss_mlp": 0.01027485,
      "balance_loss_clip": 1.04922962,
      "balance_loss_mlp": 1.01965249,
      "epoch": 0.617327000541093,
      "flos": 18186959715840.0,
      "grad_norm": 1.6881268164186665,
      "language_loss": 0.73718548,
      "learning_rate": 1.349166495987298e-06,
      "loss": 0.75925535,
      "num_input_tokens_seen": 110593310,
      "step": 5134,
      "time_per_iteration": 2.6253669261932373
    },
    {
      "auxiliary_loss_clip": 0.01079852,
      "auxiliary_loss_mlp": 0.01000244,
      "balance_loss_clip": 1.01585257,
      "balance_loss_mlp": 0.99933833,
      "epoch": 0.6174472434317321,
      "flos": 61833796122240.0,
      "grad_norm": 0.8222390731958233,
      "language_loss": 0.6086337,
      "learning_rate": 1.348429972610166e-06,
      "loss": 0.62943465,
      "num_input_tokens_seen": 110657615,
      "step": 5135,
      "time_per_iteration": 4.16083288192749
    },
    {
      "auxiliary_loss_clip": 0.01088492,
      "auxiliary_loss_mlp": 0.01000347,
      "balance_loss_clip": 1.01552844,
      "balance_loss_mlp": 0.99951231,
      "epoch": 0.6175674863223712,
      "flos": 71230970494080.0,
      "grad_norm": 0.8470203919680583,
      "language_loss": 0.57828355,
      "learning_rate": 1.3476935480756897e-06,
      "loss": 0.59917194,
      "num_input_tokens_seen": 110714365,
      "step": 5136,
      "time_per_iteration": 3.0738842487335205
    },
    {
      "auxiliary_loss_clip": 0.01172856,
      "auxiliary_loss_mlp": 0.0103316,
      "balance_loss_clip": 1.04475689,
      "balance_loss_mlp": 1.02510726,
      "epoch": 0.6176877292130103,
      "flos": 21835447770240.0,
      "grad_norm": 2.5639120686007466,
      "language_loss": 0.75434649,
      "learning_rate": 1.346957222495583e-06,
      "loss": 0.77640659,
      "num_input_tokens_seen": 110732160,
      "step": 5137,
      "time_per_iteration": 2.715294599533081
    },
    {
      "auxiliary_loss_clip": 0.01184245,
      "auxiliary_loss_mlp": 0.01208389,
      "balance_loss_clip": 1.04997849,
      "balance_loss_mlp": 1.00099635,
      "epoch": 0.6178079721036493,
      "flos": 17741738638080.0,
      "grad_norm": 3.0368314519975055,
      "language_loss": 0.71580231,
      "learning_rate": 1.3462209959815466e-06,
      "loss": 0.73972869,
      "num_input_tokens_seen": 110746900,
      "step": 5138,
      "time_per_iteration": 3.573387384414673
    },
    {
      "auxiliary_loss_clip": 0.01182011,
      "auxiliary_loss_mlp": 0.01023894,
      "balance_loss_clip": 1.04956055,
      "balance_loss_mlp": 1.01593947,
      "epoch": 0.6179282149942885,
      "flos": 22633131052800.0,
      "grad_norm": 1.8446921615288272,
      "language_loss": 0.74412203,
      "learning_rate": 1.345484868645265e-06,
      "loss": 0.76618105,
      "num_input_tokens_seen": 110765710,
      "step": 5139,
      "time_per_iteration": 2.683450698852539
    },
    {
      "auxiliary_loss_clip": 0.01188225,
      "auxiliary_loss_mlp": 0.01027404,
      "balance_loss_clip": 1.04602242,
      "balance_loss_mlp": 1.0191009,
      "epoch": 0.6180484578849276,
      "flos": 22310330503680.0,
      "grad_norm": 2.2672312463954047,
      "language_loss": 0.78657067,
      "learning_rate": 1.3447488405984088e-06,
      "loss": 0.80872697,
      "num_input_tokens_seen": 110783970,
      "step": 5140,
      "time_per_iteration": 2.7159762382507324
    },
    {
      "auxiliary_loss_clip": 0.01177754,
      "auxiliary_loss_mlp": 0.01027045,
      "balance_loss_clip": 1.04727793,
      "balance_loss_mlp": 1.01930821,
      "epoch": 0.6181687007755666,
      "flos": 35225458905600.0,
      "grad_norm": 2.22507644072827,
      "language_loss": 0.70131552,
      "learning_rate": 1.3440129119526322e-06,
      "loss": 0.72336352,
      "num_input_tokens_seen": 110806395,
      "step": 5141,
      "time_per_iteration": 2.728471279144287
    },
    {
      "auxiliary_loss_clip": 0.0107152,
      "auxiliary_loss_mlp": 0.01001485,
      "balance_loss_clip": 1.01515138,
      "balance_loss_mlp": 1.00068629,
      "epoch": 0.6182889436662057,
      "flos": 61547370094080.0,
      "grad_norm": 0.8158846087753051,
      "language_loss": 0.5115751,
      "learning_rate": 1.3432770828195762e-06,
      "loss": 0.53230512,
      "num_input_tokens_seen": 110867380,
      "step": 5142,
      "time_per_iteration": 3.2643489837646484
    },
    {
      "auxiliary_loss_clip": 0.01171432,
      "auxiliary_loss_mlp": 0.01019618,
      "balance_loss_clip": 1.04375148,
      "balance_loss_mlp": 1.01142812,
      "epoch": 0.6184091865568448,
      "flos": 19609991804160.0,
      "grad_norm": 2.5013178553032387,
      "language_loss": 0.70530021,
      "learning_rate": 1.3425413533108635e-06,
      "loss": 0.7272107,
      "num_input_tokens_seen": 110885980,
      "step": 5143,
      "time_per_iteration": 2.628493547439575
    },
    {
      "auxiliary_loss_clip": 0.01184792,
      "auxiliary_loss_mlp": 0.01025232,
      "balance_loss_clip": 1.04748952,
      "balance_loss_mlp": 1.01793325,
      "epoch": 0.6185294294474839,
      "flos": 23586882929280.0,
      "grad_norm": 5.602068566905202,
      "language_loss": 0.70559871,
      "learning_rate": 1.341805723538105e-06,
      "loss": 0.72769898,
      "num_input_tokens_seen": 110906085,
      "step": 5144,
      "time_per_iteration": 3.5614840984344482
    },
    {
      "auxiliary_loss_clip": 0.01186313,
      "auxiliary_loss_mlp": 0.01026044,
      "balance_loss_clip": 1.0500586,
      "balance_loss_mlp": 1.0181464,
      "epoch": 0.618649672338123,
      "flos": 26762032535040.0,
      "grad_norm": 2.3214949218264285,
      "language_loss": 0.7728796,
      "learning_rate": 1.3410701936128948e-06,
      "loss": 0.79500318,
      "num_input_tokens_seen": 110928865,
      "step": 5145,
      "time_per_iteration": 2.63836669921875
    },
    {
      "auxiliary_loss_clip": 0.01178936,
      "auxiliary_loss_mlp": 0.01026121,
      "balance_loss_clip": 1.05301821,
      "balance_loss_mlp": 1.01850367,
      "epoch": 0.6187699152287621,
      "flos": 14456630522880.0,
      "grad_norm": 2.4836793542253086,
      "language_loss": 0.85156894,
      "learning_rate": 1.340334763646812e-06,
      "loss": 0.87361944,
      "num_input_tokens_seen": 110943000,
      "step": 5146,
      "time_per_iteration": 2.5954763889312744
    },
    {
      "auxiliary_loss_clip": 0.01177021,
      "auxiliary_loss_mlp": 0.01028593,
      "balance_loss_clip": 1.05161977,
      "balance_loss_mlp": 1.01972342,
      "epoch": 0.6188901581194012,
      "flos": 20084766796800.0,
      "grad_norm": 1.6960847630417328,
      "language_loss": 0.74419129,
      "learning_rate": 1.3395994337514218e-06,
      "loss": 0.76624739,
      "num_input_tokens_seen": 110963170,
      "step": 5147,
      "time_per_iteration": 2.6686854362487793
    },
    {
      "auxiliary_loss_clip": 0.01168321,
      "auxiliary_loss_mlp": 0.01027093,
      "balance_loss_clip": 1.04641867,
      "balance_loss_mlp": 1.01942837,
      "epoch": 0.6190104010100402,
      "flos": 25700728360320.0,
      "grad_norm": 1.766708387310731,
      "language_loss": 0.78591561,
      "learning_rate": 1.3388642040382725e-06,
      "loss": 0.80786973,
      "num_input_tokens_seen": 110983595,
      "step": 5148,
      "time_per_iteration": 2.6305925846099854
    },
    {
      "auxiliary_loss_clip": 0.01181992,
      "auxiliary_loss_mlp": 0.01024312,
      "balance_loss_clip": 1.04148531,
      "balance_loss_mlp": 1.01634908,
      "epoch": 0.6191306439006794,
      "flos": 30442372974720.0,
      "grad_norm": 1.8090120793575304,
      "language_loss": 0.8432672,
      "learning_rate": 1.3381290746188975e-06,
      "loss": 0.86533028,
      "num_input_tokens_seen": 111002965,
      "step": 5149,
      "time_per_iteration": 2.7397513389587402
    },
    {
      "auxiliary_loss_clip": 0.01178095,
      "auxiliary_loss_mlp": 0.01028933,
      "balance_loss_clip": 1.05175233,
      "balance_loss_mlp": 1.02075243,
      "epoch": 0.6192508867913185,
      "flos": 26685793918080.0,
      "grad_norm": 1.7317856061966082,
      "language_loss": 0.67370236,
      "learning_rate": 1.3373940456048152e-06,
      "loss": 0.69577265,
      "num_input_tokens_seen": 111022990,
      "step": 5150,
      "time_per_iteration": 2.618783712387085
    },
    {
      "auxiliary_loss_clip": 0.0117362,
      "auxiliary_loss_mlp": 0.01022542,
      "balance_loss_clip": 1.0522244,
      "balance_loss_mlp": 1.01502538,
      "epoch": 0.6193711296819575,
      "flos": 36722036090880.0,
      "grad_norm": 1.9016522179737714,
      "language_loss": 0.59594631,
      "learning_rate": 1.3366591171075299e-06,
      "loss": 0.61790794,
      "num_input_tokens_seen": 111046495,
      "step": 5151,
      "time_per_iteration": 2.7704389095306396
    },
    {
      "auxiliary_loss_clip": 0.01177798,
      "auxiliary_loss_mlp": 0.01023992,
      "balance_loss_clip": 1.04832423,
      "balance_loss_mlp": 1.01651716,
      "epoch": 0.6194913725725967,
      "flos": 25192556697600.0,
      "grad_norm": 2.150022472579411,
      "language_loss": 0.91040832,
      "learning_rate": 1.335924289238529e-06,
      "loss": 0.93242621,
      "num_input_tokens_seen": 111065705,
      "step": 5152,
      "time_per_iteration": 2.6307690143585205
    },
    {
      "auxiliary_loss_clip": 0.01176631,
      "auxiliary_loss_mlp": 0.01208147,
      "balance_loss_clip": 1.05327177,
      "balance_loss_mlp": 1.00095081,
      "epoch": 0.6196116154632357,
      "flos": 21178821196800.0,
      "grad_norm": 3.933917766606457,
      "language_loss": 0.76989996,
      "learning_rate": 1.3351895621092859e-06,
      "loss": 0.79374772,
      "num_input_tokens_seen": 111086050,
      "step": 5153,
      "time_per_iteration": 2.7324492931365967
    },
    {
      "auxiliary_loss_clip": 0.01175934,
      "auxiliary_loss_mlp": 0.01027455,
      "balance_loss_clip": 1.03271174,
      "balance_loss_mlp": 1.0198977,
      "epoch": 0.6197318583538748,
      "flos": 16253744803200.0,
      "grad_norm": 1.9241725823493268,
      "language_loss": 0.76732361,
      "learning_rate": 1.3344549358312567e-06,
      "loss": 0.78935742,
      "num_input_tokens_seen": 111104450,
      "step": 5154,
      "time_per_iteration": 2.7188711166381836
    },
    {
      "auxiliary_loss_clip": 0.01181884,
      "auxiliary_loss_mlp": 0.01024133,
      "balance_loss_clip": 1.05207324,
      "balance_loss_mlp": 1.01602662,
      "epoch": 0.619852101244514,
      "flos": 24425612478720.0,
      "grad_norm": 2.4237945808317742,
      "language_loss": 0.7839601,
      "learning_rate": 1.3337204105158852e-06,
      "loss": 0.80602026,
      "num_input_tokens_seen": 111123320,
      "step": 5155,
      "time_per_iteration": 2.651740789413452
    },
    {
      "auxiliary_loss_clip": 0.01165869,
      "auxiliary_loss_mlp": 0.01027543,
      "balance_loss_clip": 1.03797162,
      "balance_loss_mlp": 1.01940143,
      "epoch": 0.619972344135153,
      "flos": 16727298733440.0,
      "grad_norm": 2.826406861052497,
      "language_loss": 0.73115778,
      "learning_rate": 1.332985986274597e-06,
      "loss": 0.75309187,
      "num_input_tokens_seen": 111140950,
      "step": 5156,
      "time_per_iteration": 2.597003221511841
    },
    {
      "auxiliary_loss_clip": 0.01177939,
      "auxiliary_loss_mlp": 0.01207931,
      "balance_loss_clip": 1.04510033,
      "balance_loss_mlp": 1.00085568,
      "epoch": 0.6200925870257921,
      "flos": 12495190498560.0,
      "grad_norm": 2.1669703516173,
      "language_loss": 0.75175589,
      "learning_rate": 1.3322516632188047e-06,
      "loss": 0.77561462,
      "num_input_tokens_seen": 111157845,
      "step": 5157,
      "time_per_iteration": 2.7352523803710938
    },
    {
      "auxiliary_loss_clip": 0.01178941,
      "auxiliary_loss_mlp": 0.01026438,
      "balance_loss_clip": 1.04569626,
      "balance_loss_mlp": 1.01830757,
      "epoch": 0.6202128299164312,
      "flos": 26539350168960.0,
      "grad_norm": 1.7333935116660517,
      "language_loss": 0.67036587,
      "learning_rate": 1.3315174414599045e-06,
      "loss": 0.69241965,
      "num_input_tokens_seen": 111179165,
      "step": 5158,
      "time_per_iteration": 3.7405612468719482
    },
    {
      "auxiliary_loss_clip": 0.01171086,
      "auxiliary_loss_mlp": 0.01024328,
      "balance_loss_clip": 1.04627967,
      "balance_loss_mlp": 1.0158937,
      "epoch": 0.6203330728070703,
      "flos": 18770508069120.0,
      "grad_norm": 1.7555061654033097,
      "language_loss": 0.75634646,
      "learning_rate": 1.3307833211092768e-06,
      "loss": 0.77830064,
      "num_input_tokens_seen": 111197830,
      "step": 5159,
      "time_per_iteration": 2.604675054550171
    },
    {
      "auxiliary_loss_clip": 0.01176528,
      "auxiliary_loss_mlp": 0.01029443,
      "balance_loss_clip": 1.05369449,
      "balance_loss_mlp": 1.02164662,
      "epoch": 0.6204533156977093,
      "flos": 20629782835200.0,
      "grad_norm": 1.6427035014983824,
      "language_loss": 0.75496066,
      "learning_rate": 1.3300493022782873e-06,
      "loss": 0.77702034,
      "num_input_tokens_seen": 111218400,
      "step": 5160,
      "time_per_iteration": 2.6346049308776855
    },
    {
      "auxiliary_loss_clip": 0.01173395,
      "auxiliary_loss_mlp": 0.01208237,
      "balance_loss_clip": 1.04381883,
      "balance_loss_mlp": 1.0008688,
      "epoch": 0.6205735585883485,
      "flos": 17348050598400.0,
      "grad_norm": 2.078694504152402,
      "language_loss": 0.72798479,
      "learning_rate": 1.3293153850782855e-06,
      "loss": 0.75180113,
      "num_input_tokens_seen": 111236720,
      "step": 5161,
      "time_per_iteration": 2.6387381553649902
    },
    {
      "auxiliary_loss_clip": 0.01171771,
      "auxiliary_loss_mlp": 0.01026703,
      "balance_loss_clip": 1.04493701,
      "balance_loss_mlp": 1.01764929,
      "epoch": 0.6206938014789876,
      "flos": 22965017742720.0,
      "grad_norm": 2.18358450563384,
      "language_loss": 0.71174985,
      "learning_rate": 1.3285815696206069e-06,
      "loss": 0.73373461,
      "num_input_tokens_seen": 111258265,
      "step": 5162,
      "time_per_iteration": 3.5948777198791504
    },
    {
      "auxiliary_loss_clip": 0.01182703,
      "auxiliary_loss_mlp": 0.01032141,
      "balance_loss_clip": 1.04476285,
      "balance_loss_mlp": 1.02379036,
      "epoch": 0.6208140443696266,
      "flos": 23983192661760.0,
      "grad_norm": 2.0045181169796087,
      "language_loss": 0.77143866,
      "learning_rate": 1.32784785601657e-06,
      "loss": 0.79358709,
      "num_input_tokens_seen": 111277675,
      "step": 5163,
      "time_per_iteration": 2.646843671798706
    },
    {
      "auxiliary_loss_clip": 0.01179121,
      "auxiliary_loss_mlp": 0.01022078,
      "balance_loss_clip": 1.04556835,
      "balance_loss_mlp": 1.0146389,
      "epoch": 0.6209342872602658,
      "flos": 35077291303680.0,
      "grad_norm": 2.0930671220564716,
      "language_loss": 0.74077827,
      "learning_rate": 1.3271142443774798e-06,
      "loss": 0.7627902,
      "num_input_tokens_seen": 111299910,
      "step": 5164,
      "time_per_iteration": 2.8105571269989014
    },
    {
      "auxiliary_loss_clip": 0.01177022,
      "auxiliary_loss_mlp": 0.01022971,
      "balance_loss_clip": 1.04929352,
      "balance_loss_mlp": 1.01535368,
      "epoch": 0.6210545301509048,
      "flos": 26979327861120.0,
      "grad_norm": 4.0183846373579035,
      "language_loss": 0.82035351,
      "learning_rate": 1.3263807348146228e-06,
      "loss": 0.84235346,
      "num_input_tokens_seen": 111319765,
      "step": 5165,
      "time_per_iteration": 3.6439123153686523
    },
    {
      "auxiliary_loss_clip": 0.01175643,
      "auxiliary_loss_mlp": 0.01036181,
      "balance_loss_clip": 1.04493475,
      "balance_loss_mlp": 1.02756786,
      "epoch": 0.6211747730415439,
      "flos": 33618240852480.0,
      "grad_norm": 5.722264862391935,
      "language_loss": 0.73364151,
      "learning_rate": 1.3256473274392733e-06,
      "loss": 0.75575972,
      "num_input_tokens_seen": 111341110,
      "step": 5166,
      "time_per_iteration": 2.7449684143066406
    },
    {
      "auxiliary_loss_clip": 0.01172096,
      "auxiliary_loss_mlp": 0.01029077,
      "balance_loss_clip": 1.05015016,
      "balance_loss_mlp": 1.02105379,
      "epoch": 0.6212950159321831,
      "flos": 34167099646080.0,
      "grad_norm": 2.7237308204147475,
      "language_loss": 0.70084983,
      "learning_rate": 1.3249140223626873e-06,
      "loss": 0.72286153,
      "num_input_tokens_seen": 111362730,
      "step": 5167,
      "time_per_iteration": 2.7034380435943604
    },
    {
      "auxiliary_loss_clip": 0.01174107,
      "auxiliary_loss_mlp": 0.01020862,
      "balance_loss_clip": 1.04955125,
      "balance_loss_mlp": 1.01345336,
      "epoch": 0.6214152588228221,
      "flos": 27965758135680.0,
      "grad_norm": 1.809847719810538,
      "language_loss": 0.75641114,
      "learning_rate": 1.3241808196961077e-06,
      "loss": 0.77836084,
      "num_input_tokens_seen": 111383855,
      "step": 5168,
      "time_per_iteration": 2.6174888610839844
    },
    {
      "auxiliary_loss_clip": 0.01163691,
      "auxiliary_loss_mlp": 0.01024266,
      "balance_loss_clip": 1.04373145,
      "balance_loss_mlp": 1.0165472,
      "epoch": 0.6215355017134612,
      "flos": 20230204965120.0,
      "grad_norm": 1.793370767981051,
      "language_loss": 0.70921499,
      "learning_rate": 1.3234477195507608e-06,
      "loss": 0.7310946,
      "num_input_tokens_seen": 111402685,
      "step": 5169,
      "time_per_iteration": 2.678964138031006
    },
    {
      "auxiliary_loss_clip": 0.01181968,
      "auxiliary_loss_mlp": 0.0102531,
      "balance_loss_clip": 1.0474875,
      "balance_loss_mlp": 1.01816058,
      "epoch": 0.6216557446041003,
      "flos": 41428129219200.0,
      "grad_norm": 2.2111107282356075,
      "language_loss": 0.62847131,
      "learning_rate": 1.322714722037857e-06,
      "loss": 0.65054405,
      "num_input_tokens_seen": 111424130,
      "step": 5170,
      "time_per_iteration": 3.7613272666931152
    },
    {
      "auxiliary_loss_clip": 0.01188187,
      "auxiliary_loss_mlp": 0.01027713,
      "balance_loss_clip": 1.04673529,
      "balance_loss_mlp": 1.01966643,
      "epoch": 0.6217759874947394,
      "flos": 27928770105600.0,
      "grad_norm": 2.104661340144117,
      "language_loss": 0.77200812,
      "learning_rate": 1.321981827268591e-06,
      "loss": 0.7941671,
      "num_input_tokens_seen": 111444785,
      "step": 5171,
      "time_per_iteration": 2.74758243560791
    },
    {
      "auxiliary_loss_clip": 0.0118006,
      "auxiliary_loss_mlp": 0.01027368,
      "balance_loss_clip": 1.04750454,
      "balance_loss_mlp": 1.01959276,
      "epoch": 0.6218962303853784,
      "flos": 21765673601280.0,
      "grad_norm": 1.8526566250569705,
      "language_loss": 0.81420112,
      "learning_rate": 1.3212490353541426e-06,
      "loss": 0.83627546,
      "num_input_tokens_seen": 111467045,
      "step": 5172,
      "time_per_iteration": 2.6800239086151123
    },
    {
      "auxiliary_loss_clip": 0.01173416,
      "auxiliary_loss_mlp": 0.01024954,
      "balance_loss_clip": 1.04944515,
      "balance_loss_mlp": 1.016729,
      "epoch": 0.6220164732760175,
      "flos": 21246260981760.0,
      "grad_norm": 2.1017453074329633,
      "language_loss": 0.80315292,
      "learning_rate": 1.3205163464056762e-06,
      "loss": 0.8251366,
      "num_input_tokens_seen": 111483650,
      "step": 5173,
      "time_per_iteration": 2.5769739151000977
    },
    {
      "auxiliary_loss_clip": 0.01173363,
      "auxiliary_loss_mlp": 0.01028514,
      "balance_loss_clip": 1.04840684,
      "balance_loss_mlp": 1.02086067,
      "epoch": 0.6221367161666567,
      "flos": 26136360506880.0,
      "grad_norm": 1.8112537150909973,
      "language_loss": 0.72664243,
      "learning_rate": 1.319783760534339e-06,
      "loss": 0.74866122,
      "num_input_tokens_seen": 111502895,
      "step": 5174,
      "time_per_iteration": 2.656947135925293
    },
    {
      "auxiliary_loss_clip": 0.01177939,
      "auxiliary_loss_mlp": 0.01027039,
      "balance_loss_clip": 1.0521338,
      "balance_loss_mlp": 1.01882601,
      "epoch": 0.6222569590572957,
      "flos": 16284196558080.0,
      "grad_norm": 2.377279160505924,
      "language_loss": 0.75449598,
      "learning_rate": 1.319051277851266e-06,
      "loss": 0.77654576,
      "num_input_tokens_seen": 111519180,
      "step": 5175,
      "time_per_iteration": 2.5980725288391113
    },
    {
      "auxiliary_loss_clip": 0.01177523,
      "auxiliary_loss_mlp": 0.01028986,
      "balance_loss_clip": 1.04876816,
      "balance_loss_mlp": 1.02131462,
      "epoch": 0.6223772019479348,
      "flos": 18223840005120.0,
      "grad_norm": 2.3908788224299946,
      "language_loss": 0.84237605,
      "learning_rate": 1.3183188984675716e-06,
      "loss": 0.86444116,
      "num_input_tokens_seen": 111537545,
      "step": 5176,
      "time_per_iteration": 2.5813167095184326
    },
    {
      "auxiliary_loss_clip": 0.01178759,
      "auxiliary_loss_mlp": 0.01031437,
      "balance_loss_clip": 1.04990852,
      "balance_loss_mlp": 1.02389383,
      "epoch": 0.6224974448385739,
      "flos": 27489797994240.0,
      "grad_norm": 3.1883734208972623,
      "language_loss": 0.71277332,
      "learning_rate": 1.3175866224943586e-06,
      "loss": 0.73487532,
      "num_input_tokens_seen": 111556265,
      "step": 5177,
      "time_per_iteration": 2.6685822010040283
    },
    {
      "auxiliary_loss_clip": 0.01182831,
      "auxiliary_loss_mlp": 0.01028315,
      "balance_loss_clip": 1.04950643,
      "balance_loss_mlp": 1.02044702,
      "epoch": 0.622617687729213,
      "flos": 19791951125760.0,
      "grad_norm": 2.1986786229670887,
      "language_loss": 0.72990048,
      "learning_rate": 1.316854450042712e-06,
      "loss": 0.7520119,
      "num_input_tokens_seen": 111574205,
      "step": 5178,
      "time_per_iteration": 2.6171371936798096
    },
    {
      "auxiliary_loss_clip": 0.01182705,
      "auxiliary_loss_mlp": 0.0102403,
      "balance_loss_clip": 1.05160069,
      "balance_loss_mlp": 1.01643658,
      "epoch": 0.622737930619852,
      "flos": 23038886062080.0,
      "grad_norm": 1.925261368485557,
      "language_loss": 0.74351114,
      "learning_rate": 1.3161223812237024e-06,
      "loss": 0.76557851,
      "num_input_tokens_seen": 111593560,
      "step": 5179,
      "time_per_iteration": 2.594533681869507
    },
    {
      "auxiliary_loss_clip": 0.01171624,
      "auxiliary_loss_mlp": 0.0102849,
      "balance_loss_clip": 1.04844761,
      "balance_loss_mlp": 1.02057409,
      "epoch": 0.6228581735104912,
      "flos": 12634271959680.0,
      "grad_norm": 3.3793395743140073,
      "language_loss": 0.85258079,
      "learning_rate": 1.3153904161483842e-06,
      "loss": 0.87458187,
      "num_input_tokens_seen": 111608860,
      "step": 5180,
      "time_per_iteration": 2.602461576461792
    },
    {
      "auxiliary_loss_clip": 0.01176061,
      "auxiliary_loss_mlp": 0.0102625,
      "balance_loss_clip": 1.04372621,
      "balance_loss_mlp": 1.01772654,
      "epoch": 0.6229784164011303,
      "flos": 23802813538560.0,
      "grad_norm": 2.027431372769476,
      "language_loss": 0.8581844,
      "learning_rate": 1.3146585549277953e-06,
      "loss": 0.88020754,
      "num_input_tokens_seen": 111627500,
      "step": 5181,
      "time_per_iteration": 2.6574268341064453
    },
    {
      "auxiliary_loss_clip": 0.01188894,
      "auxiliary_loss_mlp": 0.01029079,
      "balance_loss_clip": 1.05088246,
      "balance_loss_mlp": 1.02113414,
      "epoch": 0.6230986592917693,
      "flos": 22414219614720.0,
      "grad_norm": 2.8746982808667783,
      "language_loss": 0.78039491,
      "learning_rate": 1.3139267976729591e-06,
      "loss": 0.80257463,
      "num_input_tokens_seen": 111647690,
      "step": 5182,
      "time_per_iteration": 2.6405370235443115
    },
    {
      "auxiliary_loss_clip": 0.01178819,
      "auxiliary_loss_mlp": 0.01025427,
      "balance_loss_clip": 1.05090535,
      "balance_loss_mlp": 1.01714754,
      "epoch": 0.6232189021824085,
      "flos": 34528217028480.0,
      "grad_norm": 2.99715721128012,
      "language_loss": 0.71969497,
      "learning_rate": 1.3131951444948815e-06,
      "loss": 0.74173743,
      "num_input_tokens_seen": 111667090,
      "step": 5183,
      "time_per_iteration": 2.7280242443084717
    },
    {
      "auxiliary_loss_clip": 0.01180597,
      "auxiliary_loss_mlp": 0.01027915,
      "balance_loss_clip": 1.04929078,
      "balance_loss_mlp": 1.02019,
      "epoch": 0.6233391450730476,
      "flos": 22237000888320.0,
      "grad_norm": 1.995397145794654,
      "language_loss": 0.76345587,
      "learning_rate": 1.3124635955045546e-06,
      "loss": 0.78554094,
      "num_input_tokens_seen": 111686905,
      "step": 5184,
      "time_per_iteration": 2.780122756958008
    },
    {
      "auxiliary_loss_clip": 0.011738,
      "auxiliary_loss_mlp": 0.01208286,
      "balance_loss_clip": 1.04184508,
      "balance_loss_mlp": 1.00088048,
      "epoch": 0.6234593879636866,
      "flos": 20332693445760.0,
      "grad_norm": 2.9948126065367533,
      "language_loss": 0.84455323,
      "learning_rate": 1.3117321508129537e-06,
      "loss": 0.86837411,
      "num_input_tokens_seen": 111704985,
      "step": 5185,
      "time_per_iteration": 3.7364962100982666
    },
    {
      "auxiliary_loss_clip": 0.01182391,
      "auxiliary_loss_mlp": 0.01023627,
      "balance_loss_clip": 1.04950213,
      "balance_loss_mlp": 1.01630712,
      "epoch": 0.6235796308543258,
      "flos": 20664903358080.0,
      "grad_norm": 1.506484942832343,
      "language_loss": 0.76650119,
      "learning_rate": 1.3110008105310388e-06,
      "loss": 0.78856146,
      "num_input_tokens_seen": 111724805,
      "step": 5186,
      "time_per_iteration": 2.749976873397827
    },
    {
      "auxiliary_loss_clip": 0.0117505,
      "auxiliary_loss_mlp": 0.01029726,
      "balance_loss_clip": 1.05030966,
      "balance_loss_mlp": 1.02176869,
      "epoch": 0.6236998737449648,
      "flos": 26618641441920.0,
      "grad_norm": 1.5784786098641101,
      "language_loss": 0.77943093,
      "learning_rate": 1.3102695747697526e-06,
      "loss": 0.80147862,
      "num_input_tokens_seen": 111747675,
      "step": 5187,
      "time_per_iteration": 2.662597894668579
    },
    {
      "auxiliary_loss_clip": 0.01183383,
      "auxiliary_loss_mlp": 0.01023828,
      "balance_loss_clip": 1.04522228,
      "balance_loss_mlp": 1.01593959,
      "epoch": 0.6238201166356039,
      "flos": 12674599954560.0,
      "grad_norm": 2.6881693575065873,
      "language_loss": 0.90576953,
      "learning_rate": 1.3095384436400237e-06,
      "loss": 0.9278416,
      "num_input_tokens_seen": 111759205,
      "step": 5188,
      "time_per_iteration": 2.6888065338134766
    },
    {
      "auxiliary_loss_clip": 0.01185164,
      "auxiliary_loss_mlp": 0.01021608,
      "balance_loss_clip": 1.04792941,
      "balance_loss_mlp": 1.0141933,
      "epoch": 0.623940359526243,
      "flos": 10452160730880.0,
      "grad_norm": 2.123824198064182,
      "language_loss": 0.82395005,
      "learning_rate": 1.3088074172527633e-06,
      "loss": 0.84601784,
      "num_input_tokens_seen": 111776335,
      "step": 5189,
      "time_per_iteration": 3.5871756076812744
    },
    {
      "auxiliary_loss_clip": 0.01180743,
      "auxiliary_loss_mlp": 0.01024812,
      "balance_loss_clip": 1.04630148,
      "balance_loss_mlp": 1.01669955,
      "epoch": 0.6240606024168821,
      "flos": 29059525226880.0,
      "grad_norm": 43.42682036822351,
      "language_loss": 0.71007597,
      "learning_rate": 1.3080764957188684e-06,
      "loss": 0.73213148,
      "num_input_tokens_seen": 111796580,
      "step": 5190,
      "time_per_iteration": 2.672593832015991
    },
    {
      "auxiliary_loss_clip": 0.01186539,
      "auxiliary_loss_mlp": 0.01025137,
      "balance_loss_clip": 1.04433155,
      "balance_loss_mlp": 1.01746595,
      "epoch": 0.6241808453075212,
      "flos": 22018089450240.0,
      "grad_norm": 2.000238444425528,
      "language_loss": 0.70950228,
      "learning_rate": 1.3073456791492192e-06,
      "loss": 0.731619,
      "num_input_tokens_seen": 111816290,
      "step": 5191,
      "time_per_iteration": 2.6437954902648926
    },
    {
      "auxiliary_loss_clip": 0.01179055,
      "auxiliary_loss_mlp": 0.01025217,
      "balance_loss_clip": 1.04577839,
      "balance_loss_mlp": 1.01767135,
      "epoch": 0.6243010881981603,
      "flos": 21138708683520.0,
      "grad_norm": 1.9516816960806773,
      "language_loss": 0.7814554,
      "learning_rate": 1.3066149676546801e-06,
      "loss": 0.80349815,
      "num_input_tokens_seen": 111834470,
      "step": 5192,
      "time_per_iteration": 3.584095001220703
    },
    {
      "auxiliary_loss_clip": 0.01173766,
      "auxiliary_loss_mlp": 0.01024311,
      "balance_loss_clip": 1.04811168,
      "balance_loss_mlp": 1.01672077,
      "epoch": 0.6244213310887994,
      "flos": 22344948236160.0,
      "grad_norm": 3.1872435706412485,
      "language_loss": 0.65983123,
      "learning_rate": 1.3058843613460985e-06,
      "loss": 0.68181205,
      "num_input_tokens_seen": 111852410,
      "step": 5193,
      "time_per_iteration": 2.620347023010254
    },
    {
      "auxiliary_loss_clip": 0.01189373,
      "auxiliary_loss_mlp": 0.01024435,
      "balance_loss_clip": 1.04761291,
      "balance_loss_mlp": 1.01673365,
      "epoch": 0.6245415739794384,
      "flos": 15231978524160.0,
      "grad_norm": 2.1482056072564975,
      "language_loss": 0.74561226,
      "learning_rate": 1.3051538603343075e-06,
      "loss": 0.76775032,
      "num_input_tokens_seen": 111870340,
      "step": 5194,
      "time_per_iteration": 2.657252311706543
    },
    {
      "auxiliary_loss_clip": 0.01175786,
      "auxiliary_loss_mlp": 0.01033403,
      "balance_loss_clip": 1.04991245,
      "balance_loss_mlp": 1.02587509,
      "epoch": 0.6246618168700776,
      "flos": 18879891960960.0,
      "grad_norm": 2.056865087268236,
      "language_loss": 0.68006563,
      "learning_rate": 1.3044234647301235e-06,
      "loss": 0.7021575,
      "num_input_tokens_seen": 111888365,
      "step": 5195,
      "time_per_iteration": 2.6115760803222656
    },
    {
      "auxiliary_loss_clip": 0.01170746,
      "auxiliary_loss_mlp": 0.01026362,
      "balance_loss_clip": 1.04869211,
      "balance_loss_mlp": 1.01960015,
      "epoch": 0.6247820597607167,
      "flos": 14319201087360.0,
      "grad_norm": 1.8256605583969021,
      "language_loss": 0.72715229,
      "learning_rate": 1.303693174644347e-06,
      "loss": 0.74912339,
      "num_input_tokens_seen": 111905840,
      "step": 5196,
      "time_per_iteration": 2.7267277240753174
    },
    {
      "auxiliary_loss_clip": 0.01174447,
      "auxiliary_loss_mlp": 0.01026665,
      "balance_loss_clip": 1.04564345,
      "balance_loss_mlp": 1.01812959,
      "epoch": 0.6249023026513557,
      "flos": 22637979388800.0,
      "grad_norm": 1.8806383028434488,
      "language_loss": 0.80729675,
      "learning_rate": 1.3029629901877625e-06,
      "loss": 0.82930779,
      "num_input_tokens_seen": 111925215,
      "step": 5197,
      "time_per_iteration": 3.618350028991699
    },
    {
      "auxiliary_loss_clip": 0.01184136,
      "auxiliary_loss_mlp": 0.01024883,
      "balance_loss_clip": 1.05209661,
      "balance_loss_mlp": 1.01668131,
      "epoch": 0.6250225455419949,
      "flos": 20266690204800.0,
      "grad_norm": 2.734121438945783,
      "language_loss": 0.77438426,
      "learning_rate": 1.3022329114711376e-06,
      "loss": 0.79647446,
      "num_input_tokens_seen": 111943925,
      "step": 5198,
      "time_per_iteration": 2.585970401763916
    },
    {
      "auxiliary_loss_clip": 0.01175337,
      "auxiliary_loss_mlp": 0.0102167,
      "balance_loss_clip": 1.04719591,
      "balance_loss_mlp": 1.01383841,
      "epoch": 0.6251427884326339,
      "flos": 23437853400960.0,
      "grad_norm": 2.36786466496309,
      "language_loss": 0.6920048,
      "learning_rate": 1.3015029386052256e-06,
      "loss": 0.71397495,
      "num_input_tokens_seen": 111964095,
      "step": 5199,
      "time_per_iteration": 2.6531033515930176
    },
    {
      "auxiliary_loss_clip": 0.01191578,
      "auxiliary_loss_mlp": 0.01029282,
      "balance_loss_clip": 1.04897702,
      "balance_loss_mlp": 1.02141976,
      "epoch": 0.625263031323273,
      "flos": 31723055464320.0,
      "grad_norm": 2.5318366891475814,
      "language_loss": 0.73061776,
      "learning_rate": 1.3007730717007622e-06,
      "loss": 0.75282633,
      "num_input_tokens_seen": 111984910,
      "step": 5200,
      "time_per_iteration": 2.6906445026397705
    },
    {
      "auxiliary_loss_clip": 0.01174945,
      "auxiliary_loss_mlp": 0.0102697,
      "balance_loss_clip": 1.0511291,
      "balance_loss_mlp": 1.01872683,
      "epoch": 0.6253832742139122,
      "flos": 24134341092480.0,
      "grad_norm": 2.087085156619945,
      "language_loss": 0.75617415,
      "learning_rate": 1.3000433108684676e-06,
      "loss": 0.7781933,
      "num_input_tokens_seen": 112005410,
      "step": 5201,
      "time_per_iteration": 2.614255666732788
    },
    {
      "auxiliary_loss_clip": 0.01168984,
      "auxiliary_loss_mlp": 0.01021333,
      "balance_loss_clip": 1.04711533,
      "balance_loss_mlp": 1.01360202,
      "epoch": 0.6255035171045512,
      "flos": 27668812400640.0,
      "grad_norm": 16.246944396540343,
      "language_loss": 0.80596763,
      "learning_rate": 1.2993136562190467e-06,
      "loss": 0.82787085,
      "num_input_tokens_seen": 112024530,
      "step": 5202,
      "time_per_iteration": 2.6940855979919434
    },
    {
      "auxiliary_loss_clip": 0.01181498,
      "auxiliary_loss_mlp": 0.01028665,
      "balance_loss_clip": 1.04691935,
      "balance_loss_mlp": 1.02050471,
      "epoch": 0.6256237599951903,
      "flos": 20227798753920.0,
      "grad_norm": 1.6035206777502458,
      "language_loss": 0.70489991,
      "learning_rate": 1.2985841078631871e-06,
      "loss": 0.72700155,
      "num_input_tokens_seen": 112043850,
      "step": 5203,
      "time_per_iteration": 2.7027249336242676
    },
    {
      "auxiliary_loss_clip": 0.01182842,
      "auxiliary_loss_mlp": 0.01030696,
      "balance_loss_clip": 1.0401777,
      "balance_loss_mlp": 1.02267301,
      "epoch": 0.6257440028858293,
      "flos": 24170574936960.0,
      "grad_norm": 2.1389916776803988,
      "language_loss": 0.78393239,
      "learning_rate": 1.2978546659115608e-06,
      "loss": 0.80606776,
      "num_input_tokens_seen": 112061930,
      "step": 5204,
      "time_per_iteration": 2.6979238986968994
    },
    {
      "auxiliary_loss_clip": 0.01181601,
      "auxiliary_loss_mlp": 0.01027133,
      "balance_loss_clip": 1.04880762,
      "balance_loss_mlp": 1.01931262,
      "epoch": 0.6258642457764685,
      "flos": 15851940289920.0,
      "grad_norm": 2.0755830608271113,
      "language_loss": 0.85805571,
      "learning_rate": 1.2971253304748228e-06,
      "loss": 0.88014305,
      "num_input_tokens_seen": 112079645,
      "step": 5205,
      "time_per_iteration": 2.5589144229888916
    },
    {
      "auxiliary_loss_clip": 0.01182595,
      "auxiliary_loss_mlp": 0.01027795,
      "balance_loss_clip": 1.05221224,
      "balance_loss_mlp": 1.01964712,
      "epoch": 0.6259844886671075,
      "flos": 11911354836480.0,
      "grad_norm": 2.1143505401163702,
      "language_loss": 0.75175965,
      "learning_rate": 1.296396101663614e-06,
      "loss": 0.77386355,
      "num_input_tokens_seen": 112096205,
      "step": 5206,
      "time_per_iteration": 2.5705342292785645
    },
    {
      "auxiliary_loss_clip": 0.01178851,
      "auxiliary_loss_mlp": 0.01025253,
      "balance_loss_clip": 1.04951119,
      "balance_loss_mlp": 1.01722455,
      "epoch": 0.6261047315577466,
      "flos": 15887958652800.0,
      "grad_norm": 3.025295472802595,
      "language_loss": 0.84171653,
      "learning_rate": 1.2956669795885565e-06,
      "loss": 0.86375755,
      "num_input_tokens_seen": 112112835,
      "step": 5207,
      "time_per_iteration": 2.6488828659057617
    },
    {
      "auxiliary_loss_clip": 0.01175187,
      "auxiliary_loss_mlp": 0.01030133,
      "balance_loss_clip": 1.04663742,
      "balance_loss_mlp": 1.02214587,
      "epoch": 0.6262249744483858,
      "flos": 31248926916480.0,
      "grad_norm": 2.0466151805670005,
      "language_loss": 0.68255138,
      "learning_rate": 1.294937964360259e-06,
      "loss": 0.70460463,
      "num_input_tokens_seen": 112133105,
      "step": 5208,
      "time_per_iteration": 2.7489054203033447
    },
    {
      "auxiliary_loss_clip": 0.01183888,
      "auxiliary_loss_mlp": 0.01031689,
      "balance_loss_clip": 1.04678321,
      "balance_loss_mlp": 1.0223546,
      "epoch": 0.6263452173390248,
      "flos": 27198598435200.0,
      "grad_norm": 2.8541943539773955,
      "language_loss": 0.7138831,
      "learning_rate": 1.2942090560893108e-06,
      "loss": 0.73603892,
      "num_input_tokens_seen": 112152510,
      "step": 5209,
      "time_per_iteration": 2.6678106784820557
    },
    {
      "auxiliary_loss_clip": 0.0117315,
      "auxiliary_loss_mlp": 0.010251,
      "balance_loss_clip": 1.05048275,
      "balance_loss_mlp": 1.01791203,
      "epoch": 0.6264654602296639,
      "flos": 37342069683840.0,
      "grad_norm": 2.9943014554134995,
      "language_loss": 0.60770726,
      "learning_rate": 1.2934802548862882e-06,
      "loss": 0.62968981,
      "num_input_tokens_seen": 112175295,
      "step": 5210,
      "time_per_iteration": 2.7550318241119385
    },
    {
      "auxiliary_loss_clip": 0.01175349,
      "auxiliary_loss_mlp": 0.0102664,
      "balance_loss_clip": 1.04573822,
      "balance_loss_mlp": 1.0191716,
      "epoch": 0.626585703120303,
      "flos": 14756952136320.0,
      "grad_norm": 2.0464528523004764,
      "language_loss": 0.82644027,
      "learning_rate": 1.292751560861749e-06,
      "loss": 0.84846014,
      "num_input_tokens_seen": 112190200,
      "step": 5211,
      "time_per_iteration": 2.5877180099487305
    },
    {
      "auxiliary_loss_clip": 0.01175173,
      "auxiliary_loss_mlp": 0.01026129,
      "balance_loss_clip": 1.05094206,
      "balance_loss_mlp": 1.01777291,
      "epoch": 0.6267059460109421,
      "flos": 22347318533760.0,
      "grad_norm": 2.0881876804747623,
      "language_loss": 0.79709083,
      "learning_rate": 1.2920229741262354e-06,
      "loss": 0.81910384,
      "num_input_tokens_seen": 112208205,
      "step": 5212,
      "time_per_iteration": 3.567110538482666
    },
    {
      "auxiliary_loss_clip": 0.01179311,
      "auxiliary_loss_mlp": 0.01025391,
      "balance_loss_clip": 1.04607272,
      "balance_loss_mlp": 1.01757121,
      "epoch": 0.6268261889015811,
      "flos": 17748813617280.0,
      "grad_norm": 3.534588109944011,
      "language_loss": 0.7542997,
      "learning_rate": 1.2912944947902739e-06,
      "loss": 0.77634668,
      "num_input_tokens_seen": 112224690,
      "step": 5213,
      "time_per_iteration": 2.6532986164093018
    },
    {
      "auxiliary_loss_clip": 0.01183712,
      "auxiliary_loss_mlp": 0.01023809,
      "balance_loss_clip": 1.04801154,
      "balance_loss_mlp": 1.01529765,
      "epoch": 0.6269464317922203,
      "flos": 32846484211200.0,
      "grad_norm": 2.4775088773888974,
      "language_loss": 0.71634054,
      "learning_rate": 1.2905661229643742e-06,
      "loss": 0.73841584,
      "num_input_tokens_seen": 112244450,
      "step": 5214,
      "time_per_iteration": 2.7083685398101807
    },
    {
      "auxiliary_loss_clip": 0.01174171,
      "auxiliary_loss_mlp": 0.01026762,
      "balance_loss_clip": 1.04929531,
      "balance_loss_mlp": 1.01869738,
      "epoch": 0.6270666746828594,
      "flos": 17929192740480.0,
      "grad_norm": 2.2552663249501435,
      "language_loss": 0.84453118,
      "learning_rate": 1.2898378587590299e-06,
      "loss": 0.86654055,
      "num_input_tokens_seen": 112261050,
      "step": 5215,
      "time_per_iteration": 2.582054853439331
    },
    {
      "auxiliary_loss_clip": 0.01172943,
      "auxiliary_loss_mlp": 0.01022507,
      "balance_loss_clip": 1.04997206,
      "balance_loss_mlp": 1.0148952,
      "epoch": 0.6271869175734984,
      "flos": 17457326749440.0,
      "grad_norm": 1.7757945143030391,
      "language_loss": 0.87384093,
      "learning_rate": 1.2891097022847173e-06,
      "loss": 0.8957954,
      "num_input_tokens_seen": 112278395,
      "step": 5216,
      "time_per_iteration": 3.498952627182007
    },
    {
      "auxiliary_loss_clip": 0.01179841,
      "auxiliary_loss_mlp": 0.01031209,
      "balance_loss_clip": 1.04739857,
      "balance_loss_mlp": 1.02235186,
      "epoch": 0.6273071604641376,
      "flos": 26868615166080.0,
      "grad_norm": 1.8873172594536405,
      "language_loss": 0.66912901,
      "learning_rate": 1.2883816536518978e-06,
      "loss": 0.6912396,
      "num_input_tokens_seen": 112299535,
      "step": 5217,
      "time_per_iteration": 2.6634774208068848
    },
    {
      "auxiliary_loss_clip": 0.01172682,
      "auxiliary_loss_mlp": 0.01025244,
      "balance_loss_clip": 1.04692733,
      "balance_loss_mlp": 1.01798403,
      "epoch": 0.6274274033547766,
      "flos": 26062384446720.0,
      "grad_norm": 2.5453395625661077,
      "language_loss": 0.82090616,
      "learning_rate": 1.2876537129710155e-06,
      "loss": 0.84288538,
      "num_input_tokens_seen": 112317265,
      "step": 5218,
      "time_per_iteration": 3.614708423614502
    },
    {
      "auxiliary_loss_clip": 0.01174627,
      "auxiliary_loss_mlp": 0.01030044,
      "balance_loss_clip": 1.05037999,
      "balance_loss_mlp": 1.02250385,
      "epoch": 0.6275476462454157,
      "flos": 20266259241600.0,
      "grad_norm": 2.4813346740152116,
      "language_loss": 0.75251108,
      "learning_rate": 1.286925880352499e-06,
      "loss": 0.77455783,
      "num_input_tokens_seen": 112336125,
      "step": 5219,
      "time_per_iteration": 2.607748031616211
    },
    {
      "auxiliary_loss_clip": 0.01173412,
      "auxiliary_loss_mlp": 0.01019984,
      "balance_loss_clip": 1.0449276,
      "balance_loss_mlp": 1.01258659,
      "epoch": 0.6276678891360549,
      "flos": 26320402817280.0,
      "grad_norm": 1.8553719618669973,
      "language_loss": 0.71186876,
      "learning_rate": 1.2861981559067592e-06,
      "loss": 0.73380268,
      "num_input_tokens_seen": 112356730,
      "step": 5220,
      "time_per_iteration": 2.6835837364196777
    },
    {
      "auxiliary_loss_clip": 0.01175817,
      "auxiliary_loss_mlp": 0.01022474,
      "balance_loss_clip": 1.04360366,
      "balance_loss_mlp": 1.01506519,
      "epoch": 0.6277881320266939,
      "flos": 13912512324480.0,
      "grad_norm": 9.096416998518936,
      "language_loss": 0.80263382,
      "learning_rate": 1.2854705397441917e-06,
      "loss": 0.82461679,
      "num_input_tokens_seen": 112372270,
      "step": 5221,
      "time_per_iteration": 2.7362513542175293
    },
    {
      "auxiliary_loss_clip": 0.01174203,
      "auxiliary_loss_mlp": 0.01026884,
      "balance_loss_clip": 1.04189956,
      "balance_loss_mlp": 1.01904035,
      "epoch": 0.627908374917333,
      "flos": 27048922462080.0,
      "grad_norm": 2.3641415474082277,
      "language_loss": 0.77725583,
      "learning_rate": 1.2847430319751747e-06,
      "loss": 0.7992667,
      "num_input_tokens_seen": 112390365,
      "step": 5222,
      "time_per_iteration": 2.7332119941711426
    },
    {
      "auxiliary_loss_clip": 0.01169419,
      "auxiliary_loss_mlp": 0.01024011,
      "balance_loss_clip": 1.04870915,
      "balance_loss_mlp": 1.01667404,
      "epoch": 0.6280286178079721,
      "flos": 23769201386880.0,
      "grad_norm": 2.2739576793863954,
      "language_loss": 0.67396665,
      "learning_rate": 1.2840156327100712e-06,
      "loss": 0.69590098,
      "num_input_tokens_seen": 112407490,
      "step": 5223,
      "time_per_iteration": 2.6375389099121094
    },
    {
      "auxiliary_loss_clip": 0.01171819,
      "auxiliary_loss_mlp": 0.01021131,
      "balance_loss_clip": 1.05010915,
      "balance_loss_mlp": 1.01332271,
      "epoch": 0.6281488606986112,
      "flos": 26359150613760.0,
      "grad_norm": 2.16192237303334,
      "language_loss": 0.72593677,
      "learning_rate": 1.2832883420592272e-06,
      "loss": 0.74786627,
      "num_input_tokens_seen": 112426385,
      "step": 5224,
      "time_per_iteration": 3.536531686782837
    },
    {
      "auxiliary_loss_clip": 0.01174402,
      "auxiliary_loss_mlp": 0.01026115,
      "balance_loss_clip": 1.04686236,
      "balance_loss_mlp": 1.01788044,
      "epoch": 0.6282691035892503,
      "flos": 36137194848000.0,
      "grad_norm": 3.348650542081945,
      "language_loss": 0.64493966,
      "learning_rate": 1.282561160132972e-06,
      "loss": 0.66694486,
      "num_input_tokens_seen": 112446905,
      "step": 5225,
      "time_per_iteration": 2.7417454719543457
    },
    {
      "auxiliary_loss_clip": 0.01182981,
      "auxiliary_loss_mlp": 0.01034903,
      "balance_loss_clip": 1.04483414,
      "balance_loss_mlp": 1.02697003,
      "epoch": 0.6283893464798894,
      "flos": 26537231266560.0,
      "grad_norm": 1.6283692951641617,
      "language_loss": 0.80790728,
      "learning_rate": 1.2818340870416186e-06,
      "loss": 0.83008611,
      "num_input_tokens_seen": 112468040,
      "step": 5226,
      "time_per_iteration": 2.6966490745544434
    },
    {
      "auxiliary_loss_clip": 0.01188552,
      "auxiliary_loss_mlp": 0.01027778,
      "balance_loss_clip": 1.04600203,
      "balance_loss_mlp": 1.01920068,
      "epoch": 0.6285095893705285,
      "flos": 22237216369920.0,
      "grad_norm": 1.8683509852381237,
      "language_loss": 0.75794214,
      "learning_rate": 1.2811071228954626e-06,
      "loss": 0.78010541,
      "num_input_tokens_seen": 112486675,
      "step": 5227,
      "time_per_iteration": 2.625250816345215
    },
    {
      "auxiliary_loss_clip": 0.01177411,
      "auxiliary_loss_mlp": 0.0102377,
      "balance_loss_clip": 1.04894686,
      "balance_loss_mlp": 1.01600957,
      "epoch": 0.6286298322611675,
      "flos": 26542259170560.0,
      "grad_norm": 1.9673013735140468,
      "language_loss": 0.81074822,
      "learning_rate": 1.2803802678047846e-06,
      "loss": 0.8327601,
      "num_input_tokens_seen": 112506825,
      "step": 5228,
      "time_per_iteration": 2.6638405323028564
    },
    {
      "auxiliary_loss_clip": 0.01184949,
      "auxiliary_loss_mlp": 0.01032226,
      "balance_loss_clip": 1.05018342,
      "balance_loss_mlp": 1.02392316,
      "epoch": 0.6287500751518067,
      "flos": 21795227516160.0,
      "grad_norm": 1.871624456903156,
      "language_loss": 0.73976701,
      "learning_rate": 1.279653521879848e-06,
      "loss": 0.76193875,
      "num_input_tokens_seen": 112526890,
      "step": 5229,
      "time_per_iteration": 2.682034969329834
    },
    {
      "auxiliary_loss_clip": 0.01178452,
      "auxiliary_loss_mlp": 0.01025511,
      "balance_loss_clip": 1.0391624,
      "balance_loss_mlp": 1.01844454,
      "epoch": 0.6288703180424458,
      "flos": 20009605587840.0,
      "grad_norm": 2.12921025847067,
      "language_loss": 0.84039068,
      "learning_rate": 1.2789268852308997e-06,
      "loss": 0.86243033,
      "num_input_tokens_seen": 112542100,
      "step": 5230,
      "time_per_iteration": 2.794175386428833
    },
    {
      "auxiliary_loss_clip": 0.01169194,
      "auxiliary_loss_mlp": 0.01028359,
      "balance_loss_clip": 1.0480386,
      "balance_loss_mlp": 1.02033591,
      "epoch": 0.6289905609330848,
      "flos": 22124923476480.0,
      "grad_norm": 1.9187830296851756,
      "language_loss": 0.70513904,
      "learning_rate": 1.2782003579681688e-06,
      "loss": 0.72711456,
      "num_input_tokens_seen": 112561630,
      "step": 5231,
      "time_per_iteration": 2.694038152694702
    },
    {
      "auxiliary_loss_clip": 0.01176352,
      "auxiliary_loss_mlp": 0.01027696,
      "balance_loss_clip": 1.05202579,
      "balance_loss_mlp": 1.01940477,
      "epoch": 0.629110803823724,
      "flos": 25518481729920.0,
      "grad_norm": 1.733398789430357,
      "language_loss": 0.74442023,
      "learning_rate": 1.2774739402018701e-06,
      "loss": 0.76646072,
      "num_input_tokens_seen": 112582465,
      "step": 5232,
      "time_per_iteration": 2.613389730453491
    },
    {
      "auxiliary_loss_clip": 0.01173813,
      "auxiliary_loss_mlp": 0.01027346,
      "balance_loss_clip": 1.04984999,
      "balance_loss_mlp": 1.01877475,
      "epoch": 0.629231046714363,
      "flos": 20886616056960.0,
      "grad_norm": 2.3685826386949964,
      "language_loss": 0.73318064,
      "learning_rate": 1.2767476320422002e-06,
      "loss": 0.75519222,
      "num_input_tokens_seen": 112602390,
      "step": 5233,
      "time_per_iteration": 2.627600908279419
    },
    {
      "auxiliary_loss_clip": 0.01094144,
      "auxiliary_loss_mlp": 0.01002609,
      "balance_loss_clip": 1.01565123,
      "balance_loss_mlp": 1.00186956,
      "epoch": 0.6293512896050021,
      "flos": 65050027908480.0,
      "grad_norm": 0.6916542781113169,
      "language_loss": 0.57147837,
      "learning_rate": 1.2760214335993392e-06,
      "loss": 0.59244585,
      "num_input_tokens_seen": 112669035,
      "step": 5234,
      "time_per_iteration": 3.2362804412841797
    },
    {
      "auxiliary_loss_clip": 0.01167887,
      "auxiliary_loss_mlp": 0.01025696,
      "balance_loss_clip": 1.04601657,
      "balance_loss_mlp": 1.01863551,
      "epoch": 0.6294715324956413,
      "flos": 34677857088000.0,
      "grad_norm": 2.173126668110854,
      "language_loss": 0.5883916,
      "learning_rate": 1.2752953449834514e-06,
      "loss": 0.61032742,
      "num_input_tokens_seen": 112691485,
      "step": 5235,
      "time_per_iteration": 2.6837258338928223
    },
    {
      "auxiliary_loss_clip": 0.01173489,
      "auxiliary_loss_mlp": 0.01026168,
      "balance_loss_clip": 1.05121195,
      "balance_loss_mlp": 1.01896763,
      "epoch": 0.6295917753862803,
      "flos": 22784207656320.0,
      "grad_norm": 2.5859879868800792,
      "language_loss": 0.80231392,
      "learning_rate": 1.2745693663046836e-06,
      "loss": 0.82431054,
      "num_input_tokens_seen": 112710555,
      "step": 5236,
      "time_per_iteration": 2.550565481185913
    },
    {
      "auxiliary_loss_clip": 0.01171159,
      "auxiliary_loss_mlp": 0.01022464,
      "balance_loss_clip": 1.04753745,
      "balance_loss_mlp": 1.01556504,
      "epoch": 0.6297120182769194,
      "flos": 20850454039680.0,
      "grad_norm": 1.917670121641944,
      "language_loss": 0.80682552,
      "learning_rate": 1.2738434976731662e-06,
      "loss": 0.82876176,
      "num_input_tokens_seen": 112728740,
      "step": 5237,
      "time_per_iteration": 2.5810000896453857
    },
    {
      "auxiliary_loss_clip": 0.01179993,
      "auxiliary_loss_mlp": 0.0103233,
      "balance_loss_clip": 1.04834414,
      "balance_loss_mlp": 1.02369928,
      "epoch": 0.6298322611675584,
      "flos": 19497662997120.0,
      "grad_norm": 1.65508316107497,
      "language_loss": 0.75318301,
      "learning_rate": 1.2731177391990125e-06,
      "loss": 0.77530622,
      "num_input_tokens_seen": 112748665,
      "step": 5238,
      "time_per_iteration": 2.6487600803375244
    },
    {
      "auxiliary_loss_clip": 0.0117862,
      "auxiliary_loss_mlp": 0.0102167,
      "balance_loss_clip": 1.04535985,
      "balance_loss_mlp": 1.0142318,
      "epoch": 0.6299525040581976,
      "flos": 12604466649600.0,
      "grad_norm": 35.542462123002174,
      "language_loss": 0.8197512,
      "learning_rate": 1.2723920909923203e-06,
      "loss": 0.84175408,
      "num_input_tokens_seen": 112764410,
      "step": 5239,
      "time_per_iteration": 3.567413330078125
    },
    {
      "auxiliary_loss_clip": 0.01070745,
      "auxiliary_loss_mlp": 0.01001413,
      "balance_loss_clip": 1.01497054,
      "balance_loss_mlp": 1.00068593,
      "epoch": 0.6300727469488366,
      "flos": 57725685636480.0,
      "grad_norm": 0.8608424371972428,
      "language_loss": 0.60456026,
      "learning_rate": 1.2716665531631688e-06,
      "loss": 0.62528181,
      "num_input_tokens_seen": 112818695,
      "step": 5240,
      "time_per_iteration": 3.083372116088867
    },
    {
      "auxiliary_loss_clip": 0.0117924,
      "auxiliary_loss_mlp": 0.01022201,
      "balance_loss_clip": 1.04635215,
      "balance_loss_mlp": 1.01364207,
      "epoch": 0.6301929898394757,
      "flos": 22527302607360.0,
      "grad_norm": 1.7599546202899294,
      "language_loss": 0.77413905,
      "learning_rate": 1.270941125821623e-06,
      "loss": 0.79615343,
      "num_input_tokens_seen": 112839120,
      "step": 5241,
      "time_per_iteration": 2.585318088531494
    },
    {
      "auxiliary_loss_clip": 0.01169691,
      "auxiliary_loss_mlp": 0.01029467,
      "balance_loss_clip": 1.04466701,
      "balance_loss_mlp": 1.02174854,
      "epoch": 0.6303132327301149,
      "flos": 28293550675200.0,
      "grad_norm": 1.5928012514684284,
      "language_loss": 0.75681591,
      "learning_rate": 1.2702158090777278e-06,
      "loss": 0.77880746,
      "num_input_tokens_seen": 112860210,
      "step": 5242,
      "time_per_iteration": 3.5390408039093018
    },
    {
      "auxiliary_loss_clip": 0.01176465,
      "auxiliary_loss_mlp": 0.01025394,
      "balance_loss_clip": 1.04328847,
      "balance_loss_mlp": 1.01793468,
      "epoch": 0.6304334756207539,
      "flos": 25264521596160.0,
      "grad_norm": 7.866017855118037,
      "language_loss": 0.75421858,
      "learning_rate": 1.2694906030415148e-06,
      "loss": 0.77623719,
      "num_input_tokens_seen": 112877955,
      "step": 5243,
      "time_per_iteration": 2.682084321975708
    },
    {
      "auxiliary_loss_clip": 0.01187564,
      "auxiliary_loss_mlp": 0.0102795,
      "balance_loss_clip": 1.04836655,
      "balance_loss_mlp": 1.01982009,
      "epoch": 0.630553718511393,
      "flos": 18033548728320.0,
      "grad_norm": 2.9806581722235834,
      "language_loss": 0.82387191,
      "learning_rate": 1.2687655078229958e-06,
      "loss": 0.84602708,
      "num_input_tokens_seen": 112892285,
      "step": 5244,
      "time_per_iteration": 2.5725553035736084
    },
    {
      "auxiliary_loss_clip": 0.01176523,
      "auxiliary_loss_mlp": 0.0102131,
      "balance_loss_clip": 1.04910159,
      "balance_loss_mlp": 1.01373386,
      "epoch": 0.6306739614020321,
      "flos": 27304103658240.0,
      "grad_norm": 2.607501763772789,
      "language_loss": 0.69273126,
      "learning_rate": 1.2680405235321678e-06,
      "loss": 0.71470964,
      "num_input_tokens_seen": 112913620,
      "step": 5245,
      "time_per_iteration": 3.625922679901123
    },
    {
      "auxiliary_loss_clip": 0.01180661,
      "auxiliary_loss_mlp": 0.01208504,
      "balance_loss_clip": 1.05115068,
      "balance_loss_mlp": 1.0009383,
      "epoch": 0.6307942042926712,
      "flos": 15341434243200.0,
      "grad_norm": 2.006530666922793,
      "language_loss": 0.78699815,
      "learning_rate": 1.267315650279011e-06,
      "loss": 0.81088972,
      "num_input_tokens_seen": 112932090,
      "step": 5246,
      "time_per_iteration": 2.592593193054199
    },
    {
      "auxiliary_loss_clip": 0.0117585,
      "auxiliary_loss_mlp": 0.01021577,
      "balance_loss_clip": 1.04894185,
      "balance_loss_mlp": 1.01415622,
      "epoch": 0.6309144471833102,
      "flos": 19606400444160.0,
      "grad_norm": 1.8629708797511826,
      "language_loss": 0.74308753,
      "learning_rate": 1.2665908881734874e-06,
      "loss": 0.76506186,
      "num_input_tokens_seen": 112950925,
      "step": 5247,
      "time_per_iteration": 2.677671194076538
    },
    {
      "auxiliary_loss_clip": 0.01174779,
      "auxiliary_loss_mlp": 0.0102625,
      "balance_loss_clip": 1.04914391,
      "balance_loss_mlp": 1.01925802,
      "epoch": 0.6310346900739494,
      "flos": 17493345112320.0,
      "grad_norm": 2.846434803358855,
      "language_loss": 0.852458,
      "learning_rate": 1.2658662373255432e-06,
      "loss": 0.87446833,
      "num_input_tokens_seen": 112969315,
      "step": 5248,
      "time_per_iteration": 2.5614590644836426
    },
    {
      "auxiliary_loss_clip": 0.01083332,
      "auxiliary_loss_mlp": 0.01000465,
      "balance_loss_clip": 1.01443148,
      "balance_loss_mlp": 0.99971086,
      "epoch": 0.6311549329645885,
      "flos": 55070164131840.0,
      "grad_norm": 0.8210241049586382,
      "language_loss": 0.52190006,
      "learning_rate": 1.2651416978451063e-06,
      "loss": 0.54273802,
      "num_input_tokens_seen": 113034700,
      "step": 5249,
      "time_per_iteration": 3.266392230987549
    },
    {
      "auxiliary_loss_clip": 0.01175649,
      "auxiliary_loss_mlp": 0.01025457,
      "balance_loss_clip": 1.05039871,
      "balance_loss_mlp": 1.01781011,
      "epoch": 0.6312751758552275,
      "flos": 41902545075840.0,
      "grad_norm": 2.041847368921398,
      "language_loss": 0.65083659,
      "learning_rate": 1.2644172698420903e-06,
      "loss": 0.67284763,
      "num_input_tokens_seen": 113056805,
      "step": 5250,
      "time_per_iteration": 2.8933048248291016
    },
    {
      "auxiliary_loss_clip": 0.01180854,
      "auxiliary_loss_mlp": 0.01029898,
      "balance_loss_clip": 1.04586458,
      "balance_loss_mlp": 1.02186346,
      "epoch": 0.6313954187458667,
      "flos": 19646800266240.0,
      "grad_norm": 1.9986811983137296,
      "language_loss": 0.84537804,
      "learning_rate": 1.2636929534263892e-06,
      "loss": 0.86748552,
      "num_input_tokens_seen": 113075790,
      "step": 5251,
      "time_per_iteration": 3.53194522857666
    },
    {
      "auxiliary_loss_clip": 0.01181349,
      "auxiliary_loss_mlp": 0.01023944,
      "balance_loss_clip": 1.04167175,
      "balance_loss_mlp": 1.01645136,
      "epoch": 0.6315156616365057,
      "flos": 22894273906560.0,
      "grad_norm": 2.0976689087296405,
      "language_loss": 0.77667576,
      "learning_rate": 1.2629687487078821e-06,
      "loss": 0.7987287,
      "num_input_tokens_seen": 113094600,
      "step": 5252,
      "time_per_iteration": 2.6638028621673584
    },
    {
      "auxiliary_loss_clip": 0.0117821,
      "auxiliary_loss_mlp": 0.01028124,
      "balance_loss_clip": 1.04710293,
      "balance_loss_mlp": 1.01958859,
      "epoch": 0.6316359045271448,
      "flos": 23726251699200.0,
      "grad_norm": 2.164303216461026,
      "language_loss": 0.76839232,
      "learning_rate": 1.2622446557964293e-06,
      "loss": 0.7904557,
      "num_input_tokens_seen": 113112605,
      "step": 5253,
      "time_per_iteration": 2.6300458908081055
    },
    {
      "auxiliary_loss_clip": 0.01175172,
      "auxiliary_loss_mlp": 0.0102444,
      "balance_loss_clip": 1.04215574,
      "balance_loss_mlp": 1.01757991,
      "epoch": 0.631756147417784,
      "flos": 33108417164160.0,
      "grad_norm": 1.7039192582163463,
      "language_loss": 0.71557552,
      "learning_rate": 1.261520674801876e-06,
      "loss": 0.73757166,
      "num_input_tokens_seen": 113133200,
      "step": 5254,
      "time_per_iteration": 2.8230795860290527
    },
    {
      "auxiliary_loss_clip": 0.01176199,
      "auxiliary_loss_mlp": 0.01023982,
      "balance_loss_clip": 1.04912794,
      "balance_loss_mlp": 1.01607227,
      "epoch": 0.631876390308423,
      "flos": 31248424126080.0,
      "grad_norm": 2.0973748256978832,
      "language_loss": 0.72077763,
      "learning_rate": 1.2607968058340488e-06,
      "loss": 0.74277943,
      "num_input_tokens_seen": 113152895,
      "step": 5255,
      "time_per_iteration": 2.6987061500549316
    },
    {
      "auxiliary_loss_clip": 0.01171967,
      "auxiliary_loss_mlp": 0.0102793,
      "balance_loss_clip": 1.043872,
      "balance_loss_mlp": 1.02022934,
      "epoch": 0.6319966331990621,
      "flos": 24681152810880.0,
      "grad_norm": 2.107732934801765,
      "language_loss": 0.73222888,
      "learning_rate": 1.2600730490027583e-06,
      "loss": 0.75422782,
      "num_input_tokens_seen": 113173135,
      "step": 5256,
      "time_per_iteration": 2.6315650939941406
    },
    {
      "auxiliary_loss_clip": 0.01179719,
      "auxiliary_loss_mlp": 0.01024605,
      "balance_loss_clip": 1.04502714,
      "balance_loss_mlp": 1.01683259,
      "epoch": 0.6321168760897012,
      "flos": 17491764913920.0,
      "grad_norm": 1.6637834045411688,
      "language_loss": 0.80303776,
      "learning_rate": 1.2593494044177984e-06,
      "loss": 0.82508105,
      "num_input_tokens_seen": 113191440,
      "step": 5257,
      "time_per_iteration": 2.5853941440582275
    },
    {
      "auxiliary_loss_clip": 0.01176478,
      "auxiliary_loss_mlp": 0.01024619,
      "balance_loss_clip": 1.04894936,
      "balance_loss_mlp": 1.0163399,
      "epoch": 0.6322371189803403,
      "flos": 18295373940480.0,
      "grad_norm": 3.015013334148986,
      "language_loss": 0.80794871,
      "learning_rate": 1.2586258721889448e-06,
      "loss": 0.82995963,
      "num_input_tokens_seen": 113208790,
      "step": 5258,
      "time_per_iteration": 2.518662929534912
    },
    {
      "auxiliary_loss_clip": 0.01176712,
      "auxiliary_loss_mlp": 0.01025683,
      "balance_loss_clip": 1.04533553,
      "balance_loss_mlp": 1.01789248,
      "epoch": 0.6323573618709794,
      "flos": 20157270399360.0,
      "grad_norm": 3.0482258978567245,
      "language_loss": 0.82056034,
      "learning_rate": 1.2579024524259573e-06,
      "loss": 0.84258431,
      "num_input_tokens_seen": 113225050,
      "step": 5259,
      "time_per_iteration": 2.658355712890625
    },
    {
      "auxiliary_loss_clip": 0.01173498,
      "auxiliary_loss_mlp": 0.01024148,
      "balance_loss_clip": 1.04442465,
      "balance_loss_mlp": 1.01652431,
      "epoch": 0.6324776047616185,
      "flos": 20042391726720.0,
      "grad_norm": 2.1973197340646573,
      "language_loss": 0.91494489,
      "learning_rate": 1.2571791452385768e-06,
      "loss": 0.93692124,
      "num_input_tokens_seen": 113242315,
      "step": 5260,
      "time_per_iteration": 2.590060234069824
    },
    {
      "auxiliary_loss_clip": 0.01176363,
      "auxiliary_loss_mlp": 0.01029085,
      "balance_loss_clip": 1.04643619,
      "balance_loss_mlp": 1.02146482,
      "epoch": 0.6325978476522576,
      "flos": 30848235724800.0,
      "grad_norm": 1.621772627136363,
      "language_loss": 0.77283835,
      "learning_rate": 1.2564559507365301e-06,
      "loss": 0.79489285,
      "num_input_tokens_seen": 113264720,
      "step": 5261,
      "time_per_iteration": 2.6868410110473633
    },
    {
      "auxiliary_loss_clip": 0.01178627,
      "auxiliary_loss_mlp": 0.01025282,
      "balance_loss_clip": 1.04685998,
      "balance_loss_mlp": 1.01662123,
      "epoch": 0.6327180905428966,
      "flos": 24535104111360.0,
      "grad_norm": 1.986837342199002,
      "language_loss": 0.78733063,
      "learning_rate": 1.2557328690295244e-06,
      "loss": 0.80936968,
      "num_input_tokens_seen": 113282910,
      "step": 5262,
      "time_per_iteration": 2.6439056396484375
    },
    {
      "auxiliary_loss_clip": 0.01186446,
      "auxiliary_loss_mlp": 0.01025797,
      "balance_loss_clip": 1.04831648,
      "balance_loss_mlp": 1.01787591,
      "epoch": 0.6328383334335358,
      "flos": 21575274583680.0,
      "grad_norm": 1.7641632549973953,
      "language_loss": 0.762326,
      "learning_rate": 1.255009900227251e-06,
      "loss": 0.7844485,
      "num_input_tokens_seen": 113301935,
      "step": 5263,
      "time_per_iteration": 2.6459736824035645
    },
    {
      "auxiliary_loss_clip": 0.01169125,
      "auxiliary_loss_mlp": 0.0102443,
      "balance_loss_clip": 1.05026126,
      "balance_loss_mlp": 1.01769733,
      "epoch": 0.6329585763241748,
      "flos": 22929861306240.0,
      "grad_norm": 1.9758786884682285,
      "language_loss": 0.796754,
      "learning_rate": 1.254287044439383e-06,
      "loss": 0.81868953,
      "num_input_tokens_seen": 113321540,
      "step": 5264,
      "time_per_iteration": 2.5331249237060547
    },
    {
      "auxiliary_loss_clip": 0.0107083,
      "auxiliary_loss_mlp": 0.00999801,
      "balance_loss_clip": 1.01497102,
      "balance_loss_mlp": 0.99911845,
      "epoch": 0.6330788192148139,
      "flos": 70936897847040.0,
      "grad_norm": 0.7788642476667336,
      "language_loss": 0.54454732,
      "learning_rate": 1.2535643017755776e-06,
      "loss": 0.56525362,
      "num_input_tokens_seen": 113383730,
      "step": 5265,
      "time_per_iteration": 4.138848781585693
    },
    {
      "auxiliary_loss_clip": 0.01177717,
      "auxiliary_loss_mlp": 0.01027598,
      "balance_loss_clip": 1.04281318,
      "balance_loss_mlp": 1.01947379,
      "epoch": 0.6331990621054531,
      "flos": 21244501215360.0,
      "grad_norm": 2.3107795924468126,
      "language_loss": 0.72350252,
      "learning_rate": 1.2528416723454737e-06,
      "loss": 0.74555564,
      "num_input_tokens_seen": 113400400,
      "step": 5266,
      "time_per_iteration": 2.629540205001831
    },
    {
      "auxiliary_loss_clip": 0.01170824,
      "auxiliary_loss_mlp": 0.0102282,
      "balance_loss_clip": 1.05182302,
      "balance_loss_mlp": 1.015733,
      "epoch": 0.6333193049960921,
      "flos": 34459412526720.0,
      "grad_norm": 1.6153744792695204,
      "language_loss": 0.70949972,
      "learning_rate": 1.2521191562586945e-06,
      "loss": 0.73143619,
      "num_input_tokens_seen": 113424050,
      "step": 5267,
      "time_per_iteration": 2.68668794631958
    },
    {
      "auxiliary_loss_clip": 0.01173654,
      "auxiliary_loss_mlp": 0.01207691,
      "balance_loss_clip": 1.0509032,
      "balance_loss_mlp": 1.00080848,
      "epoch": 0.6334395478867312,
      "flos": 18329883932160.0,
      "grad_norm": 1.934020478484935,
      "language_loss": 0.7713517,
      "learning_rate": 1.2513967536248445e-06,
      "loss": 0.79516512,
      "num_input_tokens_seen": 113440370,
      "step": 5268,
      "time_per_iteration": 2.5977907180786133
    },
    {
      "auxiliary_loss_clip": 0.01169729,
      "auxiliary_loss_mlp": 0.01027052,
      "balance_loss_clip": 1.04912698,
      "balance_loss_mlp": 1.0200578,
      "epoch": 0.6335597907773702,
      "flos": 23623152687360.0,
      "grad_norm": 2.6616375347083916,
      "language_loss": 0.81251633,
      "learning_rate": 1.2506744645535117e-06,
      "loss": 0.83448416,
      "num_input_tokens_seen": 113460800,
      "step": 5269,
      "time_per_iteration": 3.500225782394409
    },
    {
      "auxiliary_loss_clip": 0.01169597,
      "auxiliary_loss_mlp": 0.01019848,
      "balance_loss_clip": 1.04171121,
      "balance_loss_mlp": 1.01229358,
      "epoch": 0.6336800336680094,
      "flos": 22710913954560.0,
      "grad_norm": 1.9487307196374584,
      "language_loss": 0.60874629,
      "learning_rate": 1.249952289154267e-06,
      "loss": 0.63064075,
      "num_input_tokens_seen": 113480840,
      "step": 5270,
      "time_per_iteration": 2.7100164890289307
    },
    {
      "auxiliary_loss_clip": 0.01171962,
      "auxiliary_loss_mlp": 0.01026945,
      "balance_loss_clip": 1.0401001,
      "balance_loss_mlp": 1.01995325,
      "epoch": 0.6338002765586485,
      "flos": 23622757637760.0,
      "grad_norm": 5.545143491114886,
      "language_loss": 0.76653802,
      "learning_rate": 1.2492302275366635e-06,
      "loss": 0.78852713,
      "num_input_tokens_seen": 113500515,
      "step": 5271,
      "time_per_iteration": 3.575730323791504
    },
    {
      "auxiliary_loss_clip": 0.01169802,
      "auxiliary_loss_mlp": 0.01026155,
      "balance_loss_clip": 1.04720342,
      "balance_loss_mlp": 1.01837111,
      "epoch": 0.6339205194492875,
      "flos": 26505450708480.0,
      "grad_norm": 2.7964533343788993,
      "language_loss": 0.65784991,
      "learning_rate": 1.2485082798102377e-06,
      "loss": 0.67980951,
      "num_input_tokens_seen": 113520930,
      "step": 5272,
      "time_per_iteration": 2.6535696983337402
    },
    {
      "auxiliary_loss_clip": 0.01185386,
      "auxiliary_loss_mlp": 0.01023885,
      "balance_loss_clip": 1.04519689,
      "balance_loss_mlp": 1.01532006,
      "epoch": 0.6340407623399267,
      "flos": 18544306170240.0,
      "grad_norm": 2.2232467284937982,
      "language_loss": 0.6898998,
      "learning_rate": 1.2477864460845084e-06,
      "loss": 0.71199256,
      "num_input_tokens_seen": 113537330,
      "step": 5273,
      "time_per_iteration": 2.721341848373413
    },
    {
      "auxiliary_loss_clip": 0.0117746,
      "auxiliary_loss_mlp": 0.01023212,
      "balance_loss_clip": 1.04557896,
      "balance_loss_mlp": 1.01516497,
      "epoch": 0.6341610052305657,
      "flos": 17712579772800.0,
      "grad_norm": 6.040733618796398,
      "language_loss": 0.73732525,
      "learning_rate": 1.2470647264689776e-06,
      "loss": 0.75933194,
      "num_input_tokens_seen": 113555810,
      "step": 5274,
      "time_per_iteration": 2.620189905166626
    },
    {
      "auxiliary_loss_clip": 0.01194195,
      "auxiliary_loss_mlp": 0.01026104,
      "balance_loss_clip": 1.04226172,
      "balance_loss_mlp": 1.01817107,
      "epoch": 0.6342812481212048,
      "flos": 23587026583680.0,
      "grad_norm": 1.9486220877561617,
      "language_loss": 0.71459264,
      "learning_rate": 1.2463431210731282e-06,
      "loss": 0.73679566,
      "num_input_tokens_seen": 113575395,
      "step": 5275,
      "time_per_iteration": 2.744953155517578
    },
    {
      "auxiliary_loss_clip": 0.01192594,
      "auxiliary_loss_mlp": 0.01024741,
      "balance_loss_clip": 1.04295814,
      "balance_loss_mlp": 1.01743984,
      "epoch": 0.634401491011844,
      "flos": 17821927751040.0,
      "grad_norm": 2.355114476369188,
      "language_loss": 0.7651388,
      "learning_rate": 1.2456216300064289e-06,
      "loss": 0.78731215,
      "num_input_tokens_seen": 113592945,
      "step": 5276,
      "time_per_iteration": 2.6746087074279785
    },
    {
      "auxiliary_loss_clip": 0.01171935,
      "auxiliary_loss_mlp": 0.01025244,
      "balance_loss_clip": 1.04489148,
      "balance_loss_mlp": 1.01695919,
      "epoch": 0.634521733902483,
      "flos": 21358158825600.0,
      "grad_norm": 1.6062825748366025,
      "language_loss": 0.78383863,
      "learning_rate": 1.244900253378328e-06,
      "loss": 0.80581039,
      "num_input_tokens_seen": 113613000,
      "step": 5277,
      "time_per_iteration": 3.693627119064331
    },
    {
      "auxiliary_loss_clip": 0.01192804,
      "auxiliary_loss_mlp": 0.01026369,
      "balance_loss_clip": 1.04255247,
      "balance_loss_mlp": 1.01884699,
      "epoch": 0.6346419767931221,
      "flos": 16545052103040.0,
      "grad_norm": 2.2162884509706076,
      "language_loss": 0.6945616,
      "learning_rate": 1.2441789912982583e-06,
      "loss": 0.71675336,
      "num_input_tokens_seen": 113630085,
      "step": 5278,
      "time_per_iteration": 2.6446373462677
    },
    {
      "auxiliary_loss_clip": 0.01179342,
      "auxiliary_loss_mlp": 0.01024874,
      "balance_loss_clip": 1.04958153,
      "balance_loss_mlp": 1.0165472,
      "epoch": 0.6347622196837612,
      "flos": 24350989973760.0,
      "grad_norm": 2.5303877193372175,
      "language_loss": 0.64783525,
      "learning_rate": 1.2434578438756346e-06,
      "loss": 0.66987753,
      "num_input_tokens_seen": 113650515,
      "step": 5279,
      "time_per_iteration": 2.6048550605773926
    },
    {
      "auxiliary_loss_clip": 0.01173651,
      "auxiliary_loss_mlp": 0.01021655,
      "balance_loss_clip": 1.04626441,
      "balance_loss_mlp": 1.01445782,
      "epoch": 0.6348824625744003,
      "flos": 64523178195840.0,
      "grad_norm": 2.7036091930961073,
      "language_loss": 0.78056973,
      "learning_rate": 1.242736811219855e-06,
      "loss": 0.80252284,
      "num_input_tokens_seen": 113676475,
      "step": 5280,
      "time_per_iteration": 3.016019344329834
    },
    {
      "auxiliary_loss_clip": 0.01169474,
      "auxiliary_loss_mlp": 0.01023322,
      "balance_loss_clip": 1.04695261,
      "balance_loss_mlp": 1.01566887,
      "epoch": 0.6350027054650393,
      "flos": 28622133313920.0,
      "grad_norm": 2.1115093813264427,
      "language_loss": 0.82285249,
      "learning_rate": 1.2420158934402988e-06,
      "loss": 0.8447805,
      "num_input_tokens_seen": 113697090,
      "step": 5281,
      "time_per_iteration": 2.6216421127319336
    },
    {
      "auxiliary_loss_clip": 0.01165073,
      "auxiliary_loss_mlp": 0.01023121,
      "balance_loss_clip": 1.03966582,
      "balance_loss_mlp": 1.01525354,
      "epoch": 0.6351229483556785,
      "flos": 23002544476800.0,
      "grad_norm": 2.5278633781385405,
      "language_loss": 0.85192406,
      "learning_rate": 1.2412950906463286e-06,
      "loss": 0.873806,
      "num_input_tokens_seen": 113714395,
      "step": 5282,
      "time_per_iteration": 2.663144826889038
    },
    {
      "auxiliary_loss_clip": 0.01182231,
      "auxiliary_loss_mlp": 0.01026498,
      "balance_loss_clip": 1.04395175,
      "balance_loss_mlp": 1.01908016,
      "epoch": 0.6352431912463176,
      "flos": 21939300967680.0,
      "grad_norm": 1.7485478152478198,
      "language_loss": 0.89961344,
      "learning_rate": 1.2405744029472902e-06,
      "loss": 0.92170072,
      "num_input_tokens_seen": 113733880,
      "step": 5283,
      "time_per_iteration": 2.7185561656951904
    },
    {
      "auxiliary_loss_clip": 0.01174692,
      "auxiliary_loss_mlp": 0.01024153,
      "balance_loss_clip": 1.04468548,
      "balance_loss_mlp": 1.01667821,
      "epoch": 0.6353634341369566,
      "flos": 13735257684480.0,
      "grad_norm": 1.9533324609852443,
      "language_loss": 0.76718497,
      "learning_rate": 1.2398538304525108e-06,
      "loss": 0.78917336,
      "num_input_tokens_seen": 113752505,
      "step": 5284,
      "time_per_iteration": 2.628086805343628
    },
    {
      "auxiliary_loss_clip": 0.01176069,
      "auxiliary_loss_mlp": 0.01029277,
      "balance_loss_clip": 1.04560256,
      "balance_loss_mlp": 1.02089071,
      "epoch": 0.6354836770275958,
      "flos": 19316170552320.0,
      "grad_norm": 2.519635377579637,
      "language_loss": 0.7593295,
      "learning_rate": 1.2391333732713016e-06,
      "loss": 0.78138304,
      "num_input_tokens_seen": 113770310,
      "step": 5285,
      "time_per_iteration": 2.6290671825408936
    },
    {
      "auxiliary_loss_clip": 0.01178667,
      "auxiliary_loss_mlp": 0.01031987,
      "balance_loss_clip": 1.04362035,
      "balance_loss_mlp": 1.02272415,
      "epoch": 0.6356039199182348,
      "flos": 21613375935360.0,
      "grad_norm": 2.4101843280652,
      "language_loss": 0.78638184,
      "learning_rate": 1.2384130315129543e-06,
      "loss": 0.80848837,
      "num_input_tokens_seen": 113788635,
      "step": 5286,
      "time_per_iteration": 2.673293113708496
    },
    {
      "auxiliary_loss_clip": 0.01197003,
      "auxiliary_loss_mlp": 0.01024435,
      "balance_loss_clip": 1.03606534,
      "balance_loss_mlp": 1.01632273,
      "epoch": 0.6357241628088739,
      "flos": 18111978074880.0,
      "grad_norm": 2.431238312284621,
      "language_loss": 0.73587382,
      "learning_rate": 1.2376928052867447e-06,
      "loss": 0.75808823,
      "num_input_tokens_seen": 113807755,
      "step": 5287,
      "time_per_iteration": 2.7590887546539307
    },
    {
      "auxiliary_loss_clip": 0.01179932,
      "auxiliary_loss_mlp": 0.01023723,
      "balance_loss_clip": 1.04960155,
      "balance_loss_mlp": 1.01640952,
      "epoch": 0.6358444056995131,
      "flos": 24935256599040.0,
      "grad_norm": 2.3501557787931993,
      "language_loss": 0.77902156,
      "learning_rate": 1.2369726947019299e-06,
      "loss": 0.80105817,
      "num_input_tokens_seen": 113828230,
      "step": 5288,
      "time_per_iteration": 2.6534130573272705
    },
    {
      "auxiliary_loss_clip": 0.0117175,
      "auxiliary_loss_mlp": 0.01021357,
      "balance_loss_clip": 1.0457052,
      "balance_loss_mlp": 1.0134716,
      "epoch": 0.6359646485901521,
      "flos": 23293348986240.0,
      "grad_norm": 2.125205382769455,
      "language_loss": 0.66847432,
      "learning_rate": 1.2362526998677511e-06,
      "loss": 0.69040537,
      "num_input_tokens_seen": 113844595,
      "step": 5289,
      "time_per_iteration": 2.553321123123169
    },
    {
      "auxiliary_loss_clip": 0.01179369,
      "auxiliary_loss_mlp": 0.01025399,
      "balance_loss_clip": 1.04657435,
      "balance_loss_mlp": 1.01846671,
      "epoch": 0.6360848914807912,
      "flos": 20887442069760.0,
      "grad_norm": 2.0543837346512226,
      "language_loss": 0.84269083,
      "learning_rate": 1.2355328208934301e-06,
      "loss": 0.86473852,
      "num_input_tokens_seen": 113863470,
      "step": 5290,
      "time_per_iteration": 2.607434034347534
    },
    {
      "auxiliary_loss_clip": 0.01172309,
      "auxiliary_loss_mlp": 0.01207788,
      "balance_loss_clip": 1.04449773,
      "balance_loss_mlp": 1.000911,
      "epoch": 0.6362051343714303,
      "flos": 18479775386880.0,
      "grad_norm": 1.5624482978377894,
      "language_loss": 0.726726,
      "learning_rate": 1.2348130578881728e-06,
      "loss": 0.75052691,
      "num_input_tokens_seen": 113881690,
      "step": 5291,
      "time_per_iteration": 3.5485286712646484
    },
    {
      "auxiliary_loss_clip": 0.01174276,
      "auxiliary_loss_mlp": 0.01027536,
      "balance_loss_clip": 1.05021858,
      "balance_loss_mlp": 1.01959085,
      "epoch": 0.6363253772620694,
      "flos": 24389594115840.0,
      "grad_norm": 3.5985750927497135,
      "language_loss": 0.76771581,
      "learning_rate": 1.2340934109611664e-06,
      "loss": 0.78973395,
      "num_input_tokens_seen": 113902450,
      "step": 5292,
      "time_per_iteration": 2.525347948074341
    },
    {
      "auxiliary_loss_clip": 0.01179329,
      "auxiliary_loss_mlp": 0.01027856,
      "balance_loss_clip": 1.04573202,
      "balance_loss_mlp": 1.01930237,
      "epoch": 0.6364456201527084,
      "flos": 25958243940480.0,
      "grad_norm": 3.336143660779107,
      "language_loss": 0.69272494,
      "learning_rate": 1.2333738802215798e-06,
      "loss": 0.71479678,
      "num_input_tokens_seen": 113922670,
      "step": 5293,
      "time_per_iteration": 2.6796975135803223
    },
    {
      "auxiliary_loss_clip": 0.01176524,
      "auxiliary_loss_mlp": 0.01025551,
      "balance_loss_clip": 1.04098356,
      "balance_loss_mlp": 1.01786494,
      "epoch": 0.6365658630433476,
      "flos": 20740711011840.0,
      "grad_norm": 1.8199319010462474,
      "language_loss": 0.81449771,
      "learning_rate": 1.2326544657785668e-06,
      "loss": 0.83651841,
      "num_input_tokens_seen": 113942360,
      "step": 5294,
      "time_per_iteration": 2.649815797805786
    },
    {
      "auxiliary_loss_clip": 0.0117081,
      "auxiliary_loss_mlp": 0.01030547,
      "balance_loss_clip": 1.0430522,
      "balance_loss_mlp": 1.02214861,
      "epoch": 0.6366861059339867,
      "flos": 21434146047360.0,
      "grad_norm": 2.7376107325470045,
      "language_loss": 0.74747396,
      "learning_rate": 1.2319351677412608e-06,
      "loss": 0.7694875,
      "num_input_tokens_seen": 113959405,
      "step": 5295,
      "time_per_iteration": 3.540152072906494
    },
    {
      "auxiliary_loss_clip": 0.01189584,
      "auxiliary_loss_mlp": 0.01022574,
      "balance_loss_clip": 1.04886031,
      "balance_loss_mlp": 1.01462913,
      "epoch": 0.6368063488246257,
      "flos": 22267093507200.0,
      "grad_norm": 1.998237245200984,
      "language_loss": 0.74341416,
      "learning_rate": 1.2312159862187796e-06,
      "loss": 0.76553571,
      "num_input_tokens_seen": 113977815,
      "step": 5296,
      "time_per_iteration": 2.715571403503418
    },
    {
      "auxiliary_loss_clip": 0.01179229,
      "auxiliary_loss_mlp": 0.01032587,
      "balance_loss_clip": 1.05347633,
      "balance_loss_mlp": 1.02483582,
      "epoch": 0.6369265917152649,
      "flos": 22420719976320.0,
      "grad_norm": 2.5535769882033086,
      "language_loss": 0.76192075,
      "learning_rate": 1.2304969213202217e-06,
      "loss": 0.7840389,
      "num_input_tokens_seen": 113999075,
      "step": 5297,
      "time_per_iteration": 2.605937957763672
    },
    {
      "auxiliary_loss_clip": 0.01171882,
      "auxiliary_loss_mlp": 0.01022017,
      "balance_loss_clip": 1.0446198,
      "balance_loss_mlp": 1.01492965,
      "epoch": 0.6370468346059039,
      "flos": 24718176754560.0,
      "grad_norm": 2.54439954386557,
      "language_loss": 0.79141033,
      "learning_rate": 1.2297779731546692e-06,
      "loss": 0.81334931,
      "num_input_tokens_seen": 114018170,
      "step": 5298,
      "time_per_iteration": 3.6598031520843506
    },
    {
      "auxiliary_loss_clip": 0.01176283,
      "auxiliary_loss_mlp": 0.01026423,
      "balance_loss_clip": 1.04826736,
      "balance_loss_mlp": 1.01841807,
      "epoch": 0.637167077496543,
      "flos": 25296589463040.0,
      "grad_norm": 3.8350564399108533,
      "language_loss": 0.78293705,
      "learning_rate": 1.2290591418311853e-06,
      "loss": 0.80496413,
      "num_input_tokens_seen": 114035565,
      "step": 5299,
      "time_per_iteration": 2.6103808879852295
    },
    {
      "auxiliary_loss_clip": 0.01172716,
      "auxiliary_loss_mlp": 0.01026062,
      "balance_loss_clip": 1.0492667,
      "balance_loss_mlp": 1.01835501,
      "epoch": 0.637287320387182,
      "flos": 27671110871040.0,
      "grad_norm": 1.5707883464076169,
      "language_loss": 0.72184402,
      "learning_rate": 1.2283404274588172e-06,
      "loss": 0.74383181,
      "num_input_tokens_seen": 114054510,
      "step": 5300,
      "time_per_iteration": 2.6683406829833984
    },
    {
      "auxiliary_loss_clip": 0.01094093,
      "auxiliary_loss_mlp": 0.01002459,
      "balance_loss_clip": 1.01168787,
      "balance_loss_mlp": 1.00171661,
      "epoch": 0.6374075632778212,
      "flos": 63173406873600.0,
      "grad_norm": 0.758606209017652,
      "language_loss": 0.52764487,
      "learning_rate": 1.227621830146592e-06,
      "loss": 0.54861033,
      "num_input_tokens_seen": 114109875,
      "step": 5301,
      "time_per_iteration": 3.105102777481079
    },
    {
      "auxiliary_loss_clip": 0.01182412,
      "auxiliary_loss_mlp": 0.01028665,
      "balance_loss_clip": 1.04645133,
      "balance_loss_mlp": 1.02102697,
      "epoch": 0.6375278061684603,
      "flos": 25558127366400.0,
      "grad_norm": 2.1417788711074444,
      "language_loss": 0.79311478,
      "learning_rate": 1.2269033500035217e-06,
      "loss": 0.81522554,
      "num_input_tokens_seen": 114130010,
      "step": 5302,
      "time_per_iteration": 2.71840763092041
    },
    {
      "auxiliary_loss_clip": 0.01178993,
      "auxiliary_loss_mlp": 0.0102763,
      "balance_loss_clip": 1.04596055,
      "balance_loss_mlp": 1.02018857,
      "epoch": 0.6376480490590993,
      "flos": 25666362023040.0,
      "grad_norm": 2.3239375457015337,
      "language_loss": 0.73663342,
      "learning_rate": 1.2261849871385988e-06,
      "loss": 0.75869966,
      "num_input_tokens_seen": 114151115,
      "step": 5303,
      "time_per_iteration": 3.6233205795288086
    },
    {
      "auxiliary_loss_clip": 0.01173094,
      "auxiliary_loss_mlp": 0.01024583,
      "balance_loss_clip": 1.04880333,
      "balance_loss_mlp": 1.01651263,
      "epoch": 0.6377682919497385,
      "flos": 31537684350720.0,
      "grad_norm": 2.3272421696630476,
      "language_loss": 0.6251297,
      "learning_rate": 1.2254667416607972e-06,
      "loss": 0.64710647,
      "num_input_tokens_seen": 114172715,
      "step": 5304,
      "time_per_iteration": 2.6347076892852783
    },
    {
      "auxiliary_loss_clip": 0.01172746,
      "auxiliary_loss_mlp": 0.01022457,
      "balance_loss_clip": 1.04888535,
      "balance_loss_mlp": 1.0145179,
      "epoch": 0.6378885348403776,
      "flos": 23039209284480.0,
      "grad_norm": 1.7105942485817451,
      "language_loss": 0.83304596,
      "learning_rate": 1.2247486136790756e-06,
      "loss": 0.85499799,
      "num_input_tokens_seen": 114192195,
      "step": 5305,
      "time_per_iteration": 2.654762029647827
    },
    {
      "auxiliary_loss_clip": 0.0117905,
      "auxiliary_loss_mlp": 0.01029412,
      "balance_loss_clip": 1.05146265,
      "balance_loss_mlp": 1.02174628,
      "epoch": 0.6380087777310166,
      "flos": 18697070712960.0,
      "grad_norm": 2.0295222322969515,
      "language_loss": 0.80555689,
      "learning_rate": 1.2240306033023726e-06,
      "loss": 0.82764149,
      "num_input_tokens_seen": 114210020,
      "step": 5306,
      "time_per_iteration": 2.6016035079956055
    },
    {
      "auxiliary_loss_clip": 0.01181486,
      "auxiliary_loss_mlp": 0.01025399,
      "balance_loss_clip": 1.04238534,
      "balance_loss_mlp": 1.01759636,
      "epoch": 0.6381290206216558,
      "flos": 23331558078720.0,
      "grad_norm": 1.932483341050328,
      "language_loss": 0.72127938,
      "learning_rate": 1.223312710639611e-06,
      "loss": 0.74334824,
      "num_input_tokens_seen": 114228740,
      "step": 5307,
      "time_per_iteration": 2.754700183868408
    },
    {
      "auxiliary_loss_clip": 0.01177789,
      "auxiliary_loss_mlp": 0.01024434,
      "balance_loss_clip": 1.04735589,
      "balance_loss_mlp": 1.01650095,
      "epoch": 0.6382492635122948,
      "flos": 18880466578560.0,
      "grad_norm": 3.621930803827962,
      "language_loss": 0.87207597,
      "learning_rate": 1.2225949357996928e-06,
      "loss": 0.89409816,
      "num_input_tokens_seen": 114246865,
      "step": 5308,
      "time_per_iteration": 2.5908987522125244
    },
    {
      "auxiliary_loss_clip": 0.01169453,
      "auxiliary_loss_mlp": 0.01025154,
      "balance_loss_clip": 1.04903305,
      "balance_loss_mlp": 1.01791167,
      "epoch": 0.6383695064029339,
      "flos": 27819134818560.0,
      "grad_norm": 1.6361341989118596,
      "language_loss": 0.80380154,
      "learning_rate": 1.221877278891505e-06,
      "loss": 0.82574755,
      "num_input_tokens_seen": 114266120,
      "step": 5309,
      "time_per_iteration": 2.6817197799682617
    },
    {
      "auxiliary_loss_clip": 0.01182812,
      "auxiliary_loss_mlp": 0.01030583,
      "balance_loss_clip": 1.04988587,
      "balance_loss_mlp": 1.02220893,
      "epoch": 0.638489749293573,
      "flos": 26395635853440.0,
      "grad_norm": 1.9835408186986856,
      "language_loss": 0.71354187,
      "learning_rate": 1.221159740023915e-06,
      "loss": 0.73567581,
      "num_input_tokens_seen": 114285950,
      "step": 5310,
      "time_per_iteration": 2.6105496883392334
    },
    {
      "auxiliary_loss_clip": 0.01190088,
      "auxiliary_loss_mlp": 0.01208044,
      "balance_loss_clip": 1.04734182,
      "balance_loss_mlp": 1.00083804,
      "epoch": 0.6386099921842121,
      "flos": 23988328306560.0,
      "grad_norm": 2.2807026677533613,
      "language_loss": 0.72367626,
      "learning_rate": 1.2204423193057735e-06,
      "loss": 0.7476576,
      "num_input_tokens_seen": 114304780,
      "step": 5311,
      "time_per_iteration": 2.715398073196411
    },
    {
      "auxiliary_loss_clip": 0.01085739,
      "auxiliary_loss_mlp": 0.01001745,
      "balance_loss_clip": 1.01626074,
      "balance_loss_mlp": 1.00100613,
      "epoch": 0.6387302350748512,
      "flos": 71731169337600.0,
      "grad_norm": 0.8545299657863689,
      "language_loss": 0.6338042,
      "learning_rate": 1.2197250168459122e-06,
      "loss": 0.65467906,
      "num_input_tokens_seen": 114361180,
      "step": 5312,
      "time_per_iteration": 3.219913959503174
    },
    {
      "auxiliary_loss_clip": 0.01177751,
      "auxiliary_loss_mlp": 0.0102335,
      "balance_loss_clip": 1.04967701,
      "balance_loss_mlp": 1.01569724,
      "epoch": 0.6388504779654903,
      "flos": 14535778141440.0,
      "grad_norm": 2.138094932605655,
      "language_loss": 0.74321109,
      "learning_rate": 1.2190078327531454e-06,
      "loss": 0.76522207,
      "num_input_tokens_seen": 114377425,
      "step": 5313,
      "time_per_iteration": 2.548762559890747
    },
    {
      "auxiliary_loss_clip": 0.01176091,
      "auxiliary_loss_mlp": 0.01028758,
      "balance_loss_clip": 1.04723763,
      "balance_loss_mlp": 1.02125931,
      "epoch": 0.6389707208561294,
      "flos": 22346133384960.0,
      "grad_norm": 1.5758530527657846,
      "language_loss": 0.72765338,
      "learning_rate": 1.2182907671362697e-06,
      "loss": 0.74970186,
      "num_input_tokens_seen": 114398120,
      "step": 5314,
      "time_per_iteration": 2.6338603496551514
    },
    {
      "auxiliary_loss_clip": 0.01175431,
      "auxiliary_loss_mlp": 0.01022279,
      "balance_loss_clip": 1.04946709,
      "balance_loss_mlp": 1.01429772,
      "epoch": 0.6390909637467684,
      "flos": 19426883247360.0,
      "grad_norm": 2.2509074140189025,
      "language_loss": 0.7852447,
      "learning_rate": 1.2175738201040626e-06,
      "loss": 0.80722189,
      "num_input_tokens_seen": 114415160,
      "step": 5315,
      "time_per_iteration": 2.560210704803467
    },
    {
      "auxiliary_loss_clip": 0.01174734,
      "auxiliary_loss_mlp": 0.01028523,
      "balance_loss_clip": 1.04781055,
      "balance_loss_mlp": 1.02057171,
      "epoch": 0.6392112066374076,
      "flos": 24090852700800.0,
      "grad_norm": 1.7099655772628106,
      "language_loss": 0.78605205,
      "learning_rate": 1.2168569917652855e-06,
      "loss": 0.80808473,
      "num_input_tokens_seen": 114435015,
      "step": 5316,
      "time_per_iteration": 2.5931203365325928
    },
    {
      "auxiliary_loss_clip": 0.01178487,
      "auxiliary_loss_mlp": 0.01021056,
      "balance_loss_clip": 1.05092311,
      "balance_loss_mlp": 1.01292562,
      "epoch": 0.6393314495280467,
      "flos": 26795141896320.0,
      "grad_norm": 1.5485840969928932,
      "language_loss": 0.63758808,
      "learning_rate": 1.2161402822286797e-06,
      "loss": 0.65958351,
      "num_input_tokens_seen": 114455700,
      "step": 5317,
      "time_per_iteration": 2.5846145153045654
    },
    {
      "auxiliary_loss_clip": 0.01176517,
      "auxiliary_loss_mlp": 0.01021448,
      "balance_loss_clip": 1.04399443,
      "balance_loss_mlp": 1.01379204,
      "epoch": 0.6394516924186857,
      "flos": 20260692633600.0,
      "grad_norm": 1.9871902419347809,
      "language_loss": 0.78939223,
      "learning_rate": 1.2154236916029703e-06,
      "loss": 0.81137192,
      "num_input_tokens_seen": 114473675,
      "step": 5318,
      "time_per_iteration": 3.6078362464904785
    },
    {
      "auxiliary_loss_clip": 0.01183918,
      "auxiliary_loss_mlp": 0.01026422,
      "balance_loss_clip": 1.04148173,
      "balance_loss_mlp": 1.01853657,
      "epoch": 0.6395719353093249,
      "flos": 18368847210240.0,
      "grad_norm": 2.698530462499152,
      "language_loss": 0.74001813,
      "learning_rate": 1.2147072199968627e-06,
      "loss": 0.7621215,
      "num_input_tokens_seen": 114492310,
      "step": 5319,
      "time_per_iteration": 2.6095049381256104
    },
    {
      "auxiliary_loss_clip": 0.01173522,
      "auxiliary_loss_mlp": 0.01026276,
      "balance_loss_clip": 1.04823518,
      "balance_loss_mlp": 1.01896238,
      "epoch": 0.6396921781999639,
      "flos": 17566315591680.0,
      "grad_norm": 1.8490657724511028,
      "language_loss": 0.71951604,
      "learning_rate": 1.2139908675190454e-06,
      "loss": 0.74151397,
      "num_input_tokens_seen": 114511520,
      "step": 5320,
      "time_per_iteration": 2.6317379474639893
    },
    {
      "auxiliary_loss_clip": 0.01176512,
      "auxiliary_loss_mlp": 0.01024018,
      "balance_loss_clip": 1.03858745,
      "balance_loss_mlp": 1.01638258,
      "epoch": 0.639812421090603,
      "flos": 21251252972160.0,
      "grad_norm": 2.116288240146158,
      "language_loss": 0.75356811,
      "learning_rate": 1.2132746342781883e-06,
      "loss": 0.77557343,
      "num_input_tokens_seen": 114532680,
      "step": 5321,
      "time_per_iteration": 2.708294630050659
    },
    {
      "auxiliary_loss_clip": 0.01174812,
      "auxiliary_loss_mlp": 0.01027265,
      "balance_loss_clip": 1.0508225,
      "balance_loss_mlp": 1.01872408,
      "epoch": 0.6399326639812422,
      "flos": 11180967684480.0,
      "grad_norm": 4.608693778277182,
      "language_loss": 0.79940403,
      "learning_rate": 1.2125585203829442e-06,
      "loss": 0.82142484,
      "num_input_tokens_seen": 114548320,
      "step": 5322,
      "time_per_iteration": 3.4954962730407715
    },
    {
      "auxiliary_loss_clip": 0.01169689,
      "auxiliary_loss_mlp": 0.01030008,
      "balance_loss_clip": 1.04559588,
      "balance_loss_mlp": 1.02171087,
      "epoch": 0.6400529068718812,
      "flos": 23911048195200.0,
      "grad_norm": 1.8469577061637588,
      "language_loss": 0.7416482,
      "learning_rate": 1.211842525941946e-06,
      "loss": 0.76364517,
      "num_input_tokens_seen": 114568115,
      "step": 5323,
      "time_per_iteration": 2.7703216075897217
    },
    {
      "auxiliary_loss_clip": 0.01178509,
      "auxiliary_loss_mlp": 0.01024918,
      "balance_loss_clip": 1.04503906,
      "balance_loss_mlp": 1.01716959,
      "epoch": 0.6401731497625203,
      "flos": 44018724890880.0,
      "grad_norm": 2.246176185163626,
      "language_loss": 0.79197901,
      "learning_rate": 1.2111266510638105e-06,
      "loss": 0.8140133,
      "num_input_tokens_seen": 114591040,
      "step": 5324,
      "time_per_iteration": 2.8674817085266113
    },
    {
      "auxiliary_loss_clip": 0.01177484,
      "auxiliary_loss_mlp": 0.01027169,
      "balance_loss_clip": 1.04166341,
      "balance_loss_mlp": 1.01925981,
      "epoch": 0.6402933926531594,
      "flos": 20662209838080.0,
      "grad_norm": 1.725958543046353,
      "language_loss": 0.80004358,
      "learning_rate": 1.2104108958571346e-06,
      "loss": 0.82209015,
      "num_input_tokens_seen": 114609310,
      "step": 5325,
      "time_per_iteration": 3.5261290073394775
    },
    {
      "auxiliary_loss_clip": 0.01172915,
      "auxiliary_loss_mlp": 0.01028255,
      "balance_loss_clip": 1.04833055,
      "balance_loss_mlp": 1.02042341,
      "epoch": 0.6404136355437985,
      "flos": 24863327614080.0,
      "grad_norm": 1.4626766352956653,
      "language_loss": 0.7575177,
      "learning_rate": 1.2096952604304975e-06,
      "loss": 0.77952939,
      "num_input_tokens_seen": 114629740,
      "step": 5326,
      "time_per_iteration": 2.631500005722046
    },
    {
      "auxiliary_loss_clip": 0.01177214,
      "auxiliary_loss_mlp": 0.01024948,
      "balance_loss_clip": 1.04761243,
      "balance_loss_mlp": 1.01690102,
      "epoch": 0.6405338784344375,
      "flos": 40479548901120.0,
      "grad_norm": 2.0523634375013198,
      "language_loss": 0.70447588,
      "learning_rate": 1.2089797448924616e-06,
      "loss": 0.72649747,
      "num_input_tokens_seen": 114653615,
      "step": 5327,
      "time_per_iteration": 2.7305221557617188
    },
    {
      "auxiliary_loss_clip": 0.01188983,
      "auxiliary_loss_mlp": 0.01028335,
      "balance_loss_clip": 1.04189086,
      "balance_loss_mlp": 1.02022254,
      "epoch": 0.6406541213250767,
      "flos": 20886041439360.0,
      "grad_norm": 2.342614662204641,
      "language_loss": 0.65898979,
      "learning_rate": 1.2082643493515692e-06,
      "loss": 0.68116295,
      "num_input_tokens_seen": 114671935,
      "step": 5328,
      "time_per_iteration": 2.6892168521881104
    },
    {
      "auxiliary_loss_clip": 0.01173368,
      "auxiliary_loss_mlp": 0.01024202,
      "balance_loss_clip": 1.04754078,
      "balance_loss_mlp": 1.01618576,
      "epoch": 0.6407743642157158,
      "flos": 23295970679040.0,
      "grad_norm": 1.851051570729426,
      "language_loss": 0.81614983,
      "learning_rate": 1.207549073916346e-06,
      "loss": 0.83812553,
      "num_input_tokens_seen": 114692870,
      "step": 5329,
      "time_per_iteration": 2.623339891433716
    },
    {
      "auxiliary_loss_clip": 0.01170132,
      "auxiliary_loss_mlp": 0.01022801,
      "balance_loss_clip": 1.04691696,
      "balance_loss_mlp": 1.0153383,
      "epoch": 0.6408946071063548,
      "flos": 15012636122880.0,
      "grad_norm": 2.066061351782444,
      "language_loss": 0.77742767,
      "learning_rate": 1.2068339186952976e-06,
      "loss": 0.79935706,
      "num_input_tokens_seen": 114710410,
      "step": 5330,
      "time_per_iteration": 3.5556864738464355
    },
    {
      "auxiliary_loss_clip": 0.01178101,
      "auxiliary_loss_mlp": 0.01027135,
      "balance_loss_clip": 1.04953551,
      "balance_loss_mlp": 1.0190587,
      "epoch": 0.6410148499969939,
      "flos": 22528595496960.0,
      "grad_norm": 1.9995683000811444,
      "language_loss": 0.73445082,
      "learning_rate": 1.2061188837969136e-06,
      "loss": 0.75650322,
      "num_input_tokens_seen": 114730020,
      "step": 5331,
      "time_per_iteration": 2.550119400024414
    },
    {
      "auxiliary_loss_clip": 0.01175411,
      "auxiliary_loss_mlp": 0.01025049,
      "balance_loss_clip": 1.04254425,
      "balance_loss_mlp": 1.0166446,
      "epoch": 0.641135092887633,
      "flos": 12422004537600.0,
      "grad_norm": 2.836755467273598,
      "language_loss": 0.84229183,
      "learning_rate": 1.2054039693296631e-06,
      "loss": 0.86429644,
      "num_input_tokens_seen": 114748015,
      "step": 5332,
      "time_per_iteration": 2.615264654159546
    },
    {
      "auxiliary_loss_clip": 0.01171151,
      "auxiliary_loss_mlp": 0.01025809,
      "balance_loss_clip": 1.04093635,
      "balance_loss_mlp": 1.01839387,
      "epoch": 0.6412553357782721,
      "flos": 22127329687680.0,
      "grad_norm": 1.7563795772524515,
      "language_loss": 0.8187843,
      "learning_rate": 1.2046891754019992e-06,
      "loss": 0.84075391,
      "num_input_tokens_seen": 114768625,
      "step": 5333,
      "time_per_iteration": 2.646627902984619
    },
    {
      "auxiliary_loss_clip": 0.01178674,
      "auxiliary_loss_mlp": 0.0102994,
      "balance_loss_clip": 1.04952407,
      "balance_loss_mlp": 1.02211976,
      "epoch": 0.6413755786689112,
      "flos": 15888605097600.0,
      "grad_norm": 2.65036902337045,
      "language_loss": 0.82222563,
      "learning_rate": 1.2039745021223548e-06,
      "loss": 0.84431171,
      "num_input_tokens_seen": 114786045,
      "step": 5334,
      "time_per_iteration": 2.624350070953369
    },
    {
      "auxiliary_loss_clip": 0.0109784,
      "auxiliary_loss_mlp": 0.01001536,
      "balance_loss_clip": 1.02318454,
      "balance_loss_mlp": 1.00073707,
      "epoch": 0.6414958215595503,
      "flos": 68039159955840.0,
      "grad_norm": 0.7905599682959163,
      "language_loss": 0.57044661,
      "learning_rate": 1.2032599495991456e-06,
      "loss": 0.59144038,
      "num_input_tokens_seen": 114850785,
      "step": 5335,
      "time_per_iteration": 3.2918312549591064
    },
    {
      "auxiliary_loss_clip": 0.01175852,
      "auxiliary_loss_mlp": 0.01026363,
      "balance_loss_clip": 1.04796088,
      "balance_loss_mlp": 1.0182091,
      "epoch": 0.6416160644501894,
      "flos": 44091300320640.0,
      "grad_norm": 1.6790855496466617,
      "language_loss": 0.69820207,
      "learning_rate": 1.2025455179407685e-06,
      "loss": 0.72022426,
      "num_input_tokens_seen": 114871945,
      "step": 5336,
      "time_per_iteration": 2.7938685417175293
    },
    {
      "auxiliary_loss_clip": 0.01172489,
      "auxiliary_loss_mlp": 0.01208208,
      "balance_loss_clip": 1.04908371,
      "balance_loss_mlp": 1.00107121,
      "epoch": 0.6417363073408284,
      "flos": 20959837931520.0,
      "grad_norm": 2.3744064583179307,
      "language_loss": 0.73930609,
      "learning_rate": 1.2018312072556022e-06,
      "loss": 0.76311314,
      "num_input_tokens_seen": 114890445,
      "step": 5337,
      "time_per_iteration": 2.5874948501586914
    },
    {
      "auxiliary_loss_clip": 0.01170243,
      "auxiliary_loss_mlp": 0.01208378,
      "balance_loss_clip": 1.04892993,
      "balance_loss_mlp": 1.00091147,
      "epoch": 0.6418565502314676,
      "flos": 22455122227200.0,
      "grad_norm": 2.2412315831733873,
      "language_loss": 0.74519902,
      "learning_rate": 1.2011170176520077e-06,
      "loss": 0.76898527,
      "num_input_tokens_seen": 114911360,
      "step": 5338,
      "time_per_iteration": 2.645413398742676
    },
    {
      "auxiliary_loss_clip": 0.01169651,
      "auxiliary_loss_mlp": 0.01023483,
      "balance_loss_clip": 1.04005647,
      "balance_loss_mlp": 1.01597846,
      "epoch": 0.6419767931221066,
      "flos": 25045502417280.0,
      "grad_norm": 1.6632795144237633,
      "language_loss": 0.81246197,
      "learning_rate": 1.2004029492383256e-06,
      "loss": 0.83439332,
      "num_input_tokens_seen": 114932700,
      "step": 5339,
      "time_per_iteration": 2.7532541751861572
    },
    {
      "auxiliary_loss_clip": 0.01174483,
      "auxiliary_loss_mlp": 0.01027234,
      "balance_loss_clip": 1.05028462,
      "balance_loss_mlp": 1.0193727,
      "epoch": 0.6420970360127457,
      "flos": 19463691709440.0,
      "grad_norm": 8.704825450719145,
      "language_loss": 0.73350561,
      "learning_rate": 1.1996890021228814e-06,
      "loss": 0.75552285,
      "num_input_tokens_seen": 114949475,
      "step": 5340,
      "time_per_iteration": 2.5887105464935303
    },
    {
      "auxiliary_loss_clip": 0.01173055,
      "auxiliary_loss_mlp": 0.0102351,
      "balance_loss_clip": 1.04437113,
      "balance_loss_mlp": 1.0160203,
      "epoch": 0.6422172789033849,
      "flos": 40406147458560.0,
      "grad_norm": 1.821749414174601,
      "language_loss": 0.70140588,
      "learning_rate": 1.1989751764139785e-06,
      "loss": 0.72337151,
      "num_input_tokens_seen": 114973125,
      "step": 5341,
      "time_per_iteration": 2.7950568199157715
    },
    {
      "auxiliary_loss_clip": 0.01178175,
      "auxiliary_loss_mlp": 0.0102308,
      "balance_loss_clip": 1.03852463,
      "balance_loss_mlp": 1.01491463,
      "epoch": 0.6423375217940239,
      "flos": 27672870637440.0,
      "grad_norm": 1.6605386210210658,
      "language_loss": 0.83280087,
      "learning_rate": 1.1982614722199044e-06,
      "loss": 0.85481346,
      "num_input_tokens_seen": 114994300,
      "step": 5342,
      "time_per_iteration": 2.7007410526275635
    },
    {
      "auxiliary_loss_clip": 0.01182628,
      "auxiliary_loss_mlp": 0.0102553,
      "balance_loss_clip": 1.04677093,
      "balance_loss_mlp": 1.01821661,
      "epoch": 0.642457764684663,
      "flos": 18369242259840.0,
      "grad_norm": 2.2291468184478886,
      "language_loss": 0.78231579,
      "learning_rate": 1.1975478896489276e-06,
      "loss": 0.80439734,
      "num_input_tokens_seen": 115012135,
      "step": 5343,
      "time_per_iteration": 2.6076953411102295
    },
    {
      "auxiliary_loss_clip": 0.01169366,
      "auxiliary_loss_mlp": 0.01022226,
      "balance_loss_clip": 1.0483911,
      "balance_loss_mlp": 1.0148139,
      "epoch": 0.6425780075753021,
      "flos": 19750509809280.0,
      "grad_norm": 2.05164991094725,
      "language_loss": 0.76508182,
      "learning_rate": 1.1968344288092981e-06,
      "loss": 0.7869978,
      "num_input_tokens_seen": 115028715,
      "step": 5344,
      "time_per_iteration": 3.4589576721191406
    },
    {
      "auxiliary_loss_clip": 0.01174979,
      "auxiliary_loss_mlp": 0.01208498,
      "balance_loss_clip": 1.04879725,
      "balance_loss_mlp": 1.00104618,
      "epoch": 0.6426982504659412,
      "flos": 20558536208640.0,
      "grad_norm": 1.8594771813914992,
      "language_loss": 0.64699584,
      "learning_rate": 1.1961210898092468e-06,
      "loss": 0.67083061,
      "num_input_tokens_seen": 115047665,
      "step": 5345,
      "time_per_iteration": 2.6190290451049805
    },
    {
      "auxiliary_loss_clip": 0.01184511,
      "auxiliary_loss_mlp": 0.01026474,
      "balance_loss_clip": 1.04942727,
      "balance_loss_mlp": 1.01828778,
      "epoch": 0.6428184933565803,
      "flos": 17851984456320.0,
      "grad_norm": 2.121550866489256,
      "language_loss": 0.79277718,
      "learning_rate": 1.1954078727569874e-06,
      "loss": 0.81488705,
      "num_input_tokens_seen": 115064965,
      "step": 5346,
      "time_per_iteration": 2.5890209674835205
    },
    {
      "auxiliary_loss_clip": 0.01183136,
      "auxiliary_loss_mlp": 0.012083,
      "balance_loss_clip": 1.044209,
      "balance_loss_mlp": 1.00104332,
      "epoch": 0.6429387362472194,
      "flos": 22456953820800.0,
      "grad_norm": 1.7422859665971029,
      "language_loss": 0.77897865,
      "learning_rate": 1.1946947777607141e-06,
      "loss": 0.80289304,
      "num_input_tokens_seen": 115086100,
      "step": 5347,
      "time_per_iteration": 2.6455352306365967
    },
    {
      "auxiliary_loss_clip": 0.01174628,
      "auxiliary_loss_mlp": 0.01025827,
      "balance_loss_clip": 1.04171181,
      "balance_loss_mlp": 1.01778078,
      "epoch": 0.6430589791378585,
      "flos": 24752579005440.0,
      "grad_norm": 2.428730699378667,
      "language_loss": 0.80382282,
      "learning_rate": 1.1939818049286024e-06,
      "loss": 0.82582736,
      "num_input_tokens_seen": 115104260,
      "step": 5348,
      "time_per_iteration": 3.633572816848755
    },
    {
      "auxiliary_loss_clip": 0.01171528,
      "auxiliary_loss_mlp": 0.01027717,
      "balance_loss_clip": 1.04013133,
      "balance_loss_mlp": 1.01994777,
      "epoch": 0.6431792220284975,
      "flos": 24901249397760.0,
      "grad_norm": 2.0037161378381367,
      "language_loss": 0.75559467,
      "learning_rate": 1.1932689543688101e-06,
      "loss": 0.77758706,
      "num_input_tokens_seen": 115125365,
      "step": 5349,
      "time_per_iteration": 2.7182652950286865
    },
    {
      "auxiliary_loss_clip": 0.01176684,
      "auxiliary_loss_mlp": 0.01025029,
      "balance_loss_clip": 1.0481199,
      "balance_loss_mlp": 1.01714885,
      "epoch": 0.6432994649191367,
      "flos": 21032305620480.0,
      "grad_norm": 3.897984783534656,
      "language_loss": 0.72617984,
      "learning_rate": 1.1925562261894756e-06,
      "loss": 0.74819702,
      "num_input_tokens_seen": 115144445,
      "step": 5350,
      "time_per_iteration": 2.7784104347229004
    },
    {
      "auxiliary_loss_clip": 0.0117461,
      "auxiliary_loss_mlp": 0.01027587,
      "balance_loss_clip": 1.04618001,
      "balance_loss_mlp": 1.02027392,
      "epoch": 0.6434197078097758,
      "flos": 30884433655680.0,
      "grad_norm": 2.2120870820798677,
      "language_loss": 0.77387828,
      "learning_rate": 1.1918436204987207e-06,
      "loss": 0.79590023,
      "num_input_tokens_seen": 115166305,
      "step": 5351,
      "time_per_iteration": 3.6232454776763916
    },
    {
      "auxiliary_loss_clip": 0.01168774,
      "auxiliary_loss_mlp": 0.01025818,
      "balance_loss_clip": 1.0478828,
      "balance_loss_mlp": 1.01806366,
      "epoch": 0.6435399507004148,
      "flos": 15012492468480.0,
      "grad_norm": 2.4330621510662813,
      "language_loss": 0.81607765,
      "learning_rate": 1.191131137404645e-06,
      "loss": 0.83802354,
      "num_input_tokens_seen": 115183045,
      "step": 5352,
      "time_per_iteration": 2.625746488571167
    },
    {
      "auxiliary_loss_clip": 0.01167528,
      "auxiliary_loss_mlp": 0.01028551,
      "balance_loss_clip": 1.04335856,
      "balance_loss_mlp": 1.02092791,
      "epoch": 0.643660193591054,
      "flos": 19901981462400.0,
      "grad_norm": 2.014639486161025,
      "language_loss": 0.77510321,
      "learning_rate": 1.190418777015333e-06,
      "loss": 0.79706395,
      "num_input_tokens_seen": 115201955,
      "step": 5353,
      "time_per_iteration": 2.811805248260498
    },
    {
      "auxiliary_loss_clip": 0.01175615,
      "auxiliary_loss_mlp": 0.01018623,
      "balance_loss_clip": 1.046736,
      "balance_loss_mlp": 1.01155114,
      "epoch": 0.643780436481693,
      "flos": 24133622820480.0,
      "grad_norm": 1.5721171980743096,
      "language_loss": 0.7370407,
      "learning_rate": 1.1897065394388487e-06,
      "loss": 0.75898308,
      "num_input_tokens_seen": 115222395,
      "step": 5354,
      "time_per_iteration": 2.6631393432617188
    },
    {
      "auxiliary_loss_clip": 0.01178434,
      "auxiliary_loss_mlp": 0.01030768,
      "balance_loss_clip": 1.05099964,
      "balance_loss_mlp": 1.0228231,
      "epoch": 0.6439006793723321,
      "flos": 23148808657920.0,
      "grad_norm": 3.1559364430654564,
      "language_loss": 0.76666915,
      "learning_rate": 1.1889944247832385e-06,
      "loss": 0.78876114,
      "num_input_tokens_seen": 115242635,
      "step": 5355,
      "time_per_iteration": 2.653484344482422
    },
    {
      "auxiliary_loss_clip": 0.01176367,
      "auxiliary_loss_mlp": 0.01028029,
      "balance_loss_clip": 1.04644716,
      "balance_loss_mlp": 1.02061391,
      "epoch": 0.6440209222629713,
      "flos": 23617909301760.0,
      "grad_norm": 1.9408122170460165,
      "language_loss": 0.70503479,
      "learning_rate": 1.1882824331565283e-06,
      "loss": 0.7270788,
      "num_input_tokens_seen": 115262095,
      "step": 5356,
      "time_per_iteration": 3.596632242202759
    },
    {
      "auxiliary_loss_clip": 0.01174441,
      "auxiliary_loss_mlp": 0.01029305,
      "balance_loss_clip": 1.0425334,
      "balance_loss_mlp": 1.02191377,
      "epoch": 0.6441411651536103,
      "flos": 16544872535040.0,
      "grad_norm": 2.8827568238735637,
      "language_loss": 0.89482975,
      "learning_rate": 1.1875705646667287e-06,
      "loss": 0.9168672,
      "num_input_tokens_seen": 115279985,
      "step": 5357,
      "time_per_iteration": 2.667168617248535
    },
    {
      "auxiliary_loss_clip": 0.0117091,
      "auxiliary_loss_mlp": 0.01026313,
      "balance_loss_clip": 1.04554236,
      "balance_loss_mlp": 1.01791477,
      "epoch": 0.6442614080442494,
      "flos": 25410965345280.0,
      "grad_norm": 2.1759959789548615,
      "language_loss": 0.7521413,
      "learning_rate": 1.1868588194218282e-06,
      "loss": 0.77411354,
      "num_input_tokens_seen": 115300365,
      "step": 5358,
      "time_per_iteration": 2.6283929347991943
    },
    {
      "auxiliary_loss_clip": 0.01183502,
      "auxiliary_loss_mlp": 0.01028124,
      "balance_loss_clip": 1.04637957,
      "balance_loss_mlp": 1.01965988,
      "epoch": 0.6443816509348885,
      "flos": 28294017552000.0,
      "grad_norm": 1.9343041667343757,
      "language_loss": 0.74330366,
      "learning_rate": 1.1861471975297979e-06,
      "loss": 0.7654199,
      "num_input_tokens_seen": 115322060,
      "step": 5359,
      "time_per_iteration": 2.655851364135742
    },
    {
      "auxiliary_loss_clip": 0.01177509,
      "auxiliary_loss_mlp": 0.01021489,
      "balance_loss_clip": 1.04678738,
      "balance_loss_mlp": 1.01298356,
      "epoch": 0.6445018938255276,
      "flos": 36690075964800.0,
      "grad_norm": 2.705398400379776,
      "language_loss": 0.70971072,
      "learning_rate": 1.185435699098591e-06,
      "loss": 0.73170066,
      "num_input_tokens_seen": 115348255,
      "step": 5360,
      "time_per_iteration": 2.845629930496216
    },
    {
      "auxiliary_loss_clip": 0.01182343,
      "auxiliary_loss_mlp": 0.01022436,
      "balance_loss_clip": 1.04776549,
      "balance_loss_mlp": 1.01474071,
      "epoch": 0.6446221367161666,
      "flos": 14501411804160.0,
      "grad_norm": 2.8513108100924023,
      "language_loss": 0.78277737,
      "learning_rate": 1.1847243242361403e-06,
      "loss": 0.80482519,
      "num_input_tokens_seen": 115366845,
      "step": 5361,
      "time_per_iteration": 2.614305019378662
    },
    {
      "auxiliary_loss_clip": 0.01176725,
      "auxiliary_loss_mlp": 0.01029866,
      "balance_loss_clip": 1.04613507,
      "balance_loss_mlp": 1.02248073,
      "epoch": 0.6447423796068057,
      "flos": 24609367480320.0,
      "grad_norm": 1.7527042492003002,
      "language_loss": 0.78248233,
      "learning_rate": 1.1840130730503624e-06,
      "loss": 0.80454826,
      "num_input_tokens_seen": 115388125,
      "step": 5362,
      "time_per_iteration": 2.6487936973571777
    },
    {
      "auxiliary_loss_clip": 0.01172965,
      "auxiliary_loss_mlp": 0.01026323,
      "balance_loss_clip": 1.05017543,
      "balance_loss_mlp": 1.01869929,
      "epoch": 0.6448626224974449,
      "flos": 25047298097280.0,
      "grad_norm": 1.9880820264109884,
      "language_loss": 0.74931479,
      "learning_rate": 1.1833019456491518e-06,
      "loss": 0.77130771,
      "num_input_tokens_seen": 115409655,
      "step": 5363,
      "time_per_iteration": 2.6750855445861816
    },
    {
      "auxiliary_loss_clip": 0.01177982,
      "auxiliary_loss_mlp": 0.01026895,
      "balance_loss_clip": 1.05064964,
      "balance_loss_mlp": 1.01896179,
      "epoch": 0.6449828653880839,
      "flos": 22530355263360.0,
      "grad_norm": 2.221742163478454,
      "language_loss": 0.79130501,
      "learning_rate": 1.1825909421403871e-06,
      "loss": 0.81335378,
      "num_input_tokens_seen": 115428750,
      "step": 5364,
      "time_per_iteration": 2.770440101623535
    },
    {
      "auxiliary_loss_clip": 0.0117624,
      "auxiliary_loss_mlp": 0.01026944,
      "balance_loss_clip": 1.04887617,
      "balance_loss_mlp": 1.01968384,
      "epoch": 0.645103108278723,
      "flos": 25695736369920.0,
      "grad_norm": 1.7489184503422608,
      "language_loss": 0.76391375,
      "learning_rate": 1.181880062631926e-06,
      "loss": 0.78594559,
      "num_input_tokens_seen": 115448085,
      "step": 5365,
      "time_per_iteration": 2.59627366065979
    },
    {
      "auxiliary_loss_clip": 0.01172293,
      "auxiliary_loss_mlp": 0.01033198,
      "balance_loss_clip": 1.04765248,
      "balance_loss_mlp": 1.02422142,
      "epoch": 0.6452233511693621,
      "flos": 27450331925760.0,
      "grad_norm": 2.632511136057878,
      "language_loss": 0.84726405,
      "learning_rate": 1.1811693072316093e-06,
      "loss": 0.86931902,
      "num_input_tokens_seen": 115465765,
      "step": 5366,
      "time_per_iteration": 2.704136371612549
    },
    {
      "auxiliary_loss_clip": 0.01171412,
      "auxiliary_loss_mlp": 0.01208534,
      "balance_loss_clip": 1.04750049,
      "balance_loss_mlp": 1.00103045,
      "epoch": 0.6453435940600012,
      "flos": 19208618254080.0,
      "grad_norm": 2.2950801675609616,
      "language_loss": 0.84528685,
      "learning_rate": 1.1804586760472574e-06,
      "loss": 0.86908633,
      "num_input_tokens_seen": 115482230,
      "step": 5367,
      "time_per_iteration": 2.5461857318878174
    },
    {
      "auxiliary_loss_clip": 0.01178942,
      "auxiliary_loss_mlp": 0.01024533,
      "balance_loss_clip": 1.04494584,
      "balance_loss_mlp": 1.01637924,
      "epoch": 0.6454638369506402,
      "flos": 25737680476800.0,
      "grad_norm": 2.2866972156414103,
      "language_loss": 0.80131537,
      "learning_rate": 1.1797481691866736e-06,
      "loss": 0.82335013,
      "num_input_tokens_seen": 115499455,
      "step": 5368,
      "time_per_iteration": 2.7248220443725586
    },
    {
      "auxiliary_loss_clip": 0.01166652,
      "auxiliary_loss_mlp": 0.01034823,
      "balance_loss_clip": 1.04570365,
      "balance_loss_mlp": 1.02718151,
      "epoch": 0.6455840798412794,
      "flos": 20989176364800.0,
      "grad_norm": 1.934836400981974,
      "language_loss": 0.82828045,
      "learning_rate": 1.1790377867576393e-06,
      "loss": 0.85029519,
      "num_input_tokens_seen": 115517205,
      "step": 5369,
      "time_per_iteration": 2.692424774169922
    },
    {
      "auxiliary_loss_clip": 0.01182,
      "auxiliary_loss_mlp": 0.01023795,
      "balance_loss_clip": 1.04800737,
      "balance_loss_mlp": 1.01602888,
      "epoch": 0.6457043227319185,
      "flos": 26067556005120.0,
      "grad_norm": 55.778957546456404,
      "language_loss": 0.7648173,
      "learning_rate": 1.1783275288679203e-06,
      "loss": 0.78687525,
      "num_input_tokens_seen": 115534370,
      "step": 5370,
      "time_per_iteration": 3.6710803508758545
    },
    {
      "auxiliary_loss_clip": 0.01080666,
      "auxiliary_loss_mlp": 0.01001576,
      "balance_loss_clip": 1.01776886,
      "balance_loss_mlp": 1.00083113,
      "epoch": 0.6458245656225575,
      "flos": 60370831088640.0,
      "grad_norm": 0.8505437359983801,
      "language_loss": 0.57077205,
      "learning_rate": 1.177617395625262e-06,
      "loss": 0.59159446,
      "num_input_tokens_seen": 115592345,
      "step": 5371,
      "time_per_iteration": 3.146365165710449
    },
    {
      "auxiliary_loss_clip": 0.01174014,
      "auxiliary_loss_mlp": 0.01028922,
      "balance_loss_clip": 1.0496012,
      "balance_loss_mlp": 1.02122748,
      "epoch": 0.6459448085131967,
      "flos": 23076771932160.0,
      "grad_norm": 1.9643989178717864,
      "language_loss": 0.75702518,
      "learning_rate": 1.1769073871373908e-06,
      "loss": 0.77905452,
      "num_input_tokens_seen": 115612550,
      "step": 5372,
      "time_per_iteration": 2.7974042892456055
    },
    {
      "auxiliary_loss_clip": 0.01176386,
      "auxiliary_loss_mlp": 0.01025032,
      "balance_loss_clip": 1.04274917,
      "balance_loss_mlp": 1.01737833,
      "epoch": 0.6460650514038357,
      "flos": 22598190097920.0,
      "grad_norm": 4.132175773190701,
      "language_loss": 0.83763307,
      "learning_rate": 1.176197503512015e-06,
      "loss": 0.85964727,
      "num_input_tokens_seen": 115632265,
      "step": 5373,
      "time_per_iteration": 2.664534091949463
    },
    {
      "auxiliary_loss_clip": 0.01175494,
      "auxiliary_loss_mlp": 0.0102575,
      "balance_loss_clip": 1.0473218,
      "balance_loss_mlp": 1.01826978,
      "epoch": 0.6461852942944748,
      "flos": 20266726118400.0,
      "grad_norm": 2.314036193543979,
      "language_loss": 0.82456416,
      "learning_rate": 1.1754877448568223e-06,
      "loss": 0.84657657,
      "num_input_tokens_seen": 115651720,
      "step": 5374,
      "time_per_iteration": 2.692265033721924
    },
    {
      "auxiliary_loss_clip": 0.01176085,
      "auxiliary_loss_mlp": 0.01027039,
      "balance_loss_clip": 1.04485178,
      "balance_loss_mlp": 1.01920712,
      "epoch": 0.646305537185114,
      "flos": 23367109564800.0,
      "grad_norm": 2.5839158090167875,
      "language_loss": 0.89821243,
      "learning_rate": 1.1747781112794837e-06,
      "loss": 0.92024374,
      "num_input_tokens_seen": 115668215,
      "step": 5375,
      "time_per_iteration": 3.5928962230682373
    },
    {
      "auxiliary_loss_clip": 0.01177247,
      "auxiliary_loss_mlp": 0.0102832,
      "balance_loss_clip": 1.04612088,
      "balance_loss_mlp": 1.02075577,
      "epoch": 0.646425780075753,
      "flos": 24277480790400.0,
      "grad_norm": 1.7337197568565863,
      "language_loss": 0.82979912,
      "learning_rate": 1.1740686028876487e-06,
      "loss": 0.85185474,
      "num_input_tokens_seen": 115687080,
      "step": 5376,
      "time_per_iteration": 2.698096990585327
    },
    {
      "auxiliary_loss_clip": 0.01171873,
      "auxiliary_loss_mlp": 0.01024605,
      "balance_loss_clip": 1.04915333,
      "balance_loss_mlp": 1.01722646,
      "epoch": 0.6465460229663921,
      "flos": 20813968800000.0,
      "grad_norm": 3.3961816730464762,
      "language_loss": 0.74620628,
      "learning_rate": 1.1733592197889507e-06,
      "loss": 0.76817101,
      "num_input_tokens_seen": 115703990,
      "step": 5377,
      "time_per_iteration": 2.6043503284454346
    },
    {
      "auxiliary_loss_clip": 0.01166002,
      "auxiliary_loss_mlp": 0.01023257,
      "balance_loss_clip": 1.0477221,
      "balance_loss_mlp": 1.01612568,
      "epoch": 0.6466662658570312,
      "flos": 22853299466880.0,
      "grad_norm": 6.908024988621738,
      "language_loss": 0.72641391,
      "learning_rate": 1.1726499620910014e-06,
      "loss": 0.74830651,
      "num_input_tokens_seen": 115724270,
      "step": 5378,
      "time_per_iteration": 3.5374836921691895
    },
    {
      "auxiliary_loss_clip": 0.01170111,
      "auxiliary_loss_mlp": 0.01022088,
      "balance_loss_clip": 1.04701197,
      "balance_loss_mlp": 1.0140475,
      "epoch": 0.6467865087476703,
      "flos": 15304553953920.0,
      "grad_norm": 2.294237261977133,
      "language_loss": 0.77774721,
      "learning_rate": 1.1719408299013955e-06,
      "loss": 0.79966927,
      "num_input_tokens_seen": 115742995,
      "step": 5379,
      "time_per_iteration": 2.5875706672668457
    },
    {
      "auxiliary_loss_clip": 0.0117148,
      "auxiliary_loss_mlp": 0.01027569,
      "balance_loss_clip": 1.05133915,
      "balance_loss_mlp": 1.01998425,
      "epoch": 0.6469067516383094,
      "flos": 19573650218880.0,
      "grad_norm": 2.1626110731885904,
      "language_loss": 0.75646913,
      "learning_rate": 1.1712318233277067e-06,
      "loss": 0.77845955,
      "num_input_tokens_seen": 115762015,
      "step": 5380,
      "time_per_iteration": 2.5670454502105713
    },
    {
      "auxiliary_loss_clip": 0.01080241,
      "auxiliary_loss_mlp": 0.01003487,
      "balance_loss_clip": 1.01821566,
      "balance_loss_mlp": 1.00274229,
      "epoch": 0.6470269945289485,
      "flos": 65098002522240.0,
      "grad_norm": 0.7536786128509534,
      "language_loss": 0.57837123,
      "learning_rate": 1.1705229424774916e-06,
      "loss": 0.59920847,
      "num_input_tokens_seen": 115816285,
      "step": 5381,
      "time_per_iteration": 3.026400089263916
    },
    {
      "auxiliary_loss_clip": 0.01172857,
      "auxiliary_loss_mlp": 0.01027934,
      "balance_loss_clip": 1.04444122,
      "balance_loss_mlp": 1.02028108,
      "epoch": 0.6471472374195876,
      "flos": 30696943639680.0,
      "grad_norm": 1.6399248558720585,
      "language_loss": 0.64335382,
      "learning_rate": 1.1698141874582867e-06,
      "loss": 0.66536176,
      "num_input_tokens_seen": 115837330,
      "step": 5382,
      "time_per_iteration": 2.6782689094543457
    },
    {
      "auxiliary_loss_clip": 0.01170251,
      "auxiliary_loss_mlp": 0.01026595,
      "balance_loss_clip": 1.05048037,
      "balance_loss_mlp": 1.01949322,
      "epoch": 0.6472674803102266,
      "flos": 20521835487360.0,
      "grad_norm": 1.6724727984184302,
      "language_loss": 0.72387421,
      "learning_rate": 1.169105558377609e-06,
      "loss": 0.7458427,
      "num_input_tokens_seen": 115857420,
      "step": 5383,
      "time_per_iteration": 3.3794867992401123
    },
    {
      "auxiliary_loss_clip": 0.01179674,
      "auxiliary_loss_mlp": 0.01208043,
      "balance_loss_clip": 1.04654241,
      "balance_loss_mlp": 1.00096607,
      "epoch": 0.6473877232008658,
      "flos": 24715447320960.0,
      "grad_norm": 1.619537626310691,
      "language_loss": 0.78619915,
      "learning_rate": 1.1683970553429587e-06,
      "loss": 0.81007636,
      "num_input_tokens_seen": 115878875,
      "step": 5384,
      "time_per_iteration": 2.603091239929199
    },
    {
      "auxiliary_loss_clip": 0.01181476,
      "auxiliary_loss_mlp": 0.01026746,
      "balance_loss_clip": 1.04562545,
      "balance_loss_mlp": 1.01877117,
      "epoch": 0.6475079660915048,
      "flos": 15885552441600.0,
      "grad_norm": 2.47263189617792,
      "language_loss": 0.82472765,
      "learning_rate": 1.1676886784618128e-06,
      "loss": 0.84680986,
      "num_input_tokens_seen": 115895540,
      "step": 5385,
      "time_per_iteration": 2.5687663555145264
    },
    {
      "auxiliary_loss_clip": 0.01174246,
      "auxiliary_loss_mlp": 0.01022842,
      "balance_loss_clip": 1.04935265,
      "balance_loss_mlp": 1.01489043,
      "epoch": 0.6476282089821439,
      "flos": 17381590922880.0,
      "grad_norm": 2.492133989467728,
      "language_loss": 0.83973992,
      "learning_rate": 1.1669804278416332e-06,
      "loss": 0.86171079,
      "num_input_tokens_seen": 115910265,
      "step": 5386,
      "time_per_iteration": 2.442326545715332
    },
    {
      "auxiliary_loss_clip": 0.01181848,
      "auxiliary_loss_mlp": 0.01023163,
      "balance_loss_clip": 1.04881835,
      "balance_loss_mlp": 1.01521754,
      "epoch": 0.6477484518727831,
      "flos": 20194078861440.0,
      "grad_norm": 2.351283182345054,
      "language_loss": 0.71636093,
      "learning_rate": 1.1662723035898602e-06,
      "loss": 0.73841101,
      "num_input_tokens_seen": 115930025,
      "step": 5387,
      "time_per_iteration": 2.6439974308013916
    },
    {
      "auxiliary_loss_clip": 0.01173769,
      "auxiliary_loss_mlp": 0.01023157,
      "balance_loss_clip": 1.04904532,
      "balance_loss_mlp": 1.0154767,
      "epoch": 0.6478686947634221,
      "flos": 25410426641280.0,
      "grad_norm": 2.023045147630186,
      "language_loss": 0.81954408,
      "learning_rate": 1.165564305813915e-06,
      "loss": 0.84151328,
      "num_input_tokens_seen": 115949025,
      "step": 5388,
      "time_per_iteration": 2.6407394409179688
    },
    {
      "auxiliary_loss_clip": 0.01173263,
      "auxiliary_loss_mlp": 0.01024749,
      "balance_loss_clip": 1.04844832,
      "balance_loss_mlp": 1.01736951,
      "epoch": 0.6479889376540612,
      "flos": 20083581648000.0,
      "grad_norm": 3.022813098623261,
      "language_loss": 0.81262952,
      "learning_rate": 1.1648564346212019e-06,
      "loss": 0.83460963,
      "num_input_tokens_seen": 115968145,
      "step": 5389,
      "time_per_iteration": 2.582846164703369
    },
    {
      "auxiliary_loss_clip": 0.01168116,
      "auxiliary_loss_mlp": 0.01026538,
      "balance_loss_clip": 1.04914439,
      "balance_loss_mlp": 1.01901567,
      "epoch": 0.6481091805447003,
      "flos": 26758082039040.0,
      "grad_norm": 2.2834410082494676,
      "language_loss": 0.76357341,
      "learning_rate": 1.164148690119104e-06,
      "loss": 0.78551996,
      "num_input_tokens_seen": 115989425,
      "step": 5390,
      "time_per_iteration": 2.649909257888794
    },
    {
      "auxiliary_loss_clip": 0.01168336,
      "auxiliary_loss_mlp": 0.01025889,
      "balance_loss_clip": 1.048455,
      "balance_loss_mlp": 1.01830697,
      "epoch": 0.6482294234353394,
      "flos": 23952094462080.0,
      "grad_norm": 1.777853599387126,
      "language_loss": 0.74150431,
      "learning_rate": 1.163441072414985e-06,
      "loss": 0.76344657,
      "num_input_tokens_seen": 116009630,
      "step": 5391,
      "time_per_iteration": 2.5837368965148926
    },
    {
      "auxiliary_loss_clip": 0.01174237,
      "auxiliary_loss_mlp": 0.01020793,
      "balance_loss_clip": 1.04991174,
      "balance_loss_mlp": 1.01323247,
      "epoch": 0.6483496663259785,
      "flos": 26209833776640.0,
      "grad_norm": 2.1626003893549166,
      "language_loss": 0.69786614,
      "learning_rate": 1.16273358161619e-06,
      "loss": 0.71981645,
      "num_input_tokens_seen": 116029965,
      "step": 5392,
      "time_per_iteration": 2.6406190395355225
    },
    {
      "auxiliary_loss_clip": 0.0118697,
      "auxiliary_loss_mlp": 0.01029012,
      "balance_loss_clip": 1.04998446,
      "balance_loss_mlp": 1.02131093,
      "epoch": 0.6484699092166175,
      "flos": 20922239370240.0,
      "grad_norm": 6.186208005911358,
      "language_loss": 0.83717847,
      "learning_rate": 1.1620262178300446e-06,
      "loss": 0.85933828,
      "num_input_tokens_seen": 116048580,
      "step": 5393,
      "time_per_iteration": 2.6203176975250244
    },
    {
      "auxiliary_loss_clip": 0.01179404,
      "auxiliary_loss_mlp": 0.01025658,
      "balance_loss_clip": 1.04402506,
      "balance_loss_mlp": 1.01801634,
      "epoch": 0.6485901521072567,
      "flos": 33072865678080.0,
      "grad_norm": 1.8204037437840983,
      "language_loss": 0.75799197,
      "learning_rate": 1.1613189811638563e-06,
      "loss": 0.78004253,
      "num_input_tokens_seen": 116070305,
      "step": 5394,
      "time_per_iteration": 2.757779359817505
    },
    {
      "auxiliary_loss_clip": 0.01177315,
      "auxiliary_loss_mlp": 0.01025555,
      "balance_loss_clip": 1.05087209,
      "balance_loss_mlp": 1.0181669,
      "epoch": 0.6487103949978957,
      "flos": 22274060745600.0,
      "grad_norm": 2.4385104550604795,
      "language_loss": 0.78232831,
      "learning_rate": 1.1606118717249117e-06,
      "loss": 0.80435699,
      "num_input_tokens_seen": 116090405,
      "step": 5395,
      "time_per_iteration": 2.5783379077911377
    },
    {
      "auxiliary_loss_clip": 0.01175937,
      "auxiliary_loss_mlp": 0.01024896,
      "balance_loss_clip": 1.05048704,
      "balance_loss_mlp": 1.01703966,
      "epoch": 0.6488306378885348,
      "flos": 22930400010240.0,
      "grad_norm": 2.028750482639329,
      "language_loss": 0.6776402,
      "learning_rate": 1.1599048896204787e-06,
      "loss": 0.6996485,
      "num_input_tokens_seen": 116110285,
      "step": 5396,
      "time_per_iteration": 2.6161742210388184
    },
    {
      "auxiliary_loss_clip": 0.01181717,
      "auxiliary_loss_mlp": 0.01023505,
      "balance_loss_clip": 1.04599619,
      "balance_loss_mlp": 1.01620388,
      "epoch": 0.648950880779174,
      "flos": 20376110010240.0,
      "grad_norm": 3.248608981650227,
      "language_loss": 0.80267537,
      "learning_rate": 1.1591980349578061e-06,
      "loss": 0.82472765,
      "num_input_tokens_seen": 116128955,
      "step": 5397,
      "time_per_iteration": 3.7178797721862793
    },
    {
      "auxiliary_loss_clip": 0.01089685,
      "auxiliary_loss_mlp": 0.00999736,
      "balance_loss_clip": 1.01582766,
      "balance_loss_mlp": 0.99896699,
      "epoch": 0.649071123669813,
      "flos": 59930889310080.0,
      "grad_norm": 0.7325430280540999,
      "language_loss": 0.54279757,
      "learning_rate": 1.158491307844123e-06,
      "loss": 0.56369174,
      "num_input_tokens_seen": 116188875,
      "step": 5398,
      "time_per_iteration": 3.1788177490234375
    },
    {
      "auxiliary_loss_clip": 0.01177012,
      "auxiliary_loss_mlp": 0.01023989,
      "balance_loss_clip": 1.04929566,
      "balance_loss_mlp": 1.01683021,
      "epoch": 0.6491913665604521,
      "flos": 20446566537600.0,
      "grad_norm": 1.7521656412152176,
      "language_loss": 0.84149086,
      "learning_rate": 1.1577847083866387e-06,
      "loss": 0.86350083,
      "num_input_tokens_seen": 116207910,
      "step": 5399,
      "time_per_iteration": 2.6167352199554443
    },
    {
      "auxiliary_loss_clip": 0.01166793,
      "auxiliary_loss_mlp": 0.01025312,
      "balance_loss_clip": 1.04636812,
      "balance_loss_mlp": 1.01735449,
      "epoch": 0.6493116094510912,
      "flos": 16946820702720.0,
      "grad_norm": 1.9742300040988732,
      "language_loss": 0.72252876,
      "learning_rate": 1.1570782366925453e-06,
      "loss": 0.74444985,
      "num_input_tokens_seen": 116226425,
      "step": 5400,
      "time_per_iteration": 2.5853688716888428
    },
    {
      "auxiliary_loss_clip": 0.0117659,
      "auxiliary_loss_mlp": 0.01023285,
      "balance_loss_clip": 1.04351532,
      "balance_loss_mlp": 1.01581931,
      "epoch": 0.6494318523417303,
      "flos": 18802935072000.0,
      "grad_norm": 1.6141929234139645,
      "language_loss": 0.75499487,
      "learning_rate": 1.1563718928690132e-06,
      "loss": 0.77699363,
      "num_input_tokens_seen": 116243860,
      "step": 5401,
      "time_per_iteration": 2.659794807434082
    },
    {
      "auxiliary_loss_clip": 0.01179539,
      "auxiliary_loss_mlp": 0.01028315,
      "balance_loss_clip": 1.04670751,
      "balance_loss_mlp": 1.02027178,
      "epoch": 0.6495520952323693,
      "flos": 18982847318400.0,
      "grad_norm": 2.719244180264706,
      "language_loss": 0.71432394,
      "learning_rate": 1.1556656770231942e-06,
      "loss": 0.73640251,
      "num_input_tokens_seen": 116260055,
      "step": 5402,
      "time_per_iteration": 3.5646934509277344
    },
    {
      "auxiliary_loss_clip": 0.0117519,
      "auxiliary_loss_mlp": 0.01024206,
      "balance_loss_clip": 1.04799736,
      "balance_loss_mlp": 1.01714873,
      "epoch": 0.6496723381230085,
      "flos": 22745388032640.0,
      "grad_norm": 1.7367744573195427,
      "language_loss": 0.76472294,
      "learning_rate": 1.1549595892622207e-06,
      "loss": 0.78671694,
      "num_input_tokens_seen": 116278825,
      "step": 5403,
      "time_per_iteration": 2.6069202423095703
    },
    {
      "auxiliary_loss_clip": 0.01092249,
      "auxiliary_loss_mlp": 0.0100078,
      "balance_loss_clip": 1.01949108,
      "balance_loss_mlp": 0.99999315,
      "epoch": 0.6497925810136476,
      "flos": 62145283887360.0,
      "grad_norm": 0.8326182801327149,
      "language_loss": 0.59006071,
      "learning_rate": 1.1542536296932047e-06,
      "loss": 0.610991,
      "num_input_tokens_seen": 116342360,
      "step": 5404,
      "time_per_iteration": 3.1789565086364746
    },
    {
      "auxiliary_loss_clip": 0.01184758,
      "auxiliary_loss_mlp": 0.01029369,
      "balance_loss_clip": 1.04368067,
      "balance_loss_mlp": 1.0209651,
      "epoch": 0.6499128239042866,
      "flos": 20156731695360.0,
      "grad_norm": 2.8369824870025373,
      "language_loss": 0.70209336,
      "learning_rate": 1.1535477984232414e-06,
      "loss": 0.72423458,
      "num_input_tokens_seen": 116362235,
      "step": 5405,
      "time_per_iteration": 3.5749096870422363
    },
    {
      "auxiliary_loss_clip": 0.01184454,
      "auxiliary_loss_mlp": 0.01026617,
      "balance_loss_clip": 1.04026866,
      "balance_loss_mlp": 1.018785,
      "epoch": 0.6500330667949258,
      "flos": 24462420940800.0,
      "grad_norm": 1.9872426598038488,
      "language_loss": 0.76895785,
      "learning_rate": 1.152842095559404e-06,
      "loss": 0.79106861,
      "num_input_tokens_seen": 116382895,
      "step": 5406,
      "time_per_iteration": 2.7522225379943848
    },
    {
      "auxiliary_loss_clip": 0.01179129,
      "auxiliary_loss_mlp": 0.01026517,
      "balance_loss_clip": 1.04488826,
      "balance_loss_mlp": 1.01913238,
      "epoch": 0.6501533096855648,
      "flos": 25477399549440.0,
      "grad_norm": 4.850145284780769,
      "language_loss": 0.76915336,
      "learning_rate": 1.1521365212087474e-06,
      "loss": 0.79120982,
      "num_input_tokens_seen": 116402880,
      "step": 5407,
      "time_per_iteration": 2.6880605220794678
    },
    {
      "auxiliary_loss_clip": 0.01174495,
      "auxiliary_loss_mlp": 0.01025189,
      "balance_loss_clip": 1.04730177,
      "balance_loss_mlp": 1.01736879,
      "epoch": 0.6502735525762039,
      "flos": 44819245347840.0,
      "grad_norm": 1.7360068699147708,
      "language_loss": 0.7083295,
      "learning_rate": 1.1514310754783062e-06,
      "loss": 0.73032629,
      "num_input_tokens_seen": 116425830,
      "step": 5408,
      "time_per_iteration": 2.8473458290100098
    },
    {
      "auxiliary_loss_clip": 0.0117903,
      "auxiliary_loss_mlp": 0.0102175,
      "balance_loss_clip": 1.04761291,
      "balance_loss_mlp": 1.01376319,
      "epoch": 0.6503937954668431,
      "flos": 28658546726400.0,
      "grad_norm": 2.136079332954455,
      "language_loss": 0.73366642,
      "learning_rate": 1.1507257584750964e-06,
      "loss": 0.75567424,
      "num_input_tokens_seen": 116446010,
      "step": 5409,
      "time_per_iteration": 3.5896618366241455
    },
    {
      "auxiliary_loss_clip": 0.01171975,
      "auxiliary_loss_mlp": 0.01027544,
      "balance_loss_clip": 1.04956055,
      "balance_loss_mlp": 1.01956284,
      "epoch": 0.6505140383574821,
      "flos": 20922562592640.0,
      "grad_norm": 2.3697284664034344,
      "language_loss": 0.77523178,
      "learning_rate": 1.150020570306113e-06,
      "loss": 0.79722697,
      "num_input_tokens_seen": 116465150,
      "step": 5410,
      "time_per_iteration": 2.5596370697021484
    },
    {
      "auxiliary_loss_clip": 0.01170164,
      "auxiliary_loss_mlp": 0.01024828,
      "balance_loss_clip": 1.04271269,
      "balance_loss_mlp": 1.01684058,
      "epoch": 0.6506342812481212,
      "flos": 20595236929920.0,
      "grad_norm": 1.9173221257095086,
      "language_loss": 0.75333548,
      "learning_rate": 1.1493155110783338e-06,
      "loss": 0.77528536,
      "num_input_tokens_seen": 116483675,
      "step": 5411,
      "time_per_iteration": 2.643226385116577
    },
    {
      "auxiliary_loss_clip": 0.01174923,
      "auxiliary_loss_mlp": 0.01024074,
      "balance_loss_clip": 1.04882789,
      "balance_loss_mlp": 1.01614058,
      "epoch": 0.6507545241387603,
      "flos": 30226478279040.0,
      "grad_norm": 2.181524270654717,
      "language_loss": 0.70966446,
      "learning_rate": 1.1486105808987155e-06,
      "loss": 0.73165441,
      "num_input_tokens_seen": 116505165,
      "step": 5412,
      "time_per_iteration": 2.6741573810577393
    },
    {
      "auxiliary_loss_clip": 0.01177499,
      "auxiliary_loss_mlp": 0.01020775,
      "balance_loss_clip": 1.05063772,
      "balance_loss_mlp": 1.01274574,
      "epoch": 0.6508747670293994,
      "flos": 17128241320320.0,
      "grad_norm": 2.062373767898089,
      "language_loss": 0.81220299,
      "learning_rate": 1.1479057798741947e-06,
      "loss": 0.83418572,
      "num_input_tokens_seen": 116523220,
      "step": 5413,
      "time_per_iteration": 2.597597122192383
    },
    {
      "auxiliary_loss_clip": 0.01078211,
      "auxiliary_loss_mlp": 0.01000546,
      "balance_loss_clip": 1.01602972,
      "balance_loss_mlp": 0.99975359,
      "epoch": 0.6509950099200384,
      "flos": 68559826573440.0,
      "grad_norm": 0.7882329858556897,
      "language_loss": 0.53342521,
      "learning_rate": 1.14720110811169e-06,
      "loss": 0.55421281,
      "num_input_tokens_seen": 116580450,
      "step": 5414,
      "time_per_iteration": 3.1686136722564697
    },
    {
      "auxiliary_loss_clip": 0.01178663,
      "auxiliary_loss_mlp": 0.01025344,
      "balance_loss_clip": 1.04907823,
      "balance_loss_mlp": 1.01782537,
      "epoch": 0.6511152528106776,
      "flos": 22347462188160.0,
      "grad_norm": 1.8477016072261485,
      "language_loss": 0.76578408,
      "learning_rate": 1.146496565718098e-06,
      "loss": 0.78782415,
      "num_input_tokens_seen": 116601020,
      "step": 5415,
      "time_per_iteration": 2.6775147914886475
    },
    {
      "auxiliary_loss_clip": 0.01176588,
      "auxiliary_loss_mlp": 0.01025229,
      "balance_loss_clip": 1.04808569,
      "balance_loss_mlp": 1.01683068,
      "epoch": 0.6512354957013167,
      "flos": 20522158709760.0,
      "grad_norm": 1.9727267908089072,
      "language_loss": 0.75872803,
      "learning_rate": 1.1457921528002996e-06,
      "loss": 0.78074622,
      "num_input_tokens_seen": 116619455,
      "step": 5416,
      "time_per_iteration": 2.689486503601074
    },
    {
      "auxiliary_loss_clip": 0.01172333,
      "auxiliary_loss_mlp": 0.01208223,
      "balance_loss_clip": 1.04988492,
      "balance_loss_mlp": 1.00101089,
      "epoch": 0.6513557385919557,
      "flos": 32337342881280.0,
      "grad_norm": 2.3421785755808777,
      "language_loss": 0.72137737,
      "learning_rate": 1.1450878694651522e-06,
      "loss": 0.74518287,
      "num_input_tokens_seen": 116640020,
      "step": 5417,
      "time_per_iteration": 2.7091803550720215
    },
    {
      "auxiliary_loss_clip": 0.01179362,
      "auxiliary_loss_mlp": 0.01023961,
      "balance_loss_clip": 1.04035735,
      "balance_loss_mlp": 1.01623619,
      "epoch": 0.6514759814825949,
      "flos": 12093206417280.0,
      "grad_norm": 5.601488526315422,
      "language_loss": 0.62975103,
      "learning_rate": 1.1443837158194954e-06,
      "loss": 0.65178424,
      "num_input_tokens_seen": 116655165,
      "step": 5418,
      "time_per_iteration": 2.6648800373077393
    },
    {
      "auxiliary_loss_clip": 0.01179646,
      "auxiliary_loss_mlp": 0.01026288,
      "balance_loss_clip": 1.05054617,
      "balance_loss_mlp": 1.01839614,
      "epoch": 0.651596224373234,
      "flos": 22526907557760.0,
      "grad_norm": 1.5188826998518103,
      "language_loss": 0.74297297,
      "learning_rate": 1.1436796919701484e-06,
      "loss": 0.76503229,
      "num_input_tokens_seen": 116673880,
      "step": 5419,
      "time_per_iteration": 2.6028757095336914
    },
    {
      "auxiliary_loss_clip": 0.01176166,
      "auxiliary_loss_mlp": 0.0102257,
      "balance_loss_clip": 1.04798698,
      "balance_loss_mlp": 1.01487494,
      "epoch": 0.651716467263873,
      "flos": 27818955250560.0,
      "grad_norm": 1.8138439029072366,
      "language_loss": 0.61823428,
      "learning_rate": 1.1429757980239115e-06,
      "loss": 0.6402216,
      "num_input_tokens_seen": 116694305,
      "step": 5420,
      "time_per_iteration": 2.655756950378418
    },
    {
      "auxiliary_loss_clip": 0.0117367,
      "auxiliary_loss_mlp": 0.01032902,
      "balance_loss_clip": 1.04938316,
      "balance_loss_mlp": 1.02427757,
      "epoch": 0.6518367101545122,
      "flos": 24316300414080.0,
      "grad_norm": 4.525358091487842,
      "language_loss": 0.81633908,
      "learning_rate": 1.1422720340875636e-06,
      "loss": 0.83840477,
      "num_input_tokens_seen": 116713055,
      "step": 5421,
      "time_per_iteration": 2.5781807899475098
    },
    {
      "auxiliary_loss_clip": 0.01181511,
      "auxiliary_loss_mlp": 0.01023729,
      "balance_loss_clip": 1.04875946,
      "balance_loss_mlp": 1.01625204,
      "epoch": 0.6519569530451512,
      "flos": 20011939971840.0,
      "grad_norm": 1.825208509575354,
      "language_loss": 0.79129344,
      "learning_rate": 1.1415684002678671e-06,
      "loss": 0.81334585,
      "num_input_tokens_seen": 116731815,
      "step": 5422,
      "time_per_iteration": 2.585618019104004
    },
    {
      "auxiliary_loss_clip": 0.01181459,
      "auxiliary_loss_mlp": 0.01025786,
      "balance_loss_clip": 1.04622996,
      "balance_loss_mlp": 1.01753056,
      "epoch": 0.6520771959357903,
      "flos": 21576064682880.0,
      "grad_norm": 2.6846305356927234,
      "language_loss": 0.77811009,
      "learning_rate": 1.1408648966715617e-06,
      "loss": 0.80018258,
      "num_input_tokens_seen": 116749335,
      "step": 5423,
      "time_per_iteration": 2.6188411712646484
    },
    {
      "auxiliary_loss_clip": 0.01178341,
      "auxiliary_loss_mlp": 0.01030912,
      "balance_loss_clip": 1.04422343,
      "balance_loss_mlp": 1.02293754,
      "epoch": 0.6521974388264293,
      "flos": 22711021695360.0,
      "grad_norm": 2.3608953990529735,
      "language_loss": 0.7279954,
      "learning_rate": 1.1401615234053683e-06,
      "loss": 0.75008798,
      "num_input_tokens_seen": 116768155,
      "step": 5424,
      "time_per_iteration": 3.54685378074646
    },
    {
      "auxiliary_loss_clip": 0.01180344,
      "auxiliary_loss_mlp": 0.0102781,
      "balance_loss_clip": 1.04694951,
      "balance_loss_mlp": 1.01968634,
      "epoch": 0.6523176817170685,
      "flos": 23002939526400.0,
      "grad_norm": 2.2279239836794207,
      "language_loss": 0.76210612,
      "learning_rate": 1.1394582805759885e-06,
      "loss": 0.78418767,
      "num_input_tokens_seen": 116787435,
      "step": 5425,
      "time_per_iteration": 2.6390628814697266
    },
    {
      "auxiliary_loss_clip": 0.01177824,
      "auxiliary_loss_mlp": 0.01027906,
      "balance_loss_clip": 1.05082798,
      "balance_loss_mlp": 1.0199604,
      "epoch": 0.6524379246077076,
      "flos": 21688249835520.0,
      "grad_norm": 2.2457690953597877,
      "language_loss": 0.75813866,
      "learning_rate": 1.1387551682901022e-06,
      "loss": 0.78019595,
      "num_input_tokens_seen": 116808040,
      "step": 5426,
      "time_per_iteration": 2.6344261169433594
    },
    {
      "auxiliary_loss_clip": 0.01177806,
      "auxiliary_loss_mlp": 0.01023349,
      "balance_loss_clip": 1.04511356,
      "balance_loss_mlp": 1.01553178,
      "epoch": 0.6525581674983466,
      "flos": 19390936711680.0,
      "grad_norm": 2.1643658494633855,
      "language_loss": 0.70940542,
      "learning_rate": 1.138052186654373e-06,
      "loss": 0.73141694,
      "num_input_tokens_seen": 116825510,
      "step": 5427,
      "time_per_iteration": 2.6047704219818115
    },
    {
      "auxiliary_loss_clip": 0.01179926,
      "auxiliary_loss_mlp": 0.01024343,
      "balance_loss_clip": 1.04696584,
      "balance_loss_mlp": 1.01581371,
      "epoch": 0.6526784103889858,
      "flos": 17165444832000.0,
      "grad_norm": 2.3760875393048155,
      "language_loss": 0.88182014,
      "learning_rate": 1.1373493357754417e-06,
      "loss": 0.90386283,
      "num_input_tokens_seen": 116844415,
      "step": 5428,
      "time_per_iteration": 2.621805429458618
    },
    {
      "auxiliary_loss_clip": 0.01171085,
      "auxiliary_loss_mlp": 0.01021543,
      "balance_loss_clip": 1.04757357,
      "balance_loss_mlp": 1.01421714,
      "epoch": 0.6527986532796248,
      "flos": 18989168112000.0,
      "grad_norm": 1.6334938587280725,
      "language_loss": 0.77453232,
      "learning_rate": 1.1366466157599303e-06,
      "loss": 0.7964586,
      "num_input_tokens_seen": 116863690,
      "step": 5429,
      "time_per_iteration": 3.651095151901245
    },
    {
      "auxiliary_loss_clip": 0.01180361,
      "auxiliary_loss_mlp": 0.01208712,
      "balance_loss_clip": 1.04332352,
      "balance_loss_mlp": 1.00105023,
      "epoch": 0.6529188961702639,
      "flos": 14238581011200.0,
      "grad_norm": 2.7411594473874423,
      "language_loss": 0.75815701,
      "learning_rate": 1.1359440267144412e-06,
      "loss": 0.78204775,
      "num_input_tokens_seen": 116881145,
      "step": 5430,
      "time_per_iteration": 2.6186254024505615
    },
    {
      "auxiliary_loss_clip": 0.0117778,
      "auxiliary_loss_mlp": 0.010209,
      "balance_loss_clip": 1.04916191,
      "balance_loss_mlp": 1.01371789,
      "epoch": 0.653039139060903,
      "flos": 36682929158400.0,
      "grad_norm": 3.803748853656919,
      "language_loss": 0.74358618,
      "learning_rate": 1.1352415687455556e-06,
      "loss": 0.76557297,
      "num_input_tokens_seen": 116902405,
      "step": 5431,
      "time_per_iteration": 2.7763569355010986
    },
    {
      "auxiliary_loss_clip": 0.01175751,
      "auxiliary_loss_mlp": 0.01029547,
      "balance_loss_clip": 1.04921174,
      "balance_loss_mlp": 1.02154183,
      "epoch": 0.6531593819515421,
      "flos": 25376275785600.0,
      "grad_norm": 2.7091621879272467,
      "language_loss": 0.63699877,
      "learning_rate": 1.1345392419598362e-06,
      "loss": 0.65905166,
      "num_input_tokens_seen": 116921285,
      "step": 5432,
      "time_per_iteration": 3.551527976989746
    },
    {
      "auxiliary_loss_clip": 0.01169104,
      "auxiliary_loss_mlp": 0.01020119,
      "balance_loss_clip": 1.04631162,
      "balance_loss_mlp": 1.01168466,
      "epoch": 0.6532796248421812,
      "flos": 21178533888000.0,
      "grad_norm": 2.680854649501171,
      "language_loss": 0.71629798,
      "learning_rate": 1.1338370464638263e-06,
      "loss": 0.73819029,
      "num_input_tokens_seen": 116940685,
      "step": 5433,
      "time_per_iteration": 2.621037721633911
    },
    {
      "auxiliary_loss_clip": 0.01172237,
      "auxiliary_loss_mlp": 0.01023924,
      "balance_loss_clip": 1.04834723,
      "balance_loss_mlp": 1.01583576,
      "epoch": 0.6533998677328203,
      "flos": 17675950878720.0,
      "grad_norm": 2.2377653298903484,
      "language_loss": 0.64601374,
      "learning_rate": 1.1331349823640474e-06,
      "loss": 0.66797537,
      "num_input_tokens_seen": 116958115,
      "step": 5434,
      "time_per_iteration": 2.5370657444000244
    },
    {
      "auxiliary_loss_clip": 0.01176246,
      "auxiliary_loss_mlp": 0.01207765,
      "balance_loss_clip": 1.04777002,
      "balance_loss_mlp": 1.00124693,
      "epoch": 0.6535201106234594,
      "flos": 28400384701440.0,
      "grad_norm": 2.46311734524971,
      "language_loss": 0.78387618,
      "learning_rate": 1.132433049767003e-06,
      "loss": 0.80771625,
      "num_input_tokens_seen": 116976030,
      "step": 5435,
      "time_per_iteration": 2.658707857131958
    },
    {
      "auxiliary_loss_clip": 0.01177677,
      "auxiliary_loss_mlp": 0.0102644,
      "balance_loss_clip": 1.04778051,
      "balance_loss_mlp": 1.01934409,
      "epoch": 0.6536403535140984,
      "flos": 23586667447680.0,
      "grad_norm": 1.5956573509990282,
      "language_loss": 0.81143904,
      "learning_rate": 1.1317312487791748e-06,
      "loss": 0.83348018,
      "num_input_tokens_seen": 116997680,
      "step": 5436,
      "time_per_iteration": 3.5324559211730957
    },
    {
      "auxiliary_loss_clip": 0.01168051,
      "auxiliary_loss_mlp": 0.01027592,
      "balance_loss_clip": 1.04510307,
      "balance_loss_mlp": 1.01957548,
      "epoch": 0.6537605964047376,
      "flos": 21579476474880.0,
      "grad_norm": 1.9497747081376928,
      "language_loss": 0.73012799,
      "learning_rate": 1.1310295795070253e-06,
      "loss": 0.75208443,
      "num_input_tokens_seen": 117017620,
      "step": 5437,
      "time_per_iteration": 2.596266269683838
    },
    {
      "auxiliary_loss_clip": 0.01187729,
      "auxiliary_loss_mlp": 0.01027529,
      "balance_loss_clip": 1.04362857,
      "balance_loss_mlp": 1.0198102,
      "epoch": 0.6538808392953767,
      "flos": 26833997433600.0,
      "grad_norm": 1.746443525113366,
      "language_loss": 0.80866927,
      "learning_rate": 1.1303280420569982e-06,
      "loss": 0.83082187,
      "num_input_tokens_seen": 117039505,
      "step": 5438,
      "time_per_iteration": 2.686140537261963
    },
    {
      "auxiliary_loss_clip": 0.01171402,
      "auxiliary_loss_mlp": 0.0102211,
      "balance_loss_clip": 1.04786229,
      "balance_loss_mlp": 1.01441514,
      "epoch": 0.6540010821860157,
      "flos": 30738241301760.0,
      "grad_norm": 1.9111245755970978,
      "language_loss": 0.77223396,
      "learning_rate": 1.1296266365355158e-06,
      "loss": 0.79416919,
      "num_input_tokens_seen": 117062890,
      "step": 5439,
      "time_per_iteration": 2.716661214828491
    },
    {
      "auxiliary_loss_clip": 0.01184748,
      "auxiliary_loss_mlp": 0.01024584,
      "balance_loss_clip": 1.04694295,
      "balance_loss_mlp": 1.01623297,
      "epoch": 0.6541213250766549,
      "flos": 26907147480960.0,
      "grad_norm": 1.854445115816418,
      "language_loss": 0.73920953,
      "learning_rate": 1.1289253630489806e-06,
      "loss": 0.76130277,
      "num_input_tokens_seen": 117083940,
      "step": 5440,
      "time_per_iteration": 2.7381906509399414
    },
    {
      "auxiliary_loss_clip": 0.01179509,
      "auxiliary_loss_mlp": 0.01026965,
      "balance_loss_clip": 1.04674482,
      "balance_loss_mlp": 1.01866174,
      "epoch": 0.6542415679672939,
      "flos": 19172384409600.0,
      "grad_norm": 2.0019724743926974,
      "language_loss": 0.72294325,
      "learning_rate": 1.1282242217037753e-06,
      "loss": 0.74500799,
      "num_input_tokens_seen": 117101440,
      "step": 5441,
      "time_per_iteration": 2.5795633792877197
    },
    {
      "auxiliary_loss_clip": 0.01180253,
      "auxiliary_loss_mlp": 0.010251,
      "balance_loss_clip": 1.04119587,
      "balance_loss_mlp": 1.0169636,
      "epoch": 0.654361810857933,
      "flos": 48173517100800.0,
      "grad_norm": 2.431922389839427,
      "language_loss": 0.61983114,
      "learning_rate": 1.127523212606262e-06,
      "loss": 0.64188462,
      "num_input_tokens_seen": 117124265,
      "step": 5442,
      "time_per_iteration": 2.8913910388946533
    },
    {
      "auxiliary_loss_clip": 0.01173207,
      "auxiliary_loss_mlp": 0.01025609,
      "balance_loss_clip": 1.0482986,
      "balance_loss_mlp": 1.01781237,
      "epoch": 0.6544820537485722,
      "flos": 26943165843840.0,
      "grad_norm": 2.066733816936066,
      "language_loss": 0.73279059,
      "learning_rate": 1.1268223358627835e-06,
      "loss": 0.75477874,
      "num_input_tokens_seen": 117146755,
      "step": 5443,
      "time_per_iteration": 2.637824535369873
    },
    {
      "auxiliary_loss_clip": 0.01173009,
      "auxiliary_loss_mlp": 0.01021956,
      "balance_loss_clip": 1.04952097,
      "balance_loss_mlp": 1.01376653,
      "epoch": 0.6546022966392112,
      "flos": 20886328748160.0,
      "grad_norm": 1.8151291811391952,
      "language_loss": 0.7240479,
      "learning_rate": 1.126121591579663e-06,
      "loss": 0.74599755,
      "num_input_tokens_seen": 117165960,
      "step": 5444,
      "time_per_iteration": 2.6089329719543457
    },
    {
      "auxiliary_loss_clip": 0.01172263,
      "auxiliary_loss_mlp": 0.01023945,
      "balance_loss_clip": 1.04950666,
      "balance_loss_mlp": 1.01607132,
      "epoch": 0.6547225395298503,
      "flos": 24936693143040.0,
      "grad_norm": 1.5596874179947315,
      "language_loss": 0.6923126,
      "learning_rate": 1.1254209798632018e-06,
      "loss": 0.71427476,
      "num_input_tokens_seen": 117186980,
      "step": 5445,
      "time_per_iteration": 2.675947427749634
    },
    {
      "auxiliary_loss_clip": 0.01171695,
      "auxiliary_loss_mlp": 0.01021745,
      "balance_loss_clip": 1.03909087,
      "balance_loss_mlp": 1.01403785,
      "epoch": 0.6548427824204894,
      "flos": 22565942663040.0,
      "grad_norm": 1.694148163604591,
      "language_loss": 0.84922147,
      "learning_rate": 1.124720500819683e-06,
      "loss": 0.87115586,
      "num_input_tokens_seen": 117205135,
      "step": 5446,
      "time_per_iteration": 2.684962034225464
    },
    {
      "auxiliary_loss_clip": 0.01175575,
      "auxiliary_loss_mlp": 0.01031408,
      "balance_loss_clip": 1.05151165,
      "balance_loss_mlp": 1.02305186,
      "epoch": 0.6549630253111285,
      "flos": 18442500048000.0,
      "grad_norm": 2.0707444974154416,
      "language_loss": 0.82348579,
      "learning_rate": 1.1240201545553682e-06,
      "loss": 0.84555566,
      "num_input_tokens_seen": 117222935,
      "step": 5447,
      "time_per_iteration": 2.576019525527954
    },
    {
      "auxiliary_loss_clip": 0.01177901,
      "auxiliary_loss_mlp": 0.01025296,
      "balance_loss_clip": 1.04481363,
      "balance_loss_mlp": 1.01727867,
      "epoch": 0.6550832682017675,
      "flos": 25187313312000.0,
      "grad_norm": 1.8897587473947421,
      "language_loss": 0.72949851,
      "learning_rate": 1.1233199411764987e-06,
      "loss": 0.75153053,
      "num_input_tokens_seen": 117242370,
      "step": 5448,
      "time_per_iteration": 2.6469247341156006
    },
    {
      "auxiliary_loss_clip": 0.01168793,
      "auxiliary_loss_mlp": 0.01024244,
      "balance_loss_clip": 1.04262459,
      "balance_loss_mlp": 1.01654291,
      "epoch": 0.6552035110924067,
      "flos": 22748153379840.0,
      "grad_norm": 1.7441079201322751,
      "language_loss": 0.68828237,
      "learning_rate": 1.1226198607892978e-06,
      "loss": 0.71021277,
      "num_input_tokens_seen": 117262930,
      "step": 5449,
      "time_per_iteration": 2.7423415184020996
    },
    {
      "auxiliary_loss_clip": 0.01185955,
      "auxiliary_loss_mlp": 0.01024221,
      "balance_loss_clip": 1.04529381,
      "balance_loss_mlp": 1.01637101,
      "epoch": 0.6553237539830458,
      "flos": 21799178012160.0,
      "grad_norm": 1.7408326985402764,
      "language_loss": 0.79792297,
      "learning_rate": 1.1219199134999664e-06,
      "loss": 0.82002467,
      "num_input_tokens_seen": 117281430,
      "step": 5450,
      "time_per_iteration": 2.692660093307495
    },
    {
      "auxiliary_loss_clip": 0.01178506,
      "auxiliary_loss_mlp": 0.01026108,
      "balance_loss_clip": 1.04621172,
      "balance_loss_mlp": 1.01714325,
      "epoch": 0.6554439968736848,
      "flos": 20887226588160.0,
      "grad_norm": 3.2156544456647564,
      "language_loss": 0.78717732,
      "learning_rate": 1.1212200994146863e-06,
      "loss": 0.80922353,
      "num_input_tokens_seen": 117299185,
      "step": 5451,
      "time_per_iteration": 3.5591676235198975
    },
    {
      "auxiliary_loss_clip": 0.01176946,
      "auxiliary_loss_mlp": 0.0102689,
      "balance_loss_clip": 1.04073882,
      "balance_loss_mlp": 1.01914752,
      "epoch": 0.655564239764324,
      "flos": 16139045698560.0,
      "grad_norm": 1.849940104404877,
      "language_loss": 0.75818491,
      "learning_rate": 1.120520418639618e-06,
      "loss": 0.78022325,
      "num_input_tokens_seen": 117317720,
      "step": 5452,
      "time_per_iteration": 2.6472878456115723
    },
    {
      "auxiliary_loss_clip": 0.01174691,
      "auxiliary_loss_mlp": 0.01028425,
      "balance_loss_clip": 1.04901004,
      "balance_loss_mlp": 1.0209322,
      "epoch": 0.655684482654963,
      "flos": 29570354496000.0,
      "grad_norm": 2.4587743188436066,
      "language_loss": 0.83605474,
      "learning_rate": 1.119820871280903e-06,
      "loss": 0.85808587,
      "num_input_tokens_seen": 117338795,
      "step": 5453,
      "time_per_iteration": 2.7490246295928955
    },
    {
      "auxiliary_loss_clip": 0.01172546,
      "auxiliary_loss_mlp": 0.01025011,
      "balance_loss_clip": 1.0474745,
      "balance_loss_mlp": 1.01701856,
      "epoch": 0.6558047255456021,
      "flos": 29789409588480.0,
      "grad_norm": 2.0102593971634346,
      "language_loss": 0.73929095,
      "learning_rate": 1.1191214574446614e-06,
      "loss": 0.76126653,
      "num_input_tokens_seen": 117359040,
      "step": 5454,
      "time_per_iteration": 2.7074055671691895
    },
    {
      "auxiliary_loss_clip": 0.01172338,
      "auxiliary_loss_mlp": 0.01025704,
      "balance_loss_clip": 1.0450865,
      "balance_loss_mlp": 1.01761532,
      "epoch": 0.6559249684362413,
      "flos": 29059166090880.0,
      "grad_norm": 1.6201250501396385,
      "language_loss": 0.80184019,
      "learning_rate": 1.118422177236995e-06,
      "loss": 0.82382065,
      "num_input_tokens_seen": 117380865,
      "step": 5455,
      "time_per_iteration": 3.6538028717041016
    },
    {
      "auxiliary_loss_clip": 0.0117825,
      "auxiliary_loss_mlp": 0.01029623,
      "balance_loss_clip": 1.04588032,
      "balance_loss_mlp": 1.02108216,
      "epoch": 0.6560452113268803,
      "flos": 20225464369920.0,
      "grad_norm": 2.0571913699901403,
      "language_loss": 0.85474336,
      "learning_rate": 1.1177230307639835e-06,
      "loss": 0.87682211,
      "num_input_tokens_seen": 117398405,
      "step": 5456,
      "time_per_iteration": 2.634270668029785
    },
    {
      "auxiliary_loss_clip": 0.01176236,
      "auxiliary_loss_mlp": 0.01024267,
      "balance_loss_clip": 1.04316592,
      "balance_loss_mlp": 1.01653075,
      "epoch": 0.6561654542175194,
      "flos": 25045538330880.0,
      "grad_norm": 1.9759859725626205,
      "language_loss": 0.78873765,
      "learning_rate": 1.1170240181316865e-06,
      "loss": 0.81074268,
      "num_input_tokens_seen": 117419850,
      "step": 5457,
      "time_per_iteration": 2.6436691284179688
    },
    {
      "auxiliary_loss_clip": 0.01174792,
      "auxiliary_loss_mlp": 0.01025762,
      "balance_loss_clip": 1.04110122,
      "balance_loss_mlp": 1.01773942,
      "epoch": 0.6562856971081584,
      "flos": 22856711258880.0,
      "grad_norm": 2.0705916102365443,
      "language_loss": 0.79314101,
      "learning_rate": 1.1163251394461442e-06,
      "loss": 0.81514657,
      "num_input_tokens_seen": 117438330,
      "step": 5458,
      "time_per_iteration": 3.7132394313812256
    },
    {
      "auxiliary_loss_clip": 0.01172943,
      "auxiliary_loss_mlp": 0.01029214,
      "balance_loss_clip": 1.04825807,
      "balance_loss_mlp": 1.02149487,
      "epoch": 0.6564059399987976,
      "flos": 18872565586560.0,
      "grad_norm": 2.116266424265479,
      "language_loss": 0.82485759,
      "learning_rate": 1.1156263948133746e-06,
      "loss": 0.84687918,
      "num_input_tokens_seen": 117454985,
      "step": 5459,
      "time_per_iteration": 2.6565327644348145
    },
    {
      "auxiliary_loss_clip": 0.01176328,
      "auxiliary_loss_mlp": 0.01208293,
      "balance_loss_clip": 1.04396892,
      "balance_loss_mlp": 1.00092959,
      "epoch": 0.6565261828894366,
      "flos": 25484187219840.0,
      "grad_norm": 3.7660818793557165,
      "language_loss": 0.77802587,
      "learning_rate": 1.1149277843393787e-06,
      "loss": 0.80187201,
      "num_input_tokens_seen": 117476145,
      "step": 5460,
      "time_per_iteration": 2.710329055786133
    },
    {
      "auxiliary_loss_clip": 0.01179153,
      "auxiliary_loss_mlp": 0.01208416,
      "balance_loss_clip": 1.03663778,
      "balance_loss_mlp": 1.00089526,
      "epoch": 0.6566464257800757,
      "flos": 19683500987520.0,
      "grad_norm": 2.6825827221628606,
      "language_loss": 0.63542509,
      "learning_rate": 1.1142293081301342e-06,
      "loss": 0.65930074,
      "num_input_tokens_seen": 117494025,
      "step": 5461,
      "time_per_iteration": 2.7249419689178467
    },
    {
      "auxiliary_loss_clip": 0.01171478,
      "auxiliary_loss_mlp": 0.01021295,
      "balance_loss_clip": 1.04513729,
      "balance_loss_mlp": 1.01441395,
      "epoch": 0.6567666686707149,
      "flos": 23514127931520.0,
      "grad_norm": 2.4329212292816336,
      "language_loss": 0.68359268,
      "learning_rate": 1.1135309662915995e-06,
      "loss": 0.70552039,
      "num_input_tokens_seen": 117514190,
      "step": 5462,
      "time_per_iteration": 3.525587797164917
    },
    {
      "auxiliary_loss_clip": 0.01188437,
      "auxiliary_loss_mlp": 0.01023953,
      "balance_loss_clip": 1.04243159,
      "balance_loss_mlp": 1.01626444,
      "epoch": 0.6568869115613539,
      "flos": 32781342896640.0,
      "grad_norm": 3.4337098862930664,
      "language_loss": 0.60227108,
      "learning_rate": 1.112832758929712e-06,
      "loss": 0.62439501,
      "num_input_tokens_seen": 117536800,
      "step": 5463,
      "time_per_iteration": 2.8363959789276123
    },
    {
      "auxiliary_loss_clip": 0.01169568,
      "auxiliary_loss_mlp": 0.01031118,
      "balance_loss_clip": 1.04629338,
      "balance_loss_mlp": 1.02328014,
      "epoch": 0.657007154451993,
      "flos": 18442428220800.0,
      "grad_norm": 1.7277339642219764,
      "language_loss": 0.75050962,
      "learning_rate": 1.11213468615039e-06,
      "loss": 0.77251649,
      "num_input_tokens_seen": 117556230,
      "step": 5464,
      "time_per_iteration": 2.61753249168396
    },
    {
      "auxiliary_loss_clip": 0.01183252,
      "auxiliary_loss_mlp": 0.01024973,
      "balance_loss_clip": 1.04158282,
      "balance_loss_mlp": 1.01748955,
      "epoch": 0.6571273973426321,
      "flos": 25156717902720.0,
      "grad_norm": 1.6609355231292893,
      "language_loss": 0.75542462,
      "learning_rate": 1.1114367480595292e-06,
      "loss": 0.77750683,
      "num_input_tokens_seen": 117577310,
      "step": 5465,
      "time_per_iteration": 2.824056625366211
    },
    {
      "auxiliary_loss_clip": 0.01181502,
      "auxiliary_loss_mlp": 0.01031414,
      "balance_loss_clip": 1.04287124,
      "balance_loss_mlp": 1.02302158,
      "epoch": 0.6572476402332712,
      "flos": 17529830352000.0,
      "grad_norm": 2.1440745380858806,
      "language_loss": 0.81787205,
      "learning_rate": 1.1107389447630086e-06,
      "loss": 0.84000123,
      "num_input_tokens_seen": 117596010,
      "step": 5466,
      "time_per_iteration": 2.779059648513794
    },
    {
      "auxiliary_loss_clip": 0.01171269,
      "auxiliary_loss_mlp": 0.01207485,
      "balance_loss_clip": 1.04280591,
      "balance_loss_mlp": 1.00107658,
      "epoch": 0.6573678831239103,
      "flos": 17014260487680.0,
      "grad_norm": 2.163160886368807,
      "language_loss": 0.78397167,
      "learning_rate": 1.1100412763666818e-06,
      "loss": 0.80775928,
      "num_input_tokens_seen": 117611270,
      "step": 5467,
      "time_per_iteration": 2.6152503490448
    },
    {
      "auxiliary_loss_clip": 0.01177328,
      "auxiliary_loss_mlp": 0.01020368,
      "balance_loss_clip": 1.04594398,
      "balance_loss_mlp": 1.01256585,
      "epoch": 0.6574881260145494,
      "flos": 23910078528000.0,
      "grad_norm": 1.5329934717760163,
      "language_loss": 0.80208409,
      "learning_rate": 1.1093437429763865e-06,
      "loss": 0.8240611,
      "num_input_tokens_seen": 117631535,
      "step": 5468,
      "time_per_iteration": 2.6709139347076416
    },
    {
      "auxiliary_loss_clip": 0.01173898,
      "auxiliary_loss_mlp": 0.01020773,
      "balance_loss_clip": 1.04860365,
      "balance_loss_mlp": 1.01370728,
      "epoch": 0.6576083689051885,
      "flos": 11218458504960.0,
      "grad_norm": 2.0417396979862152,
      "language_loss": 0.74089146,
      "learning_rate": 1.1086463446979361e-06,
      "loss": 0.76283818,
      "num_input_tokens_seen": 117649885,
      "step": 5469,
      "time_per_iteration": 2.58674693107605
    },
    {
      "auxiliary_loss_clip": 0.01177324,
      "auxiliary_loss_mlp": 0.01024116,
      "balance_loss_clip": 1.05052042,
      "balance_loss_mlp": 1.01651323,
      "epoch": 0.6577286117958275,
      "flos": 22455553190400.0,
      "grad_norm": 1.9739394442851723,
      "language_loss": 0.77587652,
      "learning_rate": 1.1079490816371277e-06,
      "loss": 0.7978909,
      "num_input_tokens_seen": 117669650,
      "step": 5470,
      "time_per_iteration": 2.619565963745117
    },
    {
      "auxiliary_loss_clip": 0.01173117,
      "auxiliary_loss_mlp": 0.01208067,
      "balance_loss_clip": 1.04554045,
      "balance_loss_mlp": 1.00097215,
      "epoch": 0.6578488546864667,
      "flos": 21872184405120.0,
      "grad_norm": 2.054280976135606,
      "language_loss": 0.74808514,
      "learning_rate": 1.1072519538997352e-06,
      "loss": 0.77189696,
      "num_input_tokens_seen": 117688790,
      "step": 5471,
      "time_per_iteration": 2.59972882270813
    },
    {
      "auxiliary_loss_clip": 0.01176894,
      "auxiliary_loss_mlp": 0.0102111,
      "balance_loss_clip": 1.04206347,
      "balance_loss_mlp": 1.01380801,
      "epoch": 0.6579690975771058,
      "flos": 23543753673600.0,
      "grad_norm": 4.611839066406241,
      "language_loss": 0.82299984,
      "learning_rate": 1.1065549615915095e-06,
      "loss": 0.84497982,
      "num_input_tokens_seen": 117708620,
      "step": 5472,
      "time_per_iteration": 2.649855852127075
    },
    {
      "auxiliary_loss_clip": 0.0117412,
      "auxiliary_loss_mlp": 0.01026285,
      "balance_loss_clip": 1.05068731,
      "balance_loss_mlp": 1.01830077,
      "epoch": 0.6580893404677448,
      "flos": 32743995730560.0,
      "grad_norm": 2.22254691897154,
      "language_loss": 0.78830016,
      "learning_rate": 1.105858104818187e-06,
      "loss": 0.81030422,
      "num_input_tokens_seen": 117729775,
      "step": 5473,
      "time_per_iteration": 2.876558780670166
    },
    {
      "auxiliary_loss_clip": 0.0117735,
      "auxiliary_loss_mlp": 0.01023298,
      "balance_loss_clip": 1.04837584,
      "balance_loss_mlp": 1.01494694,
      "epoch": 0.658209583358384,
      "flos": 15888138220800.0,
      "grad_norm": 2.9525563408339215,
      "language_loss": 0.75027871,
      "learning_rate": 1.105161383685478e-06,
      "loss": 0.77228522,
      "num_input_tokens_seen": 117746160,
      "step": 5474,
      "time_per_iteration": 2.656675100326538
    },
    {
      "auxiliary_loss_clip": 0.01088309,
      "auxiliary_loss_mlp": 0.01000757,
      "balance_loss_clip": 1.0154624,
      "balance_loss_mlp": 0.99997598,
      "epoch": 0.658329826249023,
      "flos": 62695902447360.0,
      "grad_norm": 0.7229733661105185,
      "language_loss": 0.56271571,
      "learning_rate": 1.1044647982990771e-06,
      "loss": 0.58360636,
      "num_input_tokens_seen": 117808045,
      "step": 5475,
      "time_per_iteration": 3.2132527828216553
    },
    {
      "auxiliary_loss_clip": 0.0117837,
      "auxiliary_loss_mlp": 0.01024811,
      "balance_loss_clip": 1.04634881,
      "balance_loss_mlp": 1.01684761,
      "epoch": 0.6584500691396621,
      "flos": 31722624501120.0,
      "grad_norm": 2.2675548842873905,
      "language_loss": 0.64624763,
      "learning_rate": 1.1037683487646536e-06,
      "loss": 0.66827941,
      "num_input_tokens_seen": 117828330,
      "step": 5476,
      "time_per_iteration": 2.787292003631592
    },
    {
      "auxiliary_loss_clip": 0.01175378,
      "auxiliary_loss_mlp": 0.01208129,
      "balance_loss_clip": 1.04817653,
      "balance_loss_mlp": 1.00089312,
      "epoch": 0.6585703120303013,
      "flos": 18406086635520.0,
      "grad_norm": 1.753171165915038,
      "language_loss": 0.77123404,
      "learning_rate": 1.1030720351878583e-06,
      "loss": 0.7950691,
      "num_input_tokens_seen": 117846450,
      "step": 5477,
      "time_per_iteration": 3.619826316833496
    },
    {
      "auxiliary_loss_clip": 0.01084077,
      "auxiliary_loss_mlp": 0.01000391,
      "balance_loss_clip": 1.01521873,
      "balance_loss_mlp": 0.99959201,
      "epoch": 0.6586905549209403,
      "flos": 58309880434560.0,
      "grad_norm": 0.803272666732254,
      "language_loss": 0.57660687,
      "learning_rate": 1.102375857674323e-06,
      "loss": 0.59745157,
      "num_input_tokens_seen": 117908365,
      "step": 5478,
      "time_per_iteration": 3.146594524383545
    },
    {
      "auxiliary_loss_clip": 0.01174079,
      "auxiliary_loss_mlp": 0.01022572,
      "balance_loss_clip": 1.04361379,
      "balance_loss_mlp": 1.01503515,
      "epoch": 0.6588107978115794,
      "flos": 22782627457920.0,
      "grad_norm": 2.8236579506175685,
      "language_loss": 0.90342402,
      "learning_rate": 1.1016798163296561e-06,
      "loss": 0.92539048,
      "num_input_tokens_seen": 117927565,
      "step": 5479,
      "time_per_iteration": 2.6883621215820312
    },
    {
      "auxiliary_loss_clip": 0.01174167,
      "auxiliary_loss_mlp": 0.01021283,
      "balance_loss_clip": 1.04627812,
      "balance_loss_mlp": 1.01345706,
      "epoch": 0.6589310407022185,
      "flos": 20667525050880.0,
      "grad_norm": 2.1857988811313405,
      "language_loss": 0.66315603,
      "learning_rate": 1.1009839112594471e-06,
      "loss": 0.68511045,
      "num_input_tokens_seen": 117945590,
      "step": 5480,
      "time_per_iteration": 2.566622495651245
    },
    {
      "auxiliary_loss_clip": 0.01175508,
      "auxiliary_loss_mlp": 0.01028233,
      "balance_loss_clip": 1.04735017,
      "balance_loss_mlp": 1.02025175,
      "epoch": 0.6590512835928576,
      "flos": 25630595055360.0,
      "grad_norm": 2.095172067689959,
      "language_loss": 0.72042257,
      "learning_rate": 1.1002881425692638e-06,
      "loss": 0.74246001,
      "num_input_tokens_seen": 117966020,
      "step": 5481,
      "time_per_iteration": 2.6473641395568848
    },
    {
      "auxiliary_loss_clip": 0.01167234,
      "auxiliary_loss_mlp": 0.01023395,
      "balance_loss_clip": 1.04457474,
      "balance_loss_mlp": 1.01569128,
      "epoch": 0.6591715264834966,
      "flos": 23726108044800.0,
      "grad_norm": 1.7057263809449548,
      "language_loss": 0.75521177,
      "learning_rate": 1.0995925103646532e-06,
      "loss": 0.77711809,
      "num_input_tokens_seen": 117984620,
      "step": 5482,
      "time_per_iteration": 3.491441011428833
    },
    {
      "auxiliary_loss_clip": 0.01175123,
      "auxiliary_loss_mlp": 0.0102366,
      "balance_loss_clip": 1.04580903,
      "balance_loss_mlp": 1.01581562,
      "epoch": 0.6592917693741358,
      "flos": 35773850822400.0,
      "grad_norm": 1.5569481326600858,
      "language_loss": 0.66963089,
      "learning_rate": 1.0988970147511437e-06,
      "loss": 0.69161868,
      "num_input_tokens_seen": 118006500,
      "step": 5483,
      "time_per_iteration": 2.7423934936523438
    },
    {
      "auxiliary_loss_clip": 0.01176833,
      "auxiliary_loss_mlp": 0.01028293,
      "balance_loss_clip": 1.04770303,
      "balance_loss_mlp": 1.02031744,
      "epoch": 0.6594120122647749,
      "flos": 21396834794880.0,
      "grad_norm": 2.328025030737524,
      "language_loss": 0.80619287,
      "learning_rate": 1.0982016558342405e-06,
      "loss": 0.82824415,
      "num_input_tokens_seen": 118025470,
      "step": 5484,
      "time_per_iteration": 3.6368956565856934
    },
    {
      "auxiliary_loss_clip": 0.01171841,
      "auxiliary_loss_mlp": 0.01025268,
      "balance_loss_clip": 1.04961729,
      "balance_loss_mlp": 1.01804399,
      "epoch": 0.6595322551554139,
      "flos": 19351829779200.0,
      "grad_norm": 3.443886981875417,
      "language_loss": 0.71332592,
      "learning_rate": 1.0975064337194291e-06,
      "loss": 0.73529702,
      "num_input_tokens_seen": 118043515,
      "step": 5485,
      "time_per_iteration": 2.6089529991149902
    },
    {
      "auxiliary_loss_clip": 0.01170947,
      "auxiliary_loss_mlp": 0.01034574,
      "balance_loss_clip": 1.04205859,
      "balance_loss_mlp": 1.02711415,
      "epoch": 0.6596524980460531,
      "flos": 16837113588480.0,
      "grad_norm": 1.576818863725046,
      "language_loss": 0.70537513,
      "learning_rate": 1.0968113485121743e-06,
      "loss": 0.72743034,
      "num_input_tokens_seen": 118063105,
      "step": 5486,
      "time_per_iteration": 2.707404851913452
    },
    {
      "auxiliary_loss_clip": 0.01174477,
      "auxiliary_loss_mlp": 0.01208329,
      "balance_loss_clip": 1.04532099,
      "balance_loss_mlp": 1.0009743,
      "epoch": 0.6597727409366921,
      "flos": 21798567480960.0,
      "grad_norm": 1.9063090144046286,
      "language_loss": 0.79998326,
      "learning_rate": 1.0961164003179185e-06,
      "loss": 0.82381129,
      "num_input_tokens_seen": 118081615,
      "step": 5487,
      "time_per_iteration": 2.725680351257324
    },
    {
      "auxiliary_loss_clip": 0.01177223,
      "auxiliary_loss_mlp": 0.01024965,
      "balance_loss_clip": 1.04347479,
      "balance_loss_mlp": 1.01710892,
      "epoch": 0.6598929838273312,
      "flos": 23730704985600.0,
      "grad_norm": 2.17532721492987,
      "language_loss": 0.84257591,
      "learning_rate": 1.0954215892420884e-06,
      "loss": 0.8645978,
      "num_input_tokens_seen": 118102315,
      "step": 5488,
      "time_per_iteration": 3.5761914253234863
    },
    {
      "auxiliary_loss_clip": 0.01184699,
      "auxiliary_loss_mlp": 0.01035158,
      "balance_loss_clip": 1.04719186,
      "balance_loss_mlp": 1.02661109,
      "epoch": 0.6600132267179702,
      "flos": 19974520978560.0,
      "grad_norm": 2.0247389074859194,
      "language_loss": 0.70829678,
      "learning_rate": 1.094726915390082e-06,
      "loss": 0.73049533,
      "num_input_tokens_seen": 118120650,
      "step": 5489,
      "time_per_iteration": 2.7242202758789062
    },
    {
      "auxiliary_loss_clip": 0.01175098,
      "auxiliary_loss_mlp": 0.01024882,
      "balance_loss_clip": 1.04759574,
      "balance_loss_mlp": 1.01724052,
      "epoch": 0.6601334696086094,
      "flos": 22342649765760.0,
      "grad_norm": 1.979567637522351,
      "language_loss": 0.69646782,
      "learning_rate": 1.0940323788672836e-06,
      "loss": 0.71846759,
      "num_input_tokens_seen": 118139825,
      "step": 5490,
      "time_per_iteration": 2.589648962020874
    },
    {
      "auxiliary_loss_clip": 0.01169362,
      "auxiliary_loss_mlp": 0.01022404,
      "balance_loss_clip": 1.04704857,
      "balance_loss_mlp": 1.01436901,
      "epoch": 0.6602537124992485,
      "flos": 25703098657920.0,
      "grad_norm": 2.1259121237120686,
      "language_loss": 0.73577505,
      "learning_rate": 1.0933379797790522e-06,
      "loss": 0.75769269,
      "num_input_tokens_seen": 118159240,
      "step": 5491,
      "time_per_iteration": 2.6428658962249756
    },
    {
      "auxiliary_loss_clip": 0.0117276,
      "auxiliary_loss_mlp": 0.01026062,
      "balance_loss_clip": 1.05018497,
      "balance_loss_mlp": 1.0178901,
      "epoch": 0.6603739553898875,
      "flos": 25848572739840.0,
      "grad_norm": 2.626315353891208,
      "language_loss": 0.71994799,
      "learning_rate": 1.0926437182307293e-06,
      "loss": 0.74193621,
      "num_input_tokens_seen": 118178050,
      "step": 5492,
      "time_per_iteration": 2.6186554431915283
    },
    {
      "auxiliary_loss_clip": 0.01179714,
      "auxiliary_loss_mlp": 0.01025867,
      "balance_loss_clip": 1.04483795,
      "balance_loss_mlp": 1.0181843,
      "epoch": 0.6604941982805267,
      "flos": 24570296461440.0,
      "grad_norm": 1.8075961663065612,
      "language_loss": 0.77659237,
      "learning_rate": 1.0919495943276338e-06,
      "loss": 0.79864818,
      "num_input_tokens_seen": 118199070,
      "step": 5493,
      "time_per_iteration": 2.6665306091308594
    },
    {
      "auxiliary_loss_clip": 0.01181023,
      "auxiliary_loss_mlp": 0.0102522,
      "balance_loss_clip": 1.04115355,
      "balance_loss_mlp": 1.01690531,
      "epoch": 0.6606144411711657,
      "flos": 13261775581440.0,
      "grad_norm": 2.246143252118766,
      "language_loss": 0.76511973,
      "learning_rate": 1.0912556081750611e-06,
      "loss": 0.78718221,
      "num_input_tokens_seen": 118217000,
      "step": 5494,
      "time_per_iteration": 2.666192054748535
    },
    {
      "auxiliary_loss_clip": 0.01176109,
      "auxiliary_loss_mlp": 0.01024762,
      "balance_loss_clip": 1.04822314,
      "balance_loss_mlp": 1.01755571,
      "epoch": 0.6607346840618048,
      "flos": 25155281358720.0,
      "grad_norm": 8.667335115329108,
      "language_loss": 0.76767886,
      "learning_rate": 1.0905617598782909e-06,
      "loss": 0.78968751,
      "num_input_tokens_seen": 118237205,
      "step": 5495,
      "time_per_iteration": 2.6645233631134033
    },
    {
      "auxiliary_loss_clip": 0.01174153,
      "auxiliary_loss_mlp": 0.0102977,
      "balance_loss_clip": 1.04200947,
      "balance_loss_mlp": 1.02236116,
      "epoch": 0.660854926952444,
      "flos": 17638029095040.0,
      "grad_norm": 2.115002436110331,
      "language_loss": 0.81700528,
      "learning_rate": 1.0898680495425775e-06,
      "loss": 0.83904445,
      "num_input_tokens_seen": 118255495,
      "step": 5496,
      "time_per_iteration": 2.8009731769561768
    },
    {
      "auxiliary_loss_clip": 0.01181237,
      "auxiliary_loss_mlp": 0.01030875,
      "balance_loss_clip": 1.04838443,
      "balance_loss_mlp": 1.02333212,
      "epoch": 0.660975169843083,
      "flos": 16836000266880.0,
      "grad_norm": 2.1756411806215894,
      "language_loss": 0.80547583,
      "learning_rate": 1.0891744772731594e-06,
      "loss": 0.82759696,
      "num_input_tokens_seen": 118273310,
      "step": 5497,
      "time_per_iteration": 2.616798162460327
    },
    {
      "auxiliary_loss_clip": 0.01174909,
      "auxiliary_loss_mlp": 0.01029857,
      "balance_loss_clip": 1.04688096,
      "balance_loss_mlp": 1.02260947,
      "epoch": 0.6610954127337221,
      "flos": 26870410846080.0,
      "grad_norm": 1.7313306996541185,
      "language_loss": 0.66247147,
      "learning_rate": 1.088481043175248e-06,
      "loss": 0.68451917,
      "num_input_tokens_seen": 118293880,
      "step": 5498,
      "time_per_iteration": 2.6858773231506348
    },
    {
      "auxiliary_loss_clip": 0.01162969,
      "auxiliary_loss_mlp": 0.01025212,
      "balance_loss_clip": 1.040362,
      "balance_loss_mlp": 1.01726699,
      "epoch": 0.6612156556243612,
      "flos": 26465697331200.0,
      "grad_norm": 1.9191445231081887,
      "language_loss": 0.75932705,
      "learning_rate": 1.0877877473540368e-06,
      "loss": 0.78120887,
      "num_input_tokens_seen": 118314465,
      "step": 5499,
      "time_per_iteration": 2.724864959716797
    },
    {
      "auxiliary_loss_clip": 0.01170413,
      "auxiliary_loss_mlp": 0.01024104,
      "balance_loss_clip": 1.04703188,
      "balance_loss_mlp": 1.01656449,
      "epoch": 0.6613358985150003,
      "flos": 19791915212160.0,
      "grad_norm": 1.9679082779302024,
      "language_loss": 0.72589505,
      "learning_rate": 1.0870945899147002e-06,
      "loss": 0.74784017,
      "num_input_tokens_seen": 118331110,
      "step": 5500,
      "time_per_iteration": 2.6547658443450928
    },
    {
      "auxiliary_loss_clip": 0.01172891,
      "auxiliary_loss_mlp": 0.01026482,
      "balance_loss_clip": 1.04887605,
      "balance_loss_mlp": 1.01933527,
      "epoch": 0.6614561414056394,
      "flos": 26831627136000.0,
      "grad_norm": 1.8749281416765409,
      "language_loss": 0.76418084,
      "learning_rate": 1.0864015709623879e-06,
      "loss": 0.78617454,
      "num_input_tokens_seen": 118351980,
      "step": 5501,
      "time_per_iteration": 2.734714984893799
    },
    {
      "auxiliary_loss_clip": 0.0117697,
      "auxiliary_loss_mlp": 0.01025083,
      "balance_loss_clip": 1.04704261,
      "balance_loss_mlp": 1.0178169,
      "epoch": 0.6615763842962785,
      "flos": 22894597128960.0,
      "grad_norm": 2.2896620863106807,
      "language_loss": 0.80031627,
      "learning_rate": 1.0857086906022313e-06,
      "loss": 0.82233679,
      "num_input_tokens_seen": 118370315,
      "step": 5502,
      "time_per_iteration": 2.781888723373413
    },
    {
      "auxiliary_loss_clip": 0.01171158,
      "auxiliary_loss_mlp": 0.01023216,
      "balance_loss_clip": 1.04201531,
      "balance_loss_mlp": 1.01531243,
      "epoch": 0.6616966271869176,
      "flos": 24790321221120.0,
      "grad_norm": 1.862373426015124,
      "language_loss": 0.72887295,
      "learning_rate": 1.0850159489393388e-06,
      "loss": 0.75081664,
      "num_input_tokens_seen": 118389575,
      "step": 5503,
      "time_per_iteration": 2.7074718475341797
    },
    {
      "auxiliary_loss_clip": 0.01169623,
      "auxiliary_loss_mlp": 0.01024416,
      "balance_loss_clip": 1.03971648,
      "balance_loss_mlp": 1.0166496,
      "epoch": 0.6618168700775566,
      "flos": 17202109639680.0,
      "grad_norm": 2.0786979938659704,
      "language_loss": 0.82722199,
      "learning_rate": 1.0843233460787992e-06,
      "loss": 0.84916234,
      "num_input_tokens_seen": 118406790,
      "step": 5504,
      "time_per_iteration": 3.6367993354797363
    },
    {
      "auxiliary_loss_clip": 0.01165814,
      "auxiliary_loss_mlp": 0.01025959,
      "balance_loss_clip": 1.04510653,
      "balance_loss_mlp": 1.01838887,
      "epoch": 0.6619371129681958,
      "flos": 25447091448960.0,
      "grad_norm": 2.0565746434742547,
      "language_loss": 0.78047526,
      "learning_rate": 1.0836308821256805e-06,
      "loss": 0.80239296,
      "num_input_tokens_seen": 118427590,
      "step": 5505,
      "time_per_iteration": 2.7664425373077393
    },
    {
      "auxiliary_loss_clip": 0.01173158,
      "auxiliary_loss_mlp": 0.01026144,
      "balance_loss_clip": 1.04739094,
      "balance_loss_mlp": 1.0186162,
      "epoch": 0.6620573558588349,
      "flos": 18040444139520.0,
      "grad_norm": 2.952674384551895,
      "language_loss": 0.78271312,
      "learning_rate": 1.0829385571850282e-06,
      "loss": 0.8047061,
      "num_input_tokens_seen": 118444570,
      "step": 5506,
      "time_per_iteration": 2.5662078857421875
    },
    {
      "auxiliary_loss_clip": 0.01177125,
      "auxiliary_loss_mlp": 0.01023705,
      "balance_loss_clip": 1.05061877,
      "balance_loss_mlp": 1.01547384,
      "epoch": 0.6621775987494739,
      "flos": 17785586165760.0,
      "grad_norm": 2.345788261935345,
      "language_loss": 0.83406019,
      "learning_rate": 1.0822463713618679e-06,
      "loss": 0.85606849,
      "num_input_tokens_seen": 118461425,
      "step": 5507,
      "time_per_iteration": 2.607344388961792
    },
    {
      "auxiliary_loss_clip": 0.01179935,
      "auxiliary_loss_mlp": 0.01024863,
      "balance_loss_clip": 1.04457688,
      "balance_loss_mlp": 1.01740026,
      "epoch": 0.6622978416401131,
      "flos": 17492590926720.0,
      "grad_norm": 1.999536222554496,
      "language_loss": 0.85057092,
      "learning_rate": 1.0815543247612034e-06,
      "loss": 0.87261891,
      "num_input_tokens_seen": 118478495,
      "step": 5508,
      "time_per_iteration": 3.722590446472168
    },
    {
      "auxiliary_loss_clip": 0.01172741,
      "auxiliary_loss_mlp": 0.01022405,
      "balance_loss_clip": 1.04076517,
      "balance_loss_mlp": 1.01452529,
      "epoch": 0.6624180845307521,
      "flos": 21648352803840.0,
      "grad_norm": 1.7805059754151689,
      "language_loss": 0.83079112,
      "learning_rate": 1.0808624174880168e-06,
      "loss": 0.85274255,
      "num_input_tokens_seen": 118499145,
      "step": 5509,
      "time_per_iteration": 2.712170362472534
    },
    {
      "auxiliary_loss_clip": 0.01168027,
      "auxiliary_loss_mlp": 0.01020334,
      "balance_loss_clip": 1.04819226,
      "balance_loss_mlp": 1.01314902,
      "epoch": 0.6625383274213912,
      "flos": 23805902108160.0,
      "grad_norm": 2.114907027071194,
      "language_loss": 0.80085957,
      "learning_rate": 1.080170649647272e-06,
      "loss": 0.82274318,
      "num_input_tokens_seen": 118518950,
      "step": 5510,
      "time_per_iteration": 2.5813560485839844
    },
    {
      "auxiliary_loss_clip": 0.01169044,
      "auxiliary_loss_mlp": 0.01023282,
      "balance_loss_clip": 1.04762888,
      "balance_loss_mlp": 1.01535463,
      "epoch": 0.6626585703120303,
      "flos": 33262941473280.0,
      "grad_norm": 1.790962521666516,
      "language_loss": 0.67214453,
      "learning_rate": 1.0794790213439068e-06,
      "loss": 0.69406778,
      "num_input_tokens_seen": 118545850,
      "step": 5511,
      "time_per_iteration": 3.721050500869751
    },
    {
      "auxiliary_loss_clip": 0.01185617,
      "auxiliary_loss_mlp": 0.01028674,
      "balance_loss_clip": 1.04511809,
      "balance_loss_mlp": 1.02029991,
      "epoch": 0.6627788132026694,
      "flos": 22085780630400.0,
      "grad_norm": 2.0221914505443253,
      "language_loss": 0.78761715,
      "learning_rate": 1.078787532682843e-06,
      "loss": 0.80976009,
      "num_input_tokens_seen": 118563325,
      "step": 5512,
      "time_per_iteration": 2.697680950164795
    },
    {
      "auxiliary_loss_clip": 0.01168228,
      "auxiliary_loss_mlp": 0.01026241,
      "balance_loss_clip": 1.04537165,
      "balance_loss_mlp": 1.01885593,
      "epoch": 0.6628990560933085,
      "flos": 36173608260480.0,
      "grad_norm": 2.018967543708999,
      "language_loss": 0.75771236,
      "learning_rate": 1.0780961837689773e-06,
      "loss": 0.77965713,
      "num_input_tokens_seen": 118582835,
      "step": 5513,
      "time_per_iteration": 2.6969738006591797
    },
    {
      "auxiliary_loss_clip": 0.01169849,
      "auxiliary_loss_mlp": 0.01024588,
      "balance_loss_clip": 1.04596043,
      "balance_loss_mlp": 1.01719451,
      "epoch": 0.6630192989839476,
      "flos": 18513567106560.0,
      "grad_norm": 1.7290050994677695,
      "language_loss": 0.70043671,
      "learning_rate": 1.0774049747071883e-06,
      "loss": 0.72238111,
      "num_input_tokens_seen": 118600715,
      "step": 5514,
      "time_per_iteration": 2.770678758621216
    },
    {
      "auxiliary_loss_clip": 0.01176958,
      "auxiliary_loss_mlp": 0.01031811,
      "balance_loss_clip": 1.04304492,
      "balance_loss_mlp": 1.02345157,
      "epoch": 0.6631395418745867,
      "flos": 35809510049280.0,
      "grad_norm": 11.743171008728659,
      "language_loss": 0.68546379,
      "learning_rate": 1.076713905602332e-06,
      "loss": 0.70755148,
      "num_input_tokens_seen": 118621290,
      "step": 5515,
      "time_per_iteration": 3.7522220611572266
    },
    {
      "auxiliary_loss_clip": 0.01175923,
      "auxiliary_loss_mlp": 0.0102228,
      "balance_loss_clip": 1.04879773,
      "balance_loss_mlp": 1.01502037,
      "epoch": 0.6632597847652257,
      "flos": 20047742853120.0,
      "grad_norm": 1.909438935282953,
      "language_loss": 0.81280017,
      "learning_rate": 1.07602297655924e-06,
      "loss": 0.83478224,
      "num_input_tokens_seen": 118639610,
      "step": 5516,
      "time_per_iteration": 2.6099421977996826
    },
    {
      "auxiliary_loss_clip": 0.01171122,
      "auxiliary_loss_mlp": 0.01025305,
      "balance_loss_clip": 1.04933918,
      "balance_loss_mlp": 1.0179317,
      "epoch": 0.6633800276558649,
      "flos": 21214480423680.0,
      "grad_norm": 3.427066886748327,
      "language_loss": 0.81304383,
      "learning_rate": 1.0753321876827292e-06,
      "loss": 0.83500814,
      "num_input_tokens_seen": 118658895,
      "step": 5517,
      "time_per_iteration": 2.585496425628662
    },
    {
      "auxiliary_loss_clip": 0.0116868,
      "auxiliary_loss_mlp": 0.01023844,
      "balance_loss_clip": 1.04568791,
      "balance_loss_mlp": 1.01605356,
      "epoch": 0.663500270546504,
      "flos": 23987753688960.0,
      "grad_norm": 1.9982635628127854,
      "language_loss": 0.74514443,
      "learning_rate": 1.0746415390775893e-06,
      "loss": 0.76706964,
      "num_input_tokens_seen": 118677025,
      "step": 5518,
      "time_per_iteration": 2.5551533699035645
    },
    {
      "auxiliary_loss_clip": 0.01169462,
      "auxiliary_loss_mlp": 0.01025764,
      "balance_loss_clip": 1.04935658,
      "balance_loss_mlp": 1.0185101,
      "epoch": 0.663620513437143,
      "flos": 17932389050880.0,
      "grad_norm": 1.9475239563896174,
      "language_loss": 0.76477724,
      "learning_rate": 1.0739510308485939e-06,
      "loss": 0.78672951,
      "num_input_tokens_seen": 118694240,
      "step": 5519,
      "time_per_iteration": 2.52020525932312
    },
    {
      "auxiliary_loss_clip": 0.01092378,
      "auxiliary_loss_mlp": 0.01000639,
      "balance_loss_clip": 1.01509643,
      "balance_loss_mlp": 0.99984664,
      "epoch": 0.6637407563277821,
      "flos": 57840241086720.0,
      "grad_norm": 0.803915921373363,
      "language_loss": 0.62496507,
      "learning_rate": 1.07326066310049e-06,
      "loss": 0.64589524,
      "num_input_tokens_seen": 118758365,
      "step": 5520,
      "time_per_iteration": 3.1562387943267822
    },
    {
      "auxiliary_loss_clip": 0.01174171,
      "auxiliary_loss_mlp": 0.0102622,
      "balance_loss_clip": 1.04188061,
      "balance_loss_mlp": 1.01754761,
      "epoch": 0.6638609992184212,
      "flos": 27306007079040.0,
      "grad_norm": 2.30029130601476,
      "language_loss": 0.79790294,
      "learning_rate": 1.0725704359380059e-06,
      "loss": 0.81990683,
      "num_input_tokens_seen": 118778220,
      "step": 5521,
      "time_per_iteration": 2.700929880142212
    },
    {
      "auxiliary_loss_clip": 0.01169928,
      "auxiliary_loss_mlp": 0.01022218,
      "balance_loss_clip": 1.04801011,
      "balance_loss_mlp": 1.01521111,
      "epoch": 0.6639812421090603,
      "flos": 18624854419200.0,
      "grad_norm": 2.024718160792922,
      "language_loss": 0.72023314,
      "learning_rate": 1.0718803494658497e-06,
      "loss": 0.7421546,
      "num_input_tokens_seen": 118797110,
      "step": 5522,
      "time_per_iteration": 2.5237350463867188
    },
    {
      "auxiliary_loss_clip": 0.01181481,
      "auxiliary_loss_mlp": 0.0103039,
      "balance_loss_clip": 1.03832126,
      "balance_loss_mlp": 1.02220011,
      "epoch": 0.6641014849996993,
      "flos": 15924479806080.0,
      "grad_norm": 2.401401731089482,
      "language_loss": 0.83961666,
      "learning_rate": 1.071190403788707e-06,
      "loss": 0.86173534,
      "num_input_tokens_seen": 118812415,
      "step": 5523,
      "time_per_iteration": 2.8364975452423096
    },
    {
      "auxiliary_loss_clip": 0.01187731,
      "auxiliary_loss_mlp": 0.01026712,
      "balance_loss_clip": 1.04855895,
      "balance_loss_mlp": 1.01868892,
      "epoch": 0.6642217278903385,
      "flos": 26505486622080.0,
      "grad_norm": 1.8326370966638859,
      "language_loss": 0.76060796,
      "learning_rate": 1.0705005990112415e-06,
      "loss": 0.78275239,
      "num_input_tokens_seen": 118832195,
      "step": 5524,
      "time_per_iteration": 2.713728904724121
    },
    {
      "auxiliary_loss_clip": 0.01168591,
      "auxiliary_loss_mlp": 0.01030309,
      "balance_loss_clip": 1.04188037,
      "balance_loss_mlp": 1.02266741,
      "epoch": 0.6643419707809776,
      "flos": 15377308951680.0,
      "grad_norm": 2.333683952183928,
      "language_loss": 0.74417996,
      "learning_rate": 1.0698109352380957e-06,
      "loss": 0.76616895,
      "num_input_tokens_seen": 118849795,
      "step": 5525,
      "time_per_iteration": 2.6569225788116455
    },
    {
      "auxiliary_loss_clip": 0.01168829,
      "auxiliary_loss_mlp": 0.01025705,
      "balance_loss_clip": 1.04781103,
      "balance_loss_mlp": 1.01854348,
      "epoch": 0.6644622136716166,
      "flos": 25117610970240.0,
      "grad_norm": 1.7404795083026303,
      "language_loss": 0.77976751,
      "learning_rate": 1.0691214125738909e-06,
      "loss": 0.80171287,
      "num_input_tokens_seen": 118870000,
      "step": 5526,
      "time_per_iteration": 2.6506104469299316
    },
    {
      "auxiliary_loss_clip": 0.01069558,
      "auxiliary_loss_mlp": 0.01000289,
      "balance_loss_clip": 1.01482129,
      "balance_loss_mlp": 0.99946636,
      "epoch": 0.6645824565622558,
      "flos": 66201717680640.0,
      "grad_norm": 0.7870574714020449,
      "language_loss": 0.57539469,
      "learning_rate": 1.0684320311232287e-06,
      "loss": 0.59609312,
      "num_input_tokens_seen": 118932905,
      "step": 5527,
      "time_per_iteration": 3.1924211978912354
    },
    {
      "auxiliary_loss_clip": 0.01172581,
      "auxiliary_loss_mlp": 0.01021637,
      "balance_loss_clip": 1.04408395,
      "balance_loss_mlp": 1.01397789,
      "epoch": 0.6647026994528948,
      "flos": 25082131311360.0,
      "grad_norm": 2.2002598759538388,
      "language_loss": 0.81737024,
      "learning_rate": 1.0677427909906865e-06,
      "loss": 0.83931243,
      "num_input_tokens_seen": 118953355,
      "step": 5528,
      "time_per_iteration": 2.7062788009643555
    },
    {
      "auxiliary_loss_clip": 0.01174304,
      "auxiliary_loss_mlp": 0.01032514,
      "balance_loss_clip": 1.04997361,
      "balance_loss_mlp": 1.02408576,
      "epoch": 0.6648229423435339,
      "flos": 18222187979520.0,
      "grad_norm": 1.9201069926474945,
      "language_loss": 0.72062182,
      "learning_rate": 1.0670536922808216e-06,
      "loss": 0.74268997,
      "num_input_tokens_seen": 118973480,
      "step": 5529,
      "time_per_iteration": 2.6055068969726562
    },
    {
      "auxiliary_loss_clip": 0.01176177,
      "auxiliary_loss_mlp": 0.01021434,
      "balance_loss_clip": 1.04665017,
      "balance_loss_mlp": 1.01422226,
      "epoch": 0.6649431852341731,
      "flos": 18296882311680.0,
      "grad_norm": 2.2168213378427146,
      "language_loss": 0.72011703,
      "learning_rate": 1.06636473509817e-06,
      "loss": 0.74209321,
      "num_input_tokens_seen": 118989860,
      "step": 5530,
      "time_per_iteration": 3.709803581237793
    },
    {
      "auxiliary_loss_clip": 0.01171281,
      "auxiliary_loss_mlp": 0.01208528,
      "balance_loss_clip": 1.04463172,
      "balance_loss_mlp": 1.00079632,
      "epoch": 0.6650634281248121,
      "flos": 17019575700480.0,
      "grad_norm": 2.4854927738443946,
      "language_loss": 0.81195062,
      "learning_rate": 1.0656759195472447e-06,
      "loss": 0.83574873,
      "num_input_tokens_seen": 119007150,
      "step": 5531,
      "time_per_iteration": 2.689937114715576
    },
    {
      "auxiliary_loss_clip": 0.01081077,
      "auxiliary_loss_mlp": 0.01000951,
      "balance_loss_clip": 1.01540911,
      "balance_loss_mlp": 1.00014007,
      "epoch": 0.6651836710154512,
      "flos": 69294810666240.0,
      "grad_norm": 0.7706288242684968,
      "language_loss": 0.59714174,
      "learning_rate": 1.0649872457325414e-06,
      "loss": 0.61796206,
      "num_input_tokens_seen": 119068435,
      "step": 5532,
      "time_per_iteration": 3.169297933578491
    },
    {
      "auxiliary_loss_clip": 0.01076621,
      "auxiliary_loss_mlp": 0.01000781,
      "balance_loss_clip": 1.01404393,
      "balance_loss_mlp": 1.00001192,
      "epoch": 0.6653039139060903,
      "flos": 66883444882560.0,
      "grad_norm": 0.8418059842277705,
      "language_loss": 0.5506351,
      "learning_rate": 1.0642987137585278e-06,
      "loss": 0.57140911,
      "num_input_tokens_seen": 119127960,
      "step": 5533,
      "time_per_iteration": 3.2009925842285156
    },
    {
      "auxiliary_loss_clip": 0.01173823,
      "auxiliary_loss_mlp": 0.01024386,
      "balance_loss_clip": 1.04475856,
      "balance_loss_mlp": 1.01720095,
      "epoch": 0.6654241567967294,
      "flos": 21470056669440.0,
      "grad_norm": 2.9091758642076098,
      "language_loss": 0.82545877,
      "learning_rate": 1.0636103237296561e-06,
      "loss": 0.8474409,
      "num_input_tokens_seen": 119146885,
      "step": 5534,
      "time_per_iteration": 2.610154867172241
    },
    {
      "auxiliary_loss_clip": 0.01173102,
      "auxiliary_loss_mlp": 0.01025386,
      "balance_loss_clip": 1.05040121,
      "balance_loss_mlp": 1.01840615,
      "epoch": 0.6655443996873684,
      "flos": 25119514391040.0,
      "grad_norm": 3.032794433248202,
      "language_loss": 0.84638715,
      "learning_rate": 1.062922075750353e-06,
      "loss": 0.86837208,
      "num_input_tokens_seen": 119166900,
      "step": 5535,
      "time_per_iteration": 3.590972423553467
    },
    {
      "auxiliary_loss_clip": 0.0117923,
      "auxiliary_loss_mlp": 0.01020221,
      "balance_loss_clip": 1.04455376,
      "balance_loss_mlp": 1.01287735,
      "epoch": 0.6656646425780076,
      "flos": 17457326749440.0,
      "grad_norm": 2.1379045412273565,
      "language_loss": 0.72161973,
      "learning_rate": 1.0622339699250267e-06,
      "loss": 0.7436142,
      "num_input_tokens_seen": 119184820,
      "step": 5536,
      "time_per_iteration": 2.7720158100128174
    },
    {
      "auxiliary_loss_clip": 0.01176911,
      "auxiliary_loss_mlp": 0.01022048,
      "balance_loss_clip": 1.04277229,
      "balance_loss_mlp": 1.01457369,
      "epoch": 0.6657848854686467,
      "flos": 23434190213760.0,
      "grad_norm": 3.076207496656182,
      "language_loss": 0.79665691,
      "learning_rate": 1.0615460063580624e-06,
      "loss": 0.81864649,
      "num_input_tokens_seen": 119203295,
      "step": 5537,
      "time_per_iteration": 2.7390856742858887
    },
    {
      "auxiliary_loss_clip": 0.01179016,
      "auxiliary_loss_mlp": 0.01023425,
      "balance_loss_clip": 1.04632902,
      "balance_loss_mlp": 1.01632881,
      "epoch": 0.6659051283592857,
      "flos": 11509909459200.0,
      "grad_norm": 1.8128764588236221,
      "language_loss": 0.73036796,
      "learning_rate": 1.060858185153821e-06,
      "loss": 0.75239235,
      "num_input_tokens_seen": 119221395,
      "step": 5538,
      "time_per_iteration": 3.5662155151367188
    },
    {
      "auxiliary_loss_clip": 0.01182466,
      "auxiliary_loss_mlp": 0.01026808,
      "balance_loss_clip": 1.04659295,
      "balance_loss_mlp": 1.01899433,
      "epoch": 0.6660253712499249,
      "flos": 20594554571520.0,
      "grad_norm": 2.289935701371409,
      "language_loss": 0.75819552,
      "learning_rate": 1.0601705064166474e-06,
      "loss": 0.78028822,
      "num_input_tokens_seen": 119239790,
      "step": 5539,
      "time_per_iteration": 2.6175761222839355
    },
    {
      "auxiliary_loss_clip": 0.01167487,
      "auxiliary_loss_mlp": 0.01027099,
      "balance_loss_clip": 1.04494929,
      "balance_loss_mlp": 1.01951766,
      "epoch": 0.666145614140564,
      "flos": 21251504367360.0,
      "grad_norm": 10.543669983215716,
      "language_loss": 0.73576641,
      "learning_rate": 1.0594829702508596e-06,
      "loss": 0.75771224,
      "num_input_tokens_seen": 119257505,
      "step": 5540,
      "time_per_iteration": 2.620797872543335
    },
    {
      "auxiliary_loss_clip": 0.01180063,
      "auxiliary_loss_mlp": 0.01022637,
      "balance_loss_clip": 1.0432744,
      "balance_loss_mlp": 1.01554155,
      "epoch": 0.666265857031203,
      "flos": 33726188200320.0,
      "grad_norm": 2.1085339017455604,
      "language_loss": 0.55591393,
      "learning_rate": 1.0587955767607592e-06,
      "loss": 0.57794088,
      "num_input_tokens_seen": 119279365,
      "step": 5541,
      "time_per_iteration": 2.754101514816284
    },
    {
      "auxiliary_loss_clip": 0.01170813,
      "auxiliary_loss_mlp": 0.01025297,
      "balance_loss_clip": 1.04886842,
      "balance_loss_mlp": 1.01761973,
      "epoch": 0.6663860999218422,
      "flos": 17456644391040.0,
      "grad_norm": 3.3755405582615747,
      "language_loss": 0.7736153,
      "learning_rate": 1.0581083260506206e-06,
      "loss": 0.79557645,
      "num_input_tokens_seen": 119296150,
      "step": 5542,
      "time_per_iteration": 3.4521520137786865
    },
    {
      "auxiliary_loss_clip": 0.01173937,
      "auxiliary_loss_mlp": 0.01022934,
      "balance_loss_clip": 1.04474354,
      "balance_loss_mlp": 1.01567125,
      "epoch": 0.6665063428124812,
      "flos": 17676740977920.0,
      "grad_norm": 2.0658613467229445,
      "language_loss": 0.76510465,
      "learning_rate": 1.0574212182246993e-06,
      "loss": 0.78707337,
      "num_input_tokens_seen": 119314845,
      "step": 5543,
      "time_per_iteration": 2.652188301086426
    },
    {
      "auxiliary_loss_clip": 0.01178558,
      "auxiliary_loss_mlp": 0.01021864,
      "balance_loss_clip": 1.04383945,
      "balance_loss_mlp": 1.01384664,
      "epoch": 0.6666265857031203,
      "flos": 27673265687040.0,
      "grad_norm": 2.5320661956195667,
      "language_loss": 0.76036179,
      "learning_rate": 1.0567342533872303e-06,
      "loss": 0.78236604,
      "num_input_tokens_seen": 119334875,
      "step": 5544,
      "time_per_iteration": 2.6577939987182617
    },
    {
      "auxiliary_loss_clip": 0.01178125,
      "auxiliary_loss_mlp": 0.01025652,
      "balance_loss_clip": 1.04737353,
      "balance_loss_mlp": 1.0178678,
      "epoch": 0.6667468285937594,
      "flos": 25046831220480.0,
      "grad_norm": 1.6354798700881759,
      "language_loss": 0.80775082,
      "learning_rate": 1.0560474316424255e-06,
      "loss": 0.82978868,
      "num_input_tokens_seen": 119354635,
      "step": 5545,
      "time_per_iteration": 2.652965784072876
    },
    {
      "auxiliary_loss_clip": 0.01175616,
      "auxiliary_loss_mlp": 0.01027977,
      "balance_loss_clip": 1.04399085,
      "balance_loss_mlp": 1.01956677,
      "epoch": 0.6668670714843985,
      "flos": 22780472641920.0,
      "grad_norm": 2.5579637056095526,
      "language_loss": 0.7364127,
      "learning_rate": 1.0553607530944746e-06,
      "loss": 0.75844866,
      "num_input_tokens_seen": 119372690,
      "step": 5546,
      "time_per_iteration": 2.6233458518981934
    },
    {
      "auxiliary_loss_clip": 0.01175776,
      "auxiliary_loss_mlp": 0.01023516,
      "balance_loss_clip": 1.04143476,
      "balance_loss_mlp": 1.01590991,
      "epoch": 0.6669873143750376,
      "flos": 22163886754560.0,
      "grad_norm": 1.8984746869431182,
      "language_loss": 0.89579046,
      "learning_rate": 1.0546742178475463e-06,
      "loss": 0.91778338,
      "num_input_tokens_seen": 119391685,
      "step": 5547,
      "time_per_iteration": 2.7013208866119385
    },
    {
      "auxiliary_loss_clip": 0.01184965,
      "auxiliary_loss_mlp": 0.01020733,
      "balance_loss_clip": 1.04475641,
      "balance_loss_mlp": 1.01372981,
      "epoch": 0.6671075572656767,
      "flos": 20514832335360.0,
      "grad_norm": 1.9256305144782477,
      "language_loss": 0.86666548,
      "learning_rate": 1.0539878260057868e-06,
      "loss": 0.88872242,
      "num_input_tokens_seen": 119410725,
      "step": 5548,
      "time_per_iteration": 2.6730968952178955
    },
    {
      "auxiliary_loss_clip": 0.01174526,
      "auxiliary_loss_mlp": 0.01021856,
      "balance_loss_clip": 1.04983532,
      "balance_loss_mlp": 1.01371443,
      "epoch": 0.6672278001563158,
      "flos": 17931203902080.0,
      "grad_norm": 2.680185374709829,
      "language_loss": 0.68564665,
      "learning_rate": 1.0533015776733226e-06,
      "loss": 0.70761043,
      "num_input_tokens_seen": 119426875,
      "step": 5549,
      "time_per_iteration": 2.608157157897949
    },
    {
      "auxiliary_loss_clip": 0.01171964,
      "auxiliary_loss_mlp": 0.01023778,
      "balance_loss_clip": 1.04563069,
      "balance_loss_mlp": 1.01551104,
      "epoch": 0.6673480430469548,
      "flos": 22342146975360.0,
      "grad_norm": 2.1631688393372097,
      "language_loss": 0.78522468,
      "learning_rate": 1.0526154729542566e-06,
      "loss": 0.80718207,
      "num_input_tokens_seen": 119446935,
      "step": 5550,
      "time_per_iteration": 2.619513750076294
    },
    {
      "auxiliary_loss_clip": 0.01182144,
      "auxiliary_loss_mlp": 0.01030889,
      "balance_loss_clip": 1.04776382,
      "balance_loss_mlp": 1.02250862,
      "epoch": 0.6674682859375939,
      "flos": 20703830722560.0,
      "grad_norm": 2.5526217929027855,
      "language_loss": 0.79777145,
      "learning_rate": 1.0519295119526699e-06,
      "loss": 0.81990182,
      "num_input_tokens_seen": 119463240,
      "step": 5551,
      "time_per_iteration": 2.7004384994506836
    },
    {
      "auxiliary_loss_clip": 0.01178123,
      "auxiliary_loss_mlp": 0.01023217,
      "balance_loss_clip": 1.04556084,
      "balance_loss_mlp": 1.0151999,
      "epoch": 0.667588528828233,
      "flos": 26206673379840.0,
      "grad_norm": 1.6447472906965628,
      "language_loss": 0.83091831,
      "learning_rate": 1.0512436947726227e-06,
      "loss": 0.85293168,
      "num_input_tokens_seen": 119484655,
      "step": 5552,
      "time_per_iteration": 2.657761812210083
    },
    {
      "auxiliary_loss_clip": 0.01178426,
      "auxiliary_loss_mlp": 0.01022557,
      "balance_loss_clip": 1.0420748,
      "balance_loss_mlp": 1.01443899,
      "epoch": 0.6677087717188721,
      "flos": 23071025756160.0,
      "grad_norm": 2.348801289495868,
      "language_loss": 0.65234208,
      "learning_rate": 1.0505580215181517e-06,
      "loss": 0.67435187,
      "num_input_tokens_seen": 119502895,
      "step": 5553,
      "time_per_iteration": 2.7051923274993896
    },
    {
      "auxiliary_loss_clip": 0.01078604,
      "auxiliary_loss_mlp": 0.01002425,
      "balance_loss_clip": 1.0141263,
      "balance_loss_mlp": 1.00159097,
      "epoch": 0.6678290146095112,
      "flos": 70941315219840.0,
      "grad_norm": 0.7775277281557812,
      "language_loss": 0.56622326,
      "learning_rate": 1.0498724922932753e-06,
      "loss": 0.58703363,
      "num_input_tokens_seen": 119561010,
      "step": 5554,
      "time_per_iteration": 3.122859239578247
    },
    {
      "auxiliary_loss_clip": 0.01177052,
      "auxiliary_loss_mlp": 0.01024698,
      "balance_loss_clip": 1.0520649,
      "balance_loss_mlp": 1.016711,
      "epoch": 0.6679492575001503,
      "flos": 18661088263680.0,
      "grad_norm": 2.3100331793844573,
      "language_loss": 0.86865175,
      "learning_rate": 1.0491871072019851e-06,
      "loss": 0.89066929,
      "num_input_tokens_seen": 119578900,
      "step": 5555,
      "time_per_iteration": 2.590184450149536
    },
    {
      "auxiliary_loss_clip": 0.01183292,
      "auxiliary_loss_mlp": 0.01026554,
      "balance_loss_clip": 1.04409647,
      "balance_loss_mlp": 1.01894808,
      "epoch": 0.6680695003907894,
      "flos": 29711985822720.0,
      "grad_norm": 1.7585739539683138,
      "language_loss": 0.63906002,
      "learning_rate": 1.0485018663482555e-06,
      "loss": 0.6611585,
      "num_input_tokens_seen": 119598920,
      "step": 5556,
      "time_per_iteration": 2.7130472660064697
    },
    {
      "auxiliary_loss_clip": 0.01170832,
      "auxiliary_loss_mlp": 0.01023765,
      "balance_loss_clip": 1.0471015,
      "balance_loss_mlp": 1.01553392,
      "epoch": 0.6681897432814284,
      "flos": 28218964083840.0,
      "grad_norm": 2.806765380996773,
      "language_loss": 0.7080766,
      "learning_rate": 1.0478167698360354e-06,
      "loss": 0.73002255,
      "num_input_tokens_seen": 119618220,
      "step": 5557,
      "time_per_iteration": 3.596163034439087
    },
    {
      "auxiliary_loss_clip": 0.01166858,
      "auxiliary_loss_mlp": 0.01024197,
      "balance_loss_clip": 1.04530609,
      "balance_loss_mlp": 1.01600707,
      "epoch": 0.6683099861720676,
      "flos": 25046543911680.0,
      "grad_norm": 3.311915074983274,
      "language_loss": 0.69979203,
      "learning_rate": 1.0471318177692556e-06,
      "loss": 0.72170258,
      "num_input_tokens_seen": 119638520,
      "step": 5558,
      "time_per_iteration": 2.669912099838257
    },
    {
      "auxiliary_loss_clip": 0.01186217,
      "auxiliary_loss_mlp": 0.01028082,
      "balance_loss_clip": 1.04380417,
      "balance_loss_mlp": 1.02048492,
      "epoch": 0.6684302290627067,
      "flos": 22996977868800.0,
      "grad_norm": 2.6299467778642076,
      "language_loss": 0.75680119,
      "learning_rate": 1.046447010251821e-06,
      "loss": 0.77894419,
      "num_input_tokens_seen": 119655850,
      "step": 5559,
      "time_per_iteration": 2.6944549083709717
    },
    {
      "auxiliary_loss_clip": 0.01177358,
      "auxiliary_loss_mlp": 0.01026798,
      "balance_loss_clip": 1.04915833,
      "balance_loss_mlp": 1.01932955,
      "epoch": 0.6685504719533457,
      "flos": 26573824247040.0,
      "grad_norm": 2.060369278890211,
      "language_loss": 0.75771958,
      "learning_rate": 1.0457623473876157e-06,
      "loss": 0.77976114,
      "num_input_tokens_seen": 119675355,
      "step": 5560,
      "time_per_iteration": 2.666151762008667
    },
    {
      "auxiliary_loss_clip": 0.011711,
      "auxiliary_loss_mlp": 0.01025273,
      "balance_loss_clip": 1.04930258,
      "balance_loss_mlp": 1.01790571,
      "epoch": 0.6686707148439849,
      "flos": 28986087870720.0,
      "grad_norm": 1.9140370674121174,
      "language_loss": 0.71209955,
      "learning_rate": 1.0450778292805046e-06,
      "loss": 0.73406327,
      "num_input_tokens_seen": 119695340,
      "step": 5561,
      "time_per_iteration": 2.6074697971343994
    },
    {
      "auxiliary_loss_clip": 0.01176099,
      "auxiliary_loss_mlp": 0.01025219,
      "balance_loss_clip": 1.04599345,
      "balance_loss_mlp": 1.01771522,
      "epoch": 0.6687909577346239,
      "flos": 23623152687360.0,
      "grad_norm": 2.441310472054835,
      "language_loss": 0.78877223,
      "learning_rate": 1.0443934560343267e-06,
      "loss": 0.81078541,
      "num_input_tokens_seen": 119716750,
      "step": 5562,
      "time_per_iteration": 3.5135953426361084
    },
    {
      "auxiliary_loss_clip": 0.01165459,
      "auxiliary_loss_mlp": 0.01023242,
      "balance_loss_clip": 1.04153037,
      "balance_loss_mlp": 1.01580334,
      "epoch": 0.668911200625263,
      "flos": 23148593176320.0,
      "grad_norm": 2.0823393745547523,
      "language_loss": 0.78283936,
      "learning_rate": 1.0437092277529034e-06,
      "loss": 0.80472642,
      "num_input_tokens_seen": 119736005,
      "step": 5563,
      "time_per_iteration": 2.808917284011841
    },
    {
      "auxiliary_loss_clip": 0.01171342,
      "auxiliary_loss_mlp": 0.01027459,
      "balance_loss_clip": 1.04341066,
      "balance_loss_mlp": 1.02013993,
      "epoch": 0.6690314435159022,
      "flos": 18551919853440.0,
      "grad_norm": 2.0012433747529657,
      "language_loss": 0.73347974,
      "learning_rate": 1.0430251445400292e-06,
      "loss": 0.75546777,
      "num_input_tokens_seen": 119754050,
      "step": 5564,
      "time_per_iteration": 2.627824306488037
    },
    {
      "auxiliary_loss_clip": 0.01186407,
      "auxiliary_loss_mlp": 0.0102709,
      "balance_loss_clip": 1.03979719,
      "balance_loss_mlp": 1.01985979,
      "epoch": 0.6691516864065412,
      "flos": 31759540704000.0,
      "grad_norm": 2.2215740395795223,
      "language_loss": 0.62707901,
      "learning_rate": 1.0423412064994787e-06,
      "loss": 0.64921403,
      "num_input_tokens_seen": 119774820,
      "step": 5565,
      "time_per_iteration": 3.7813754081726074
    },
    {
      "auxiliary_loss_clip": 0.01180165,
      "auxiliary_loss_mlp": 0.01023752,
      "balance_loss_clip": 1.04381323,
      "balance_loss_mlp": 1.01643813,
      "epoch": 0.6692719292971803,
      "flos": 34933864296960.0,
      "grad_norm": 1.991742115749448,
      "language_loss": 0.73759407,
      "learning_rate": 1.0416574137350064e-06,
      "loss": 0.75963318,
      "num_input_tokens_seen": 119795525,
      "step": 5566,
      "time_per_iteration": 2.7695810794830322
    },
    {
      "auxiliary_loss_clip": 0.01167155,
      "auxiliary_loss_mlp": 0.01026487,
      "balance_loss_clip": 1.047436,
      "balance_loss_mlp": 1.01853597,
      "epoch": 0.6693921721878194,
      "flos": 20449188230400.0,
      "grad_norm": 2.3645788584715968,
      "language_loss": 0.81237668,
      "learning_rate": 1.0409737663503428e-06,
      "loss": 0.83431309,
      "num_input_tokens_seen": 119813905,
      "step": 5567,
      "time_per_iteration": 2.6378719806671143
    },
    {
      "auxiliary_loss_clip": 0.01170351,
      "auxiliary_loss_mlp": 0.01028425,
      "balance_loss_clip": 1.04463625,
      "balance_loss_mlp": 1.01978564,
      "epoch": 0.6695124150784585,
      "flos": 16614538963200.0,
      "grad_norm": 2.323747230426692,
      "language_loss": 0.82815921,
      "learning_rate": 1.040290264449196e-06,
      "loss": 0.85014695,
      "num_input_tokens_seen": 119832010,
      "step": 5568,
      "time_per_iteration": 3.513988733291626
    },
    {
      "auxiliary_loss_clip": 0.01169793,
      "auxiliary_loss_mlp": 0.01025061,
      "balance_loss_clip": 1.04787636,
      "balance_loss_mlp": 1.01782525,
      "epoch": 0.6696326579690975,
      "flos": 26652145852800.0,
      "grad_norm": 1.9946911395247673,
      "language_loss": 0.64340502,
      "learning_rate": 1.0396069081352532e-06,
      "loss": 0.66535354,
      "num_input_tokens_seen": 119851165,
      "step": 5569,
      "time_per_iteration": 2.592273712158203
    },
    {
      "auxiliary_loss_clip": 0.01069767,
      "auxiliary_loss_mlp": 0.01001043,
      "balance_loss_clip": 1.01497555,
      "balance_loss_mlp": 1.00024986,
      "epoch": 0.6697529008597367,
      "flos": 66964603662720.0,
      "grad_norm": 0.7731036316662131,
      "language_loss": 0.55939907,
      "learning_rate": 1.0389236975121782e-06,
      "loss": 0.58010715,
      "num_input_tokens_seen": 119906015,
      "step": 5570,
      "time_per_iteration": 3.085378408432007
    },
    {
      "auxiliary_loss_clip": 0.01173366,
      "auxiliary_loss_mlp": 0.01021192,
      "balance_loss_clip": 1.04975832,
      "balance_loss_mlp": 1.01311541,
      "epoch": 0.6698731437503758,
      "flos": 20886939279360.0,
      "grad_norm": 2.2934486018466615,
      "language_loss": 0.71097821,
      "learning_rate": 1.0382406326836147e-06,
      "loss": 0.73292375,
      "num_input_tokens_seen": 119925160,
      "step": 5571,
      "time_per_iteration": 2.6009936332702637
    },
    {
      "auxiliary_loss_clip": 0.01180752,
      "auxiliary_loss_mlp": 0.01024326,
      "balance_loss_clip": 1.04948854,
      "balance_loss_mlp": 1.01607037,
      "epoch": 0.6699933866410148,
      "flos": 20409470766720.0,
      "grad_norm": 2.2352758657632363,
      "language_loss": 0.76158047,
      "learning_rate": 1.0375577137531828e-06,
      "loss": 0.78363127,
      "num_input_tokens_seen": 119943720,
      "step": 5572,
      "time_per_iteration": 2.598377227783203
    },
    {
      "auxiliary_loss_clip": 0.01179897,
      "auxiliary_loss_mlp": 0.01023531,
      "balance_loss_clip": 1.04785776,
      "balance_loss_mlp": 1.01527607,
      "epoch": 0.670113629531654,
      "flos": 29023075900800.0,
      "grad_norm": 1.7790178700371093,
      "language_loss": 0.72182715,
      "learning_rate": 1.0368749408244802e-06,
      "loss": 0.74386144,
      "num_input_tokens_seen": 119966640,
      "step": 5573,
      "time_per_iteration": 2.7333250045776367
    },
    {
      "auxiliary_loss_clip": 0.01169708,
      "auxiliary_loss_mlp": 0.0102676,
      "balance_loss_clip": 1.04836226,
      "balance_loss_mlp": 1.01915145,
      "epoch": 0.670233872422293,
      "flos": 19791699730560.0,
      "grad_norm": 1.7335454991026196,
      "language_loss": 0.78909361,
      "learning_rate": 1.0361923140010836e-06,
      "loss": 0.81105834,
      "num_input_tokens_seen": 119985125,
      "step": 5574,
      "time_per_iteration": 2.593083143234253
    },
    {
      "auxiliary_loss_clip": 0.01177922,
      "auxiliary_loss_mlp": 0.01019847,
      "balance_loss_clip": 1.04718399,
      "balance_loss_mlp": 1.01188409,
      "epoch": 0.6703541153129321,
      "flos": 24243689070720.0,
      "grad_norm": 9.657216540753344,
      "language_loss": 0.63299155,
      "learning_rate": 1.0355098333865455e-06,
      "loss": 0.65496927,
      "num_input_tokens_seen": 120004355,
      "step": 5575,
      "time_per_iteration": 2.7149312496185303
    },
    {
      "auxiliary_loss_clip": 0.01169416,
      "auxiliary_loss_mlp": 0.01027087,
      "balance_loss_clip": 1.05045319,
      "balance_loss_mlp": 1.01945782,
      "epoch": 0.6704743582035713,
      "flos": 26688523351680.0,
      "grad_norm": 1.7830736910663474,
      "language_loss": 0.69574165,
      "learning_rate": 1.0348274990844006e-06,
      "loss": 0.71770668,
      "num_input_tokens_seen": 120027115,
      "step": 5576,
      "time_per_iteration": 2.716463088989258
    },
    {
      "auxiliary_loss_clip": 0.01176228,
      "auxiliary_loss_mlp": 0.01027417,
      "balance_loss_clip": 1.05055439,
      "balance_loss_mlp": 1.01959717,
      "epoch": 0.6705946010942103,
      "flos": 23514379326720.0,
      "grad_norm": 2.8917342504523664,
      "language_loss": 0.72819036,
      "learning_rate": 1.034145311198155e-06,
      "loss": 0.7502268,
      "num_input_tokens_seen": 120047130,
      "step": 5577,
      "time_per_iteration": 2.714376926422119
    },
    {
      "auxiliary_loss_clip": 0.01166422,
      "auxiliary_loss_mlp": 0.01022006,
      "balance_loss_clip": 1.04690552,
      "balance_loss_mlp": 1.01459122,
      "epoch": 0.6707148439848494,
      "flos": 24061011477120.0,
      "grad_norm": 2.0182151913292117,
      "language_loss": 0.63933802,
      "learning_rate": 1.0334632698312989e-06,
      "loss": 0.66122222,
      "num_input_tokens_seen": 120067925,
      "step": 5578,
      "time_per_iteration": 2.562319278717041
    },
    {
      "auxiliary_loss_clip": 0.0116967,
      "auxiliary_loss_mlp": 0.0102672,
      "balance_loss_clip": 1.04521275,
      "balance_loss_mlp": 1.01873589,
      "epoch": 0.6708350868754885,
      "flos": 22528667324160.0,
      "grad_norm": 2.074845687743777,
      "language_loss": 0.75283813,
      "learning_rate": 1.032781375087295e-06,
      "loss": 0.77480203,
      "num_input_tokens_seen": 120087825,
      "step": 5579,
      "time_per_iteration": 2.671759843826294
    },
    {
      "auxiliary_loss_clip": 0.01180561,
      "auxiliary_loss_mlp": 0.01024111,
      "balance_loss_clip": 1.05061173,
      "balance_loss_mlp": 1.01687551,
      "epoch": 0.6709553297661276,
      "flos": 25227749047680.0,
      "grad_norm": 2.6081735296904505,
      "language_loss": 0.67598277,
      "learning_rate": 1.0320996270695891e-06,
      "loss": 0.69802952,
      "num_input_tokens_seen": 120108895,
      "step": 5580,
      "time_per_iteration": 2.6400651931762695
    },
    {
      "auxiliary_loss_clip": 0.01177302,
      "auxiliary_loss_mlp": 0.01023622,
      "balance_loss_clip": 1.04314446,
      "balance_loss_mlp": 1.01593876,
      "epoch": 0.6710755726567667,
      "flos": 20448757267200.0,
      "grad_norm": 2.0915140296150794,
      "language_loss": 0.73628795,
      "learning_rate": 1.0314180258815998e-06,
      "loss": 0.7582972,
      "num_input_tokens_seen": 120127535,
      "step": 5581,
      "time_per_iteration": 2.673839569091797
    },
    {
      "auxiliary_loss_clip": 0.01169405,
      "auxiliary_loss_mlp": 0.01022325,
      "balance_loss_clip": 1.04218829,
      "balance_loss_mlp": 1.01494646,
      "epoch": 0.6711958155474057,
      "flos": 25995411538560.0,
      "grad_norm": 3.0343253452539374,
      "language_loss": 0.74320918,
      "learning_rate": 1.0307365716267247e-06,
      "loss": 0.76512647,
      "num_input_tokens_seen": 120147980,
      "step": 5582,
      "time_per_iteration": 2.670111656188965
    },
    {
      "auxiliary_loss_clip": 0.01175216,
      "auxiliary_loss_mlp": 0.01024382,
      "balance_loss_clip": 1.04915965,
      "balance_loss_mlp": 1.01674056,
      "epoch": 0.6713160584380449,
      "flos": 19937712516480.0,
      "grad_norm": 2.042361001118012,
      "language_loss": 0.78320646,
      "learning_rate": 1.0300552644083423e-06,
      "loss": 0.80520242,
      "num_input_tokens_seen": 120166905,
      "step": 5583,
      "time_per_iteration": 2.608057737350464
    },
    {
      "auxiliary_loss_clip": 0.01182817,
      "auxiliary_loss_mlp": 0.01021922,
      "balance_loss_clip": 1.04708707,
      "balance_loss_mlp": 1.01416481,
      "epoch": 0.6714363013286839,
      "flos": 18223373128320.0,
      "grad_norm": 2.524674461443693,
      "language_loss": 0.72555751,
      "learning_rate": 1.0293741043298036e-06,
      "loss": 0.74760491,
      "num_input_tokens_seen": 120185255,
      "step": 5584,
      "time_per_iteration": 3.527891159057617
    },
    {
      "auxiliary_loss_clip": 0.01181518,
      "auxiliary_loss_mlp": 0.01033728,
      "balance_loss_clip": 1.04846787,
      "balance_loss_mlp": 1.02537739,
      "epoch": 0.671556544219323,
      "flos": 25812374808960.0,
      "grad_norm": 2.1849812294389404,
      "language_loss": 0.7181195,
      "learning_rate": 1.0286930914944436e-06,
      "loss": 0.74027193,
      "num_input_tokens_seen": 120205070,
      "step": 5585,
      "time_per_iteration": 2.702929735183716
    },
    {
      "auxiliary_loss_clip": 0.01170597,
      "auxiliary_loss_mlp": 0.01023624,
      "balance_loss_clip": 1.04670882,
      "balance_loss_mlp": 1.01628041,
      "epoch": 0.6716767871099621,
      "flos": 15850431918720.0,
      "grad_norm": 2.5737043680386718,
      "language_loss": 0.77092624,
      "learning_rate": 1.0280122260055684e-06,
      "loss": 0.79286849,
      "num_input_tokens_seen": 120220780,
      "step": 5586,
      "time_per_iteration": 2.5734753608703613
    },
    {
      "auxiliary_loss_clip": 0.01173583,
      "auxiliary_loss_mlp": 0.01032783,
      "balance_loss_clip": 1.05013776,
      "balance_loss_mlp": 1.02455711,
      "epoch": 0.6717970300006012,
      "flos": 19756112330880.0,
      "grad_norm": 2.0944208311423056,
      "language_loss": 0.82337695,
      "learning_rate": 1.0273315079664652e-06,
      "loss": 0.84544063,
      "num_input_tokens_seen": 120238735,
      "step": 5587,
      "time_per_iteration": 2.5896220207214355
    },
    {
      "auxiliary_loss_clip": 0.01177102,
      "auxiliary_loss_mlp": 0.01021298,
      "balance_loss_clip": 1.05016732,
      "balance_loss_mlp": 1.01392508,
      "epoch": 0.6719172728912403,
      "flos": 25485049146240.0,
      "grad_norm": 2.3155174096723212,
      "language_loss": 0.74169213,
      "learning_rate": 1.0266509374803992e-06,
      "loss": 0.76367617,
      "num_input_tokens_seen": 120259895,
      "step": 5588,
      "time_per_iteration": 3.5590341091156006
    },
    {
      "auxiliary_loss_clip": 0.01172077,
      "auxiliary_loss_mlp": 0.01208202,
      "balance_loss_clip": 1.04873228,
      "balance_loss_mlp": 1.00093162,
      "epoch": 0.6720375157818794,
      "flos": 15880344969600.0,
      "grad_norm": 7.573655890824284,
      "language_loss": 0.84333181,
      "learning_rate": 1.0259705146506123e-06,
      "loss": 0.86713463,
      "num_input_tokens_seen": 120274790,
      "step": 5589,
      "time_per_iteration": 2.533674716949463
    },
    {
      "auxiliary_loss_clip": 0.01176541,
      "auxiliary_loss_mlp": 0.01027939,
      "balance_loss_clip": 1.04790998,
      "balance_loss_mlp": 1.02038753,
      "epoch": 0.6721577586725185,
      "flos": 32010843231360.0,
      "grad_norm": 2.2785407739699637,
      "language_loss": 0.77428424,
      "learning_rate": 1.025290239580324e-06,
      "loss": 0.79632908,
      "num_input_tokens_seen": 120295460,
      "step": 5590,
      "time_per_iteration": 2.679230213165283
    },
    {
      "auxiliary_loss_clip": 0.01182264,
      "auxiliary_loss_mlp": 0.01028821,
      "balance_loss_clip": 1.04320717,
      "balance_loss_mlp": 1.02120399,
      "epoch": 0.6722780015631575,
      "flos": 20737873837440.0,
      "grad_norm": 1.8673055297105554,
      "language_loss": 0.75459927,
      "learning_rate": 1.0246101123727313e-06,
      "loss": 0.77671015,
      "num_input_tokens_seen": 120314440,
      "step": 5591,
      "time_per_iteration": 3.589691638946533
    },
    {
      "auxiliary_loss_clip": 0.0117186,
      "auxiliary_loss_mlp": 0.01029381,
      "balance_loss_clip": 1.04542494,
      "balance_loss_mlp": 1.02229667,
      "epoch": 0.6723982444537967,
      "flos": 16909617191040.0,
      "grad_norm": 1.8879593372622494,
      "language_loss": 0.78535283,
      "learning_rate": 1.0239301331310085e-06,
      "loss": 0.80736524,
      "num_input_tokens_seen": 120332060,
      "step": 5592,
      "time_per_iteration": 2.5828371047973633
    },
    {
      "auxiliary_loss_clip": 0.01170767,
      "auxiliary_loss_mlp": 0.01025441,
      "balance_loss_clip": 1.04722667,
      "balance_loss_mlp": 1.01804399,
      "epoch": 0.6725184873444358,
      "flos": 20667812359680.0,
      "grad_norm": 2.7168708497394802,
      "language_loss": 0.88489807,
      "learning_rate": 1.0232503019583088e-06,
      "loss": 0.90686011,
      "num_input_tokens_seen": 120351670,
      "step": 5593,
      "time_per_iteration": 2.6360924243927
    },
    {
      "auxiliary_loss_clip": 0.01168537,
      "auxiliary_loss_mlp": 0.01028015,
      "balance_loss_clip": 1.04747665,
      "balance_loss_mlp": 1.02025437,
      "epoch": 0.6726387302350748,
      "flos": 23727616416000.0,
      "grad_norm": 2.8370010417497067,
      "language_loss": 0.69784153,
      "learning_rate": 1.0225706189577619e-06,
      "loss": 0.71980703,
      "num_input_tokens_seen": 120370195,
      "step": 5594,
      "time_per_iteration": 2.55320405960083
    },
    {
      "auxiliary_loss_clip": 0.01175857,
      "auxiliary_loss_mlp": 0.01024757,
      "balance_loss_clip": 1.04825747,
      "balance_loss_mlp": 1.01685309,
      "epoch": 0.672758973125714,
      "flos": 15188274650880.0,
      "grad_norm": 2.1353889052679924,
      "language_loss": 0.74737698,
      "learning_rate": 1.021891084232475e-06,
      "loss": 0.76938313,
      "num_input_tokens_seen": 120388130,
      "step": 5595,
      "time_per_iteration": 3.556330919265747
    },
    {
      "auxiliary_loss_clip": 0.0117257,
      "auxiliary_loss_mlp": 0.01027097,
      "balance_loss_clip": 1.04599714,
      "balance_loss_mlp": 1.01878524,
      "epoch": 0.672879216016353,
      "flos": 18077252601600.0,
      "grad_norm": 5.632082269204781,
      "language_loss": 0.80497169,
      "learning_rate": 1.0212116978855325e-06,
      "loss": 0.82696831,
      "num_input_tokens_seen": 120406145,
      "step": 5596,
      "time_per_iteration": 2.588441848754883
    },
    {
      "auxiliary_loss_clip": 0.01177106,
      "auxiliary_loss_mlp": 0.01018385,
      "balance_loss_clip": 1.04432249,
      "balance_loss_mlp": 1.0110358,
      "epoch": 0.6729994589069921,
      "flos": 23476349802240.0,
      "grad_norm": 1.9996669841340304,
      "language_loss": 0.78775668,
      "learning_rate": 1.020532460019997e-06,
      "loss": 0.80971158,
      "num_input_tokens_seen": 120425395,
      "step": 5597,
      "time_per_iteration": 2.715425968170166
    },
    {
      "auxiliary_loss_clip": 0.01188959,
      "auxiliary_loss_mlp": 0.01025005,
      "balance_loss_clip": 1.04009962,
      "balance_loss_mlp": 1.01769757,
      "epoch": 0.6731197017976313,
      "flos": 26322018929280.0,
      "grad_norm": 2.138963165133628,
      "language_loss": 0.70967484,
      "learning_rate": 1.0198533707389096e-06,
      "loss": 0.7318145,
      "num_input_tokens_seen": 120446270,
      "step": 5598,
      "time_per_iteration": 2.8236172199249268
    },
    {
      "auxiliary_loss_clip": 0.01169182,
      "auxiliary_loss_mlp": 0.0120831,
      "balance_loss_clip": 1.04774404,
      "balance_loss_mlp": 1.00081587,
      "epoch": 0.6732399446882703,
      "flos": 21616428591360.0,
      "grad_norm": 2.113969486230489,
      "language_loss": 0.73288733,
      "learning_rate": 1.0191744301452853e-06,
      "loss": 0.75666225,
      "num_input_tokens_seen": 120465570,
      "step": 5599,
      "time_per_iteration": 2.6040878295898438
    },
    {
      "auxiliary_loss_clip": 0.01170093,
      "auxiliary_loss_mlp": 0.01026579,
      "balance_loss_clip": 1.04847431,
      "balance_loss_mlp": 1.01887822,
      "epoch": 0.6733601875789094,
      "flos": 25880173729920.0,
      "grad_norm": 1.9560405842735669,
      "language_loss": 0.70410073,
      "learning_rate": 1.0184956383421208e-06,
      "loss": 0.72606748,
      "num_input_tokens_seen": 120484220,
      "step": 5600,
      "time_per_iteration": 2.6601006984710693
    },
    {
      "auxiliary_loss_clip": 0.01176154,
      "auxiliary_loss_mlp": 0.01023623,
      "balance_loss_clip": 1.04898107,
      "balance_loss_mlp": 1.01595831,
      "epoch": 0.6734804304695485,
      "flos": 22929573997440.0,
      "grad_norm": 2.5534112869200665,
      "language_loss": 0.65270817,
      "learning_rate": 1.017816995432387e-06,
      "loss": 0.67470598,
      "num_input_tokens_seen": 120503320,
      "step": 5601,
      "time_per_iteration": 2.6754038333892822
    },
    {
      "auxiliary_loss_clip": 0.01175751,
      "auxiliary_loss_mlp": 0.01021594,
      "balance_loss_clip": 1.04685426,
      "balance_loss_mlp": 1.01402092,
      "epoch": 0.6736006733601876,
      "flos": 18697968552960.0,
      "grad_norm": 2.064538348927552,
      "language_loss": 0.74526572,
      "learning_rate": 1.0171385015190353e-06,
      "loss": 0.76723909,
      "num_input_tokens_seen": 120523180,
      "step": 5602,
      "time_per_iteration": 2.6470582485198975
    },
    {
      "auxiliary_loss_clip": 0.01170135,
      "auxiliary_loss_mlp": 0.01208247,
      "balance_loss_clip": 1.04723573,
      "balance_loss_mlp": 1.00118732,
      "epoch": 0.6737209162508266,
      "flos": 19427745173760.0,
      "grad_norm": 3.719265756288167,
      "language_loss": 0.7303642,
      "learning_rate": 1.0164601567049908e-06,
      "loss": 0.75414801,
      "num_input_tokens_seen": 120541710,
      "step": 5603,
      "time_per_iteration": 2.6102592945098877
    },
    {
      "auxiliary_loss_clip": 0.01175832,
      "auxiliary_loss_mlp": 0.01025271,
      "balance_loss_clip": 1.04694438,
      "balance_loss_mlp": 1.01712036,
      "epoch": 0.6738411591414658,
      "flos": 20158060498560.0,
      "grad_norm": 1.9672768374657206,
      "language_loss": 0.80185866,
      "learning_rate": 1.015781961093158e-06,
      "loss": 0.82386976,
      "num_input_tokens_seen": 120561030,
      "step": 5604,
      "time_per_iteration": 2.6018226146698
    },
    {
      "auxiliary_loss_clip": 0.01175764,
      "auxiliary_loss_mlp": 0.01023472,
      "balance_loss_clip": 1.0417285,
      "balance_loss_mlp": 1.01602447,
      "epoch": 0.6739614020321049,
      "flos": 21653847584640.0,
      "grad_norm": 1.5874017534787894,
      "language_loss": 0.7686373,
      "learning_rate": 1.0151039147864197e-06,
      "loss": 0.79062968,
      "num_input_tokens_seen": 120581005,
      "step": 5605,
      "time_per_iteration": 2.5977632999420166
    },
    {
      "auxiliary_loss_clip": 0.01178135,
      "auxiliary_loss_mlp": 0.01024656,
      "balance_loss_clip": 1.04349613,
      "balance_loss_mlp": 1.01632309,
      "epoch": 0.6740816449227439,
      "flos": 19171702051200.0,
      "grad_norm": 5.831471225655623,
      "language_loss": 0.65970123,
      "learning_rate": 1.0144260178876336e-06,
      "loss": 0.68172908,
      "num_input_tokens_seen": 120600350,
      "step": 5606,
      "time_per_iteration": 2.806347370147705
    },
    {
      "auxiliary_loss_clip": 0.01181214,
      "auxiliary_loss_mlp": 0.01020602,
      "balance_loss_clip": 1.04653382,
      "balance_loss_mlp": 1.01346183,
      "epoch": 0.6742018878133831,
      "flos": 21097015971840.0,
      "grad_norm": 2.365615697583356,
      "language_loss": 0.67493004,
      "learning_rate": 1.0137482704996388e-06,
      "loss": 0.69694817,
      "num_input_tokens_seen": 120614700,
      "step": 5607,
      "time_per_iteration": 2.674010753631592
    },
    {
      "auxiliary_loss_clip": 0.0118363,
      "auxiliary_loss_mlp": 0.01028991,
      "balance_loss_clip": 1.04616559,
      "balance_loss_mlp": 1.02114666,
      "epoch": 0.6743221307040221,
      "flos": 23549966726400.0,
      "grad_norm": 3.2537901901023774,
      "language_loss": 0.79113096,
      "learning_rate": 1.0130706727252461e-06,
      "loss": 0.81325722,
      "num_input_tokens_seen": 120631755,
      "step": 5608,
      "time_per_iteration": 2.6910946369171143
    },
    {
      "auxiliary_loss_clip": 0.01181423,
      "auxiliary_loss_mlp": 0.01024215,
      "balance_loss_clip": 1.04582858,
      "balance_loss_mlp": 1.01685977,
      "epoch": 0.6744423735946612,
      "flos": 16249542912000.0,
      "grad_norm": 2.39276745684465,
      "language_loss": 0.68188334,
      "learning_rate": 1.0123932246672468e-06,
      "loss": 0.7039398,
      "num_input_tokens_seen": 120645900,
      "step": 5609,
      "time_per_iteration": 2.6644840240478516
    },
    {
      "auxiliary_loss_clip": 0.01091557,
      "auxiliary_loss_mlp": 0.01198995,
      "balance_loss_clip": 1.01303232,
      "balance_loss_mlp": 1.00003707,
      "epoch": 0.6745626164853004,
      "flos": 57843257829120.0,
      "grad_norm": 0.7483326284533254,
      "language_loss": 0.55766976,
      "learning_rate": 1.0117159264284114e-06,
      "loss": 0.58057529,
      "num_input_tokens_seen": 120709070,
      "step": 5610,
      "time_per_iteration": 4.092698097229004
    },
    {
      "auxiliary_loss_clip": 0.01180566,
      "auxiliary_loss_mlp": 0.0102527,
      "balance_loss_clip": 1.04836941,
      "balance_loss_mlp": 1.01786363,
      "epoch": 0.6746828593759394,
      "flos": 20485027025280.0,
      "grad_norm": 1.6936182635626849,
      "language_loss": 0.77202153,
      "learning_rate": 1.0110387781114837e-06,
      "loss": 0.7940799,
      "num_input_tokens_seen": 120727685,
      "step": 5611,
      "time_per_iteration": 2.6193764209747314
    },
    {
      "auxiliary_loss_clip": 0.0116852,
      "auxiliary_loss_mlp": 0.01021697,
      "balance_loss_clip": 1.04811502,
      "balance_loss_mlp": 1.01422894,
      "epoch": 0.6748031022665785,
      "flos": 19208223204480.0,
      "grad_norm": 2.0684439648135338,
      "language_loss": 0.77427602,
      "learning_rate": 1.0103617798191872e-06,
      "loss": 0.79617822,
      "num_input_tokens_seen": 120747160,
      "step": 5612,
      "time_per_iteration": 2.536079168319702
    },
    {
      "auxiliary_loss_clip": 0.01171731,
      "auxiliary_loss_mlp": 0.01021339,
      "balance_loss_clip": 1.04582047,
      "balance_loss_mlp": 1.01362646,
      "epoch": 0.6749233451572175,
      "flos": 15195026407680.0,
      "grad_norm": 2.4352429159584172,
      "language_loss": 0.82551539,
      "learning_rate": 1.0096849316542217e-06,
      "loss": 0.8474462,
      "num_input_tokens_seen": 120763710,
      "step": 5613,
      "time_per_iteration": 2.5670230388641357
    },
    {
      "auxiliary_loss_clip": 0.0117125,
      "auxiliary_loss_mlp": 0.01019456,
      "balance_loss_clip": 1.03636384,
      "balance_loss_mlp": 1.0118804,
      "epoch": 0.6750435880478567,
      "flos": 26499489050880.0,
      "grad_norm": 3.623312198864256,
      "language_loss": 0.75064528,
      "learning_rate": 1.0090082337192643e-06,
      "loss": 0.77255231,
      "num_input_tokens_seen": 120783355,
      "step": 5614,
      "time_per_iteration": 2.7300302982330322
    },
    {
      "auxiliary_loss_clip": 0.0117609,
      "auxiliary_loss_mlp": 0.01027029,
      "balance_loss_clip": 1.03659999,
      "balance_loss_mlp": 1.01958418,
      "epoch": 0.6751638309384957,
      "flos": 23404313076480.0,
      "grad_norm": 4.178337267945883,
      "language_loss": 0.78753054,
      "learning_rate": 1.0083316861169705e-06,
      "loss": 0.80956173,
      "num_input_tokens_seen": 120802090,
      "step": 5615,
      "time_per_iteration": 3.6266980171203613
    },
    {
      "auxiliary_loss_clip": 0.01182694,
      "auxiliary_loss_mlp": 0.01023961,
      "balance_loss_clip": 1.0427295,
      "balance_loss_mlp": 1.01518726,
      "epoch": 0.6752840738291348,
      "flos": 23441408847360.0,
      "grad_norm": 2.1455053141499496,
      "language_loss": 0.71771938,
      "learning_rate": 1.0076552889499713e-06,
      "loss": 0.73978591,
      "num_input_tokens_seen": 120822855,
      "step": 5616,
      "time_per_iteration": 2.7155165672302246
    },
    {
      "auxiliary_loss_clip": 0.01172957,
      "auxiliary_loss_mlp": 0.01026568,
      "balance_loss_clip": 1.0492605,
      "balance_loss_mlp": 1.01957583,
      "epoch": 0.675404316719774,
      "flos": 30335826257280.0,
      "grad_norm": 2.2008178655272594,
      "language_loss": 0.7402966,
      "learning_rate": 1.006979042320876e-06,
      "loss": 0.76229191,
      "num_input_tokens_seen": 120843070,
      "step": 5617,
      "time_per_iteration": 3.61798095703125
    },
    {
      "auxiliary_loss_clip": 0.01170918,
      "auxiliary_loss_mlp": 0.0102006,
      "balance_loss_clip": 1.04298949,
      "balance_loss_mlp": 1.01226926,
      "epoch": 0.675524559610413,
      "flos": 23622613983360.0,
      "grad_norm": 1.9669592268233322,
      "language_loss": 0.63158029,
      "learning_rate": 1.0063029463322702e-06,
      "loss": 0.65349007,
      "num_input_tokens_seen": 120863345,
      "step": 5618,
      "time_per_iteration": 2.6742706298828125
    },
    {
      "auxiliary_loss_clip": 0.01174025,
      "auxiliary_loss_mlp": 0.01208633,
      "balance_loss_clip": 1.0412426,
      "balance_loss_mlp": 1.00108659,
      "epoch": 0.6756448025010521,
      "flos": 21248631279360.0,
      "grad_norm": 3.984645223926844,
      "language_loss": 0.75405431,
      "learning_rate": 1.0056270010867164e-06,
      "loss": 0.77788091,
      "num_input_tokens_seen": 120880915,
      "step": 5619,
      "time_per_iteration": 2.7682206630706787
    },
    {
      "auxiliary_loss_clip": 0.0117479,
      "auxiliary_loss_mlp": 0.01027843,
      "balance_loss_clip": 1.04176342,
      "balance_loss_mlp": 1.01959968,
      "epoch": 0.6757650453916912,
      "flos": 21646521210240.0,
      "grad_norm": 2.5815046362769296,
      "language_loss": 0.78262383,
      "learning_rate": 1.004951206686758e-06,
      "loss": 0.80465013,
      "num_input_tokens_seen": 120899190,
      "step": 5620,
      "time_per_iteration": 2.7437524795532227
    },
    {
      "auxiliary_loss_clip": 0.01168745,
      "auxiliary_loss_mlp": 0.0103231,
      "balance_loss_clip": 1.04665875,
      "balance_loss_mlp": 1.02452612,
      "epoch": 0.6758852882823303,
      "flos": 21795658479360.0,
      "grad_norm": 3.1201002596493543,
      "language_loss": 0.72093558,
      "learning_rate": 1.0042755632349087e-06,
      "loss": 0.74294615,
      "num_input_tokens_seen": 120916080,
      "step": 5621,
      "time_per_iteration": 3.6244122982025146
    },
    {
      "auxiliary_loss_clip": 0.01176828,
      "auxiliary_loss_mlp": 0.01028635,
      "balance_loss_clip": 1.04436207,
      "balance_loss_mlp": 1.02071977,
      "epoch": 0.6760055311729694,
      "flos": 27088783580160.0,
      "grad_norm": 2.702592589846946,
      "language_loss": 0.62437797,
      "learning_rate": 1.0036000708336653e-06,
      "loss": 0.64643258,
      "num_input_tokens_seen": 120935210,
      "step": 5622,
      "time_per_iteration": 2.7430381774902344
    },
    {
      "auxiliary_loss_clip": 0.01177373,
      "auxiliary_loss_mlp": 0.01026374,
      "balance_loss_clip": 1.04630399,
      "balance_loss_mlp": 1.01879835,
      "epoch": 0.6761257740636085,
      "flos": 17999792922240.0,
      "grad_norm": 2.890541077029468,
      "language_loss": 0.7999925,
      "learning_rate": 1.0029247295854984e-06,
      "loss": 0.82202995,
      "num_input_tokens_seen": 120951830,
      "step": 5623,
      "time_per_iteration": 2.6082465648651123
    },
    {
      "auxiliary_loss_clip": 0.01183885,
      "auxiliary_loss_mlp": 0.01025609,
      "balance_loss_clip": 1.04665792,
      "balance_loss_mlp": 1.0190556,
      "epoch": 0.6762460169542476,
      "flos": 15121912273920.0,
      "grad_norm": 2.0993627308093155,
      "language_loss": 0.71928155,
      "learning_rate": 1.0022495395928588e-06,
      "loss": 0.74137652,
      "num_input_tokens_seen": 120970310,
      "step": 5624,
      "time_per_iteration": 2.5704643726348877
    },
    {
      "auxiliary_loss_clip": 0.01069184,
      "auxiliary_loss_mlp": 0.01001306,
      "balance_loss_clip": 1.01469469,
      "balance_loss_mlp": 1.00051355,
      "epoch": 0.6763662598448866,
      "flos": 67886970030720.0,
      "grad_norm": 0.796973305416409,
      "language_loss": 0.62376314,
      "learning_rate": 1.0015745009581697e-06,
      "loss": 0.64446801,
      "num_input_tokens_seen": 121031915,
      "step": 5625,
      "time_per_iteration": 3.1690688133239746
    },
    {
      "auxiliary_loss_clip": 0.01170144,
      "auxiliary_loss_mlp": 0.01023093,
      "balance_loss_clip": 1.04849577,
      "balance_loss_mlp": 1.01588643,
      "epoch": 0.6764865027355258,
      "flos": 20631829910400.0,
      "grad_norm": 2.0879100291581447,
      "language_loss": 0.66911256,
      "learning_rate": 1.0008996137838343e-06,
      "loss": 0.69104493,
      "num_input_tokens_seen": 121050890,
      "step": 5626,
      "time_per_iteration": 2.591298818588257
    },
    {
      "auxiliary_loss_clip": 0.01177513,
      "auxiliary_loss_mlp": 0.01025637,
      "balance_loss_clip": 1.05168021,
      "balance_loss_mlp": 1.01747155,
      "epoch": 0.6766067456261649,
      "flos": 21215809226880.0,
      "grad_norm": 2.3813892113852324,
      "language_loss": 0.79682988,
      "learning_rate": 1.000224878172234e-06,
      "loss": 0.81886137,
      "num_input_tokens_seen": 121070015,
      "step": 5627,
      "time_per_iteration": 2.6064751148223877
    },
    {
      "auxiliary_loss_clip": 0.01174995,
      "auxiliary_loss_mlp": 0.01024977,
      "balance_loss_clip": 1.04672885,
      "balance_loss_mlp": 1.01738954,
      "epoch": 0.6767269885168039,
      "flos": 19938251220480.0,
      "grad_norm": 5.520757486369359,
      "language_loss": 0.7274909,
      "learning_rate": 9.99550294225724e-07,
      "loss": 0.74949062,
      "num_input_tokens_seen": 121089170,
      "step": 5628,
      "time_per_iteration": 2.621767044067383
    },
    {
      "auxiliary_loss_clip": 0.01180876,
      "auxiliary_loss_mlp": 0.01025531,
      "balance_loss_clip": 1.03999925,
      "balance_loss_mlp": 1.0177114,
      "epoch": 0.6768472314074431,
      "flos": 20814076540800.0,
      "grad_norm": 1.963912116671561,
      "language_loss": 0.7264939,
      "learning_rate": 9.988758620466402e-07,
      "loss": 0.74855798,
      "num_input_tokens_seen": 121108040,
      "step": 5629,
      "time_per_iteration": 2.688610553741455
    },
    {
      "auxiliary_loss_clip": 0.01187844,
      "auxiliary_loss_mlp": 0.01022766,
      "balance_loss_clip": 1.04112804,
      "balance_loss_mlp": 1.01552415,
      "epoch": 0.6769674742980821,
      "flos": 23186012169600.0,
      "grad_norm": 1.6574364677446232,
      "language_loss": 0.76289034,
      "learning_rate": 9.982015817372917e-07,
      "loss": 0.78499645,
      "num_input_tokens_seen": 121128480,
      "step": 5630,
      "time_per_iteration": 2.74533748626709
    },
    {
      "auxiliary_loss_clip": 0.01176542,
      "auxiliary_loss_mlp": 0.01027372,
      "balance_loss_clip": 1.03900695,
      "balance_loss_mlp": 1.01943922,
      "epoch": 0.6770877171887212,
      "flos": 24242934885120.0,
      "grad_norm": 3.240435388631249,
      "language_loss": 0.82050312,
      "learning_rate": 9.975274533999657e-07,
      "loss": 0.84254217,
      "num_input_tokens_seen": 121148010,
      "step": 5631,
      "time_per_iteration": 2.7065513134002686
    },
    {
      "auxiliary_loss_clip": 0.01172248,
      "auxiliary_loss_mlp": 0.01027658,
      "balance_loss_clip": 1.04817247,
      "balance_loss_mlp": 1.01948619,
      "epoch": 0.6772079600793603,
      "flos": 18141567903360.0,
      "grad_norm": 2.6335480752859803,
      "language_loss": 0.84510523,
      "learning_rate": 9.96853477136929e-07,
      "loss": 0.86710429,
      "num_input_tokens_seen": 121162755,
      "step": 5632,
      "time_per_iteration": 2.6061432361602783
    },
    {
      "auxiliary_loss_clip": 0.01169547,
      "auxiliary_loss_mlp": 0.01020571,
      "balance_loss_clip": 1.04105473,
      "balance_loss_mlp": 1.01320374,
      "epoch": 0.6773282029699994,
      "flos": 22452069571200.0,
      "grad_norm": 2.165433046819387,
      "language_loss": 0.75310647,
      "learning_rate": 9.96179653050422e-07,
      "loss": 0.77500761,
      "num_input_tokens_seen": 121182915,
      "step": 5633,
      "time_per_iteration": 2.6564781665802
    },
    {
      "auxiliary_loss_clip": 0.01173168,
      "auxiliary_loss_mlp": 0.01020146,
      "balance_loss_clip": 1.0447371,
      "balance_loss_mlp": 1.0122366,
      "epoch": 0.6774484458606385,
      "flos": 18693730748160.0,
      "grad_norm": 2.3536638508976373,
      "language_loss": 0.7386657,
      "learning_rate": 9.955059812426635e-07,
      "loss": 0.7605989,
      "num_input_tokens_seen": 121200445,
      "step": 5634,
      "time_per_iteration": 2.715759038925171
    },
    {
      "auxiliary_loss_clip": 0.0117377,
      "auxiliary_loss_mlp": 0.01025141,
      "balance_loss_clip": 1.05215597,
      "balance_loss_mlp": 1.01711178,
      "epoch": 0.6775686887512776,
      "flos": 25994046821760.0,
      "grad_norm": 2.3759441288670997,
      "language_loss": 0.82904589,
      "learning_rate": 9.948324618158493e-07,
      "loss": 0.851035,
      "num_input_tokens_seen": 121220785,
      "step": 5635,
      "time_per_iteration": 2.600069999694824
    },
    {
      "auxiliary_loss_clip": 0.01175542,
      "auxiliary_loss_mlp": 0.01026376,
      "balance_loss_clip": 1.04575384,
      "balance_loss_mlp": 1.01848447,
      "epoch": 0.6776889316419167,
      "flos": 13587987922560.0,
      "grad_norm": 2.791087206536367,
      "language_loss": 0.77762616,
      "learning_rate": 9.941590948721502e-07,
      "loss": 0.7996453,
      "num_input_tokens_seen": 121237985,
      "step": 5636,
      "time_per_iteration": 3.5786736011505127
    },
    {
      "auxiliary_loss_clip": 0.01169862,
      "auxiliary_loss_mlp": 0.01022886,
      "balance_loss_clip": 1.04532099,
      "balance_loss_mlp": 1.01590276,
      "epoch": 0.6778091745325557,
      "flos": 27601121220480.0,
      "grad_norm": 1.6642756327527612,
      "language_loss": 0.76329589,
      "learning_rate": 9.934858805137188e-07,
      "loss": 0.78522336,
      "num_input_tokens_seen": 121258635,
      "step": 5637,
      "time_per_iteration": 2.6397688388824463
    },
    {
      "auxiliary_loss_clip": 0.0116862,
      "auxiliary_loss_mlp": 0.01025798,
      "balance_loss_clip": 1.04671478,
      "balance_loss_mlp": 1.01849926,
      "epoch": 0.6779294174231949,
      "flos": 18734058743040.0,
      "grad_norm": 1.6213703825041976,
      "language_loss": 0.80822223,
      "learning_rate": 9.92812818842677e-07,
      "loss": 0.83016646,
      "num_input_tokens_seen": 121277810,
      "step": 5638,
      "time_per_iteration": 2.6399831771850586
    },
    {
      "auxiliary_loss_clip": 0.01167829,
      "auxiliary_loss_mlp": 0.0102548,
      "balance_loss_clip": 1.04616487,
      "balance_loss_mlp": 1.01774907,
      "epoch": 0.678049660313834,
      "flos": 45873797765760.0,
      "grad_norm": 2.83071046660328,
      "language_loss": 0.64026916,
      "learning_rate": 9.921399099611306e-07,
      "loss": 0.66220218,
      "num_input_tokens_seen": 121298975,
      "step": 5639,
      "time_per_iteration": 2.8490288257598877
    },
    {
      "auxiliary_loss_clip": 0.01174811,
      "auxiliary_loss_mlp": 0.01024605,
      "balance_loss_clip": 1.04508829,
      "balance_loss_mlp": 1.01725233,
      "epoch": 0.678169903204473,
      "flos": 19974556892160.0,
      "grad_norm": 1.618931425398669,
      "language_loss": 0.69477528,
      "learning_rate": 9.914671539711588e-07,
      "loss": 0.71676946,
      "num_input_tokens_seen": 121318495,
      "step": 5640,
      "time_per_iteration": 2.6315829753875732
    },
    {
      "auxiliary_loss_clip": 0.01189391,
      "auxiliary_loss_mlp": 0.01208264,
      "balance_loss_clip": 1.03977597,
      "balance_loss_mlp": 1.00107455,
      "epoch": 0.6782901460951122,
      "flos": 21395613732480.0,
      "grad_norm": 2.7697962038593342,
      "language_loss": 0.7853446,
      "learning_rate": 9.90794550974817e-07,
      "loss": 0.80932117,
      "num_input_tokens_seen": 121338890,
      "step": 5641,
      "time_per_iteration": 3.746890068054199
    },
    {
      "auxiliary_loss_clip": 0.01173686,
      "auxiliary_loss_mlp": 0.01029597,
      "balance_loss_clip": 1.04459906,
      "balance_loss_mlp": 1.02136564,
      "epoch": 0.6784103889857512,
      "flos": 21434002392960.0,
      "grad_norm": 2.5537549850002503,
      "language_loss": 0.81572962,
      "learning_rate": 9.901221010741407e-07,
      "loss": 0.83776248,
      "num_input_tokens_seen": 121358210,
      "step": 5642,
      "time_per_iteration": 2.769724130630493
    },
    {
      "auxiliary_loss_clip": 0.01177793,
      "auxiliary_loss_mlp": 0.01028074,
      "balance_loss_clip": 1.04782867,
      "balance_loss_mlp": 1.02056384,
      "epoch": 0.6785306318763903,
      "flos": 32671923091200.0,
      "grad_norm": 2.2948252046616546,
      "language_loss": 0.74551761,
      "learning_rate": 9.894498043711375e-07,
      "loss": 0.76757622,
      "num_input_tokens_seen": 121379955,
      "step": 5643,
      "time_per_iteration": 2.7093396186828613
    },
    {
      "auxiliary_loss_clip": 0.01173831,
      "auxiliary_loss_mlp": 0.01024176,
      "balance_loss_clip": 1.04492712,
      "balance_loss_mlp": 1.01612353,
      "epoch": 0.6786508747670293,
      "flos": 25632139340160.0,
      "grad_norm": 1.9633624437902866,
      "language_loss": 0.69524932,
      "learning_rate": 9.887776609677962e-07,
      "loss": 0.71722937,
      "num_input_tokens_seen": 121401325,
      "step": 5644,
      "time_per_iteration": 3.605569362640381
    },
    {
      "auxiliary_loss_clip": 0.01167837,
      "auxiliary_loss_mlp": 0.01021052,
      "balance_loss_clip": 1.04032826,
      "balance_loss_mlp": 1.01342845,
      "epoch": 0.6787711176576685,
      "flos": 19171881619200.0,
      "grad_norm": 2.6614603417311846,
      "language_loss": 0.72418356,
      "learning_rate": 9.88105670966079e-07,
      "loss": 0.74607241,
      "num_input_tokens_seen": 121419785,
      "step": 5645,
      "time_per_iteration": 2.6272263526916504
    },
    {
      "auxiliary_loss_clip": 0.01167248,
      "auxiliary_loss_mlp": 0.01021429,
      "balance_loss_clip": 1.04254127,
      "balance_loss_mlp": 1.01430345,
      "epoch": 0.6788913605483076,
      "flos": 13985159581440.0,
      "grad_norm": 2.7039875724541127,
      "language_loss": 0.78758693,
      "learning_rate": 9.874338344679283e-07,
      "loss": 0.80947369,
      "num_input_tokens_seen": 121435630,
      "step": 5646,
      "time_per_iteration": 2.636080741882324
    },
    {
      "auxiliary_loss_clip": 0.0116745,
      "auxiliary_loss_mlp": 0.01028739,
      "balance_loss_clip": 1.04891109,
      "balance_loss_mlp": 1.02141356,
      "epoch": 0.6790116034389466,
      "flos": 22017586659840.0,
      "grad_norm": 2.44061279625836,
      "language_loss": 0.74003232,
      "learning_rate": 9.86762151575259e-07,
      "loss": 0.76199418,
      "num_input_tokens_seen": 121455625,
      "step": 5647,
      "time_per_iteration": 2.6414742469787598
    },
    {
      "auxiliary_loss_clip": 0.01182969,
      "auxiliary_loss_mlp": 0.01207718,
      "balance_loss_clip": 1.04608631,
      "balance_loss_mlp": 1.00113618,
      "epoch": 0.6791318463295858,
      "flos": 20922454851840.0,
      "grad_norm": 1.4145745792942244,
      "language_loss": 0.80067271,
      "learning_rate": 9.860906223899651e-07,
      "loss": 0.8245796,
      "num_input_tokens_seen": 121475020,
      "step": 5648,
      "time_per_iteration": 3.698927640914917
    },
    {
      "auxiliary_loss_clip": 0.01182062,
      "auxiliary_loss_mlp": 0.01028416,
      "balance_loss_clip": 1.04708934,
      "balance_loss_mlp": 1.02091753,
      "epoch": 0.6792520892202248,
      "flos": 28512749422080.0,
      "grad_norm": 1.894516649421875,
      "language_loss": 0.75565815,
      "learning_rate": 9.854192470139184e-07,
      "loss": 0.77776289,
      "num_input_tokens_seen": 121496500,
      "step": 5649,
      "time_per_iteration": 2.704420566558838
    },
    {
      "auxiliary_loss_clip": 0.01176157,
      "auxiliary_loss_mlp": 0.01028133,
      "balance_loss_clip": 1.04862535,
      "balance_loss_mlp": 1.02124596,
      "epoch": 0.6793723321108639,
      "flos": 20011904058240.0,
      "grad_norm": 2.0267461442058607,
      "language_loss": 0.71805394,
      "learning_rate": 9.847480255489645e-07,
      "loss": 0.74009681,
      "num_input_tokens_seen": 121515525,
      "step": 5650,
      "time_per_iteration": 2.603910446166992
    },
    {
      "auxiliary_loss_clip": 0.01178835,
      "auxiliary_loss_mlp": 0.01024048,
      "balance_loss_clip": 1.04572666,
      "balance_loss_mlp": 1.01672578,
      "epoch": 0.6794925750015031,
      "flos": 26649488246400.0,
      "grad_norm": 1.7765475514631688,
      "language_loss": 0.69174671,
      "learning_rate": 9.840769580969295e-07,
      "loss": 0.71377563,
      "num_input_tokens_seen": 121535965,
      "step": 5651,
      "time_per_iteration": 2.6806304454803467
    },
    {
      "auxiliary_loss_clip": 0.01165804,
      "auxiliary_loss_mlp": 0.01024218,
      "balance_loss_clip": 1.0463469,
      "balance_loss_mlp": 1.01645136,
      "epoch": 0.6796128178921421,
      "flos": 21580374314880.0,
      "grad_norm": 1.975493247534401,
      "language_loss": 0.79059011,
      "learning_rate": 9.834060447596114e-07,
      "loss": 0.81249034,
      "num_input_tokens_seen": 121555235,
      "step": 5652,
      "time_per_iteration": 2.6258726119995117
    },
    {
      "auxiliary_loss_clip": 0.01174371,
      "auxiliary_loss_mlp": 0.01025193,
      "balance_loss_clip": 1.04524541,
      "balance_loss_mlp": 1.01721787,
      "epoch": 0.6797330607827812,
      "flos": 22492002516480.0,
      "grad_norm": 2.3746256378211226,
      "language_loss": 0.78327304,
      "learning_rate": 9.827352856387868e-07,
      "loss": 0.80526865,
      "num_input_tokens_seen": 121574945,
      "step": 5653,
      "time_per_iteration": 2.7304110527038574
    },
    {
      "auxiliary_loss_clip": 0.01088863,
      "auxiliary_loss_mlp": 0.01000017,
      "balance_loss_clip": 1.01453292,
      "balance_loss_mlp": 0.99928725,
      "epoch": 0.6798533036734203,
      "flos": 66306648286080.0,
      "grad_norm": 0.7682270263888122,
      "language_loss": 0.64231235,
      "learning_rate": 9.820646808362118e-07,
      "loss": 0.66320115,
      "num_input_tokens_seen": 121641200,
      "step": 5654,
      "time_per_iteration": 3.326605796813965
    },
    {
      "auxiliary_loss_clip": 0.01171957,
      "auxiliary_loss_mlp": 0.01029984,
      "balance_loss_clip": 1.04614007,
      "balance_loss_mlp": 1.02249718,
      "epoch": 0.6799735465640594,
      "flos": 16180163792640.0,
      "grad_norm": 2.2876430640799423,
      "language_loss": 0.72463983,
      "learning_rate": 9.813942304536154e-07,
      "loss": 0.74665922,
      "num_input_tokens_seen": 121659170,
      "step": 5655,
      "time_per_iteration": 2.716301202774048
    },
    {
      "auxiliary_loss_clip": 0.01175743,
      "auxiliary_loss_mlp": 0.01028299,
      "balance_loss_clip": 1.04537368,
      "balance_loss_mlp": 1.02096176,
      "epoch": 0.6800937894546984,
      "flos": 22125749489280.0,
      "grad_norm": 3.624789899896923,
      "language_loss": 0.63567579,
      "learning_rate": 9.807239345927043e-07,
      "loss": 0.65771627,
      "num_input_tokens_seen": 121679180,
      "step": 5656,
      "time_per_iteration": 2.703369379043579
    },
    {
      "auxiliary_loss_clip": 0.01177297,
      "auxiliary_loss_mlp": 0.01024545,
      "balance_loss_clip": 1.04281843,
      "balance_loss_mlp": 1.01662302,
      "epoch": 0.6802140323453376,
      "flos": 31612953300480.0,
      "grad_norm": 2.0853777457592426,
      "language_loss": 0.722018,
      "learning_rate": 9.80053793355162e-07,
      "loss": 0.74403644,
      "num_input_tokens_seen": 121697875,
      "step": 5657,
      "time_per_iteration": 2.7963449954986572
    },
    {
      "auxiliary_loss_clip": 0.01173768,
      "auxiliary_loss_mlp": 0.01031636,
      "balance_loss_clip": 1.0411489,
      "balance_loss_mlp": 1.02348828,
      "epoch": 0.6803342752359767,
      "flos": 17712938908800.0,
      "grad_norm": 2.1723564374831374,
      "language_loss": 0.75088406,
      "learning_rate": 9.793838068426472e-07,
      "loss": 0.77293807,
      "num_input_tokens_seen": 121715570,
      "step": 5658,
      "time_per_iteration": 2.856457471847534
    },
    {
      "auxiliary_loss_clip": 0.01170279,
      "auxiliary_loss_mlp": 0.01025873,
      "balance_loss_clip": 1.0487833,
      "balance_loss_mlp": 1.01813912,
      "epoch": 0.6804545181266157,
      "flos": 11326800902400.0,
      "grad_norm": 2.175716163330187,
      "language_loss": 0.61226833,
      "learning_rate": 9.78713975156799e-07,
      "loss": 0.6342299,
      "num_input_tokens_seen": 121731435,
      "step": 5659,
      "time_per_iteration": 2.572152853012085
    },
    {
      "auxiliary_loss_clip": 0.0117763,
      "auxiliary_loss_mlp": 0.01024627,
      "balance_loss_clip": 1.0469209,
      "balance_loss_mlp": 1.01653826,
      "epoch": 0.6805747610172549,
      "flos": 29350976181120.0,
      "grad_norm": 1.8827338659969757,
      "language_loss": 0.71606708,
      "learning_rate": 9.780442983992273e-07,
      "loss": 0.73808968,
      "num_input_tokens_seen": 121749950,
      "step": 5660,
      "time_per_iteration": 2.7716126441955566
    },
    {
      "auxiliary_loss_clip": 0.01169432,
      "auxiliary_loss_mlp": 0.01024651,
      "balance_loss_clip": 1.04506183,
      "balance_loss_mlp": 1.01691175,
      "epoch": 0.680695003907894,
      "flos": 37631868612480.0,
      "grad_norm": 2.1112325617881518,
      "language_loss": 0.71766937,
      "learning_rate": 9.773747766715238e-07,
      "loss": 0.7396102,
      "num_input_tokens_seen": 121770770,
      "step": 5661,
      "time_per_iteration": 2.924973249435425
    },
    {
      "auxiliary_loss_clip": 0.01177303,
      "auxiliary_loss_mlp": 0.01026696,
      "balance_loss_clip": 1.04470372,
      "balance_loss_mlp": 1.0188756,
      "epoch": 0.680815246798533,
      "flos": 22127365601280.0,
      "grad_norm": 1.8225604899204646,
      "language_loss": 0.79906708,
      "learning_rate": 9.767054100752536e-07,
      "loss": 0.82110709,
      "num_input_tokens_seen": 121790720,
      "step": 5662,
      "time_per_iteration": 2.77006459236145
    },
    {
      "auxiliary_loss_clip": 0.01180325,
      "auxiliary_loss_mlp": 0.01024943,
      "balance_loss_clip": 1.04582536,
      "balance_loss_mlp": 1.01717103,
      "epoch": 0.6809354896891722,
      "flos": 17201822330880.0,
      "grad_norm": 2.1053554110374146,
      "language_loss": 0.81440318,
      "learning_rate": 9.760361987119584e-07,
      "loss": 0.83645582,
      "num_input_tokens_seen": 121808455,
      "step": 5663,
      "time_per_iteration": 3.5856869220733643
    },
    {
      "auxiliary_loss_clip": 0.01176086,
      "auxiliary_loss_mlp": 0.01024923,
      "balance_loss_clip": 1.045681,
      "balance_loss_mlp": 1.01640582,
      "epoch": 0.6810557325798112,
      "flos": 12458166554880.0,
      "grad_norm": 2.3412567421764305,
      "language_loss": 0.67379498,
      "learning_rate": 9.753671426831592e-07,
      "loss": 0.69580513,
      "num_input_tokens_seen": 121824470,
      "step": 5664,
      "time_per_iteration": 2.7524421215057373
    },
    {
      "auxiliary_loss_clip": 0.01166335,
      "auxiliary_loss_mlp": 0.01024677,
      "balance_loss_clip": 1.0446701,
      "balance_loss_mlp": 1.01725888,
      "epoch": 0.6811759754704503,
      "flos": 22156165330560.0,
      "grad_norm": 2.1845502255479587,
      "language_loss": 0.79512799,
      "learning_rate": 9.746982420903483e-07,
      "loss": 0.81703806,
      "num_input_tokens_seen": 121842665,
      "step": 5665,
      "time_per_iteration": 2.7296974658966064
    },
    {
      "auxiliary_loss_clip": 0.01168949,
      "auxiliary_loss_mlp": 0.01022913,
      "balance_loss_clip": 1.0486486,
      "balance_loss_mlp": 1.01583815,
      "epoch": 0.6812962183610894,
      "flos": 17525377065600.0,
      "grad_norm": 1.7183252303167857,
      "language_loss": 0.74584436,
      "learning_rate": 9.740294970349993e-07,
      "loss": 0.76776302,
      "num_input_tokens_seen": 121859080,
      "step": 5666,
      "time_per_iteration": 2.6640708446502686
    },
    {
      "auxiliary_loss_clip": 0.0107912,
      "auxiliary_loss_mlp": 0.01000132,
      "balance_loss_clip": 1.01198864,
      "balance_loss_mlp": 0.99927372,
      "epoch": 0.6814164612517285,
      "flos": 60274480855680.0,
      "grad_norm": 0.8805774842847258,
      "language_loss": 0.60884881,
      "learning_rate": 9.733609076185594e-07,
      "loss": 0.62964129,
      "num_input_tokens_seen": 121915485,
      "step": 5667,
      "time_per_iteration": 3.106837749481201
    },
    {
      "auxiliary_loss_clip": 0.01173314,
      "auxiliary_loss_mlp": 0.0102882,
      "balance_loss_clip": 1.04908669,
      "balance_loss_mlp": 1.02055264,
      "epoch": 0.6815367041423676,
      "flos": 19317750750720.0,
      "grad_norm": 2.154492452372752,
      "language_loss": 0.83815885,
      "learning_rate": 9.72692473942455e-07,
      "loss": 0.8601802,
      "num_input_tokens_seen": 121932710,
      "step": 5668,
      "time_per_iteration": 3.693810224533081
    },
    {
      "auxiliary_loss_clip": 0.011863,
      "auxiliary_loss_mlp": 0.01026315,
      "balance_loss_clip": 1.04800618,
      "balance_loss_mlp": 1.01794684,
      "epoch": 0.6816569470330067,
      "flos": 22161696024960.0,
      "grad_norm": 1.7946495471004187,
      "language_loss": 0.77558935,
      "learning_rate": 9.720241961080849e-07,
      "loss": 0.79771554,
      "num_input_tokens_seen": 121952025,
      "step": 5669,
      "time_per_iteration": 2.776233434677124
    },
    {
      "auxiliary_loss_clip": 0.01168252,
      "auxiliary_loss_mlp": 0.01024417,
      "balance_loss_clip": 1.04663813,
      "balance_loss_mlp": 1.01680875,
      "epoch": 0.6817771899236458,
      "flos": 41463501137280.0,
      "grad_norm": 1.9015541580631843,
      "language_loss": 0.732122,
      "learning_rate": 9.713560742168259e-07,
      "loss": 0.75404871,
      "num_input_tokens_seen": 121974650,
      "step": 5670,
      "time_per_iteration": 2.962700843811035
    },
    {
      "auxiliary_loss_clip": 0.01176289,
      "auxiliary_loss_mlp": 0.01027834,
      "balance_loss_clip": 1.04329109,
      "balance_loss_mlp": 1.020208,
      "epoch": 0.6818974328142848,
      "flos": 21106138026240.0,
      "grad_norm": 2.005314168465442,
      "language_loss": 0.71714407,
      "learning_rate": 9.706881083700333e-07,
      "loss": 0.73918527,
      "num_input_tokens_seen": 121994335,
      "step": 5671,
      "time_per_iteration": 3.6681065559387207
    },
    {
      "auxiliary_loss_clip": 0.01180054,
      "auxiliary_loss_mlp": 0.01026409,
      "balance_loss_clip": 1.04341829,
      "balance_loss_mlp": 1.01849902,
      "epoch": 0.682017675704924,
      "flos": 20441897769600.0,
      "grad_norm": 1.9575185741626502,
      "language_loss": 0.82979536,
      "learning_rate": 9.700202986690357e-07,
      "loss": 0.85186005,
      "num_input_tokens_seen": 122012635,
      "step": 5672,
      "time_per_iteration": 2.8168952465057373
    },
    {
      "auxiliary_loss_clip": 0.01172464,
      "auxiliary_loss_mlp": 0.01208552,
      "balance_loss_clip": 1.04722929,
      "balance_loss_mlp": 1.00101519,
      "epoch": 0.682137918595563,
      "flos": 20044438801920.0,
      "grad_norm": 1.7664161072591265,
      "language_loss": 0.66447079,
      "learning_rate": 9.693526452151413e-07,
      "loss": 0.68828094,
      "num_input_tokens_seen": 122031685,
      "step": 5673,
      "time_per_iteration": 2.644284248352051
    },
    {
      "auxiliary_loss_clip": 0.01183338,
      "auxiliary_loss_mlp": 0.01021783,
      "balance_loss_clip": 1.04412079,
      "balance_loss_mlp": 1.01349807,
      "epoch": 0.6822581614862021,
      "flos": 31684559063040.0,
      "grad_norm": 1.74456728052546,
      "language_loss": 0.7580722,
      "learning_rate": 9.686851481096305e-07,
      "loss": 0.78012341,
      "num_input_tokens_seen": 122052995,
      "step": 5674,
      "time_per_iteration": 3.665011167526245
    },
    {
      "auxiliary_loss_clip": 0.01180424,
      "auxiliary_loss_mlp": 0.01028125,
      "balance_loss_clip": 1.04089713,
      "balance_loss_mlp": 1.02003717,
      "epoch": 0.6823784043768413,
      "flos": 23477570864640.0,
      "grad_norm": 1.9341730351582676,
      "language_loss": 0.71907699,
      "learning_rate": 9.68017807453762e-07,
      "loss": 0.74116254,
      "num_input_tokens_seen": 122071740,
      "step": 5675,
      "time_per_iteration": 2.758312702178955
    },
    {
      "auxiliary_loss_clip": 0.01178125,
      "auxiliary_loss_mlp": 0.01208188,
      "balance_loss_clip": 1.04821467,
      "balance_loss_mlp": 1.00098586,
      "epoch": 0.6824986472674803,
      "flos": 14137134024960.0,
      "grad_norm": 1.885388743304596,
      "language_loss": 0.73628879,
      "learning_rate": 9.673506233487721e-07,
      "loss": 0.76015186,
      "num_input_tokens_seen": 122089705,
      "step": 5676,
      "time_per_iteration": 2.6202170848846436
    },
    {
      "auxiliary_loss_clip": 0.01174554,
      "auxiliary_loss_mlp": 0.01207793,
      "balance_loss_clip": 1.04398811,
      "balance_loss_mlp": 1.0010587,
      "epoch": 0.6826188901581194,
      "flos": 21504997624320.0,
      "grad_norm": 1.816085633257513,
      "language_loss": 0.86077046,
      "learning_rate": 9.666835958958717e-07,
      "loss": 0.88459396,
      "num_input_tokens_seen": 122109025,
      "step": 5677,
      "time_per_iteration": 2.693589925765991
    },
    {
      "auxiliary_loss_clip": 0.01170161,
      "auxiliary_loss_mlp": 0.01022977,
      "balance_loss_clip": 1.04916549,
      "balance_loss_mlp": 1.01540709,
      "epoch": 0.6827391330487584,
      "flos": 20810126044800.0,
      "grad_norm": 1.9085040944507918,
      "language_loss": 0.80275571,
      "learning_rate": 9.660167251962484e-07,
      "loss": 0.82468706,
      "num_input_tokens_seen": 122127385,
      "step": 5678,
      "time_per_iteration": 2.5779285430908203
    },
    {
      "auxiliary_loss_clip": 0.01181095,
      "auxiliary_loss_mlp": 0.01024076,
      "balance_loss_clip": 1.04339397,
      "balance_loss_mlp": 1.017102,
      "epoch": 0.6828593759393976,
      "flos": 21688788539520.0,
      "grad_norm": 1.6069029815129052,
      "language_loss": 0.77773547,
      "learning_rate": 9.653500113510654e-07,
      "loss": 0.79978716,
      "num_input_tokens_seen": 122146500,
      "step": 5679,
      "time_per_iteration": 2.7046802043914795
    },
    {
      "auxiliary_loss_clip": 0.01169638,
      "auxiliary_loss_mlp": 0.01029843,
      "balance_loss_clip": 1.04336381,
      "balance_loss_mlp": 1.02189815,
      "epoch": 0.6829796188300367,
      "flos": 25337707557120.0,
      "grad_norm": 2.653021179386871,
      "language_loss": 0.67263949,
      "learning_rate": 9.646834544614627e-07,
      "loss": 0.69463432,
      "num_input_tokens_seen": 122167000,
      "step": 5680,
      "time_per_iteration": 2.6639866828918457
    },
    {
      "auxiliary_loss_clip": 0.01167438,
      "auxiliary_loss_mlp": 0.01024305,
      "balance_loss_clip": 1.04624271,
      "balance_loss_mlp": 1.01688123,
      "epoch": 0.6830998617206757,
      "flos": 20704800389760.0,
      "grad_norm": 2.6434064566565785,
      "language_loss": 0.76207376,
      "learning_rate": 9.64017054628558e-07,
      "loss": 0.78399116,
      "num_input_tokens_seen": 122185825,
      "step": 5681,
      "time_per_iteration": 2.6749281883239746
    },
    {
      "auxiliary_loss_clip": 0.01180803,
      "auxiliary_loss_mlp": 0.01025749,
      "balance_loss_clip": 1.04047942,
      "balance_loss_mlp": 1.01819408,
      "epoch": 0.6832201046113149,
      "flos": 21726638496000.0,
      "grad_norm": 1.711101459247672,
      "language_loss": 0.79281127,
      "learning_rate": 9.63350811953441e-07,
      "loss": 0.81487679,
      "num_input_tokens_seen": 122206200,
      "step": 5682,
      "time_per_iteration": 2.7249505519866943
    },
    {
      "auxiliary_loss_clip": 0.01178998,
      "auxiliary_loss_mlp": 0.01023419,
      "balance_loss_clip": 1.04300511,
      "balance_loss_mlp": 1.01572704,
      "epoch": 0.6833403475019539,
      "flos": 19536554448000.0,
      "grad_norm": 9.295071496780405,
      "language_loss": 0.70763427,
      "learning_rate": 9.626847265371826e-07,
      "loss": 0.72965848,
      "num_input_tokens_seen": 122225520,
      "step": 5683,
      "time_per_iteration": 2.72379207611084
    },
    {
      "auxiliary_loss_clip": 0.01164907,
      "auxiliary_loss_mlp": 0.01026416,
      "balance_loss_clip": 1.04246545,
      "balance_loss_mlp": 1.01906633,
      "epoch": 0.683460590392593,
      "flos": 19352153001600.0,
      "grad_norm": 2.400376979906626,
      "language_loss": 0.78827989,
      "learning_rate": 9.620187984808262e-07,
      "loss": 0.81019318,
      "num_input_tokens_seen": 122244320,
      "step": 5684,
      "time_per_iteration": 2.7516143321990967
    },
    {
      "auxiliary_loss_clip": 0.01173384,
      "auxiliary_loss_mlp": 0.01207778,
      "balance_loss_clip": 1.04626822,
      "balance_loss_mlp": 1.00099826,
      "epoch": 0.6835808332832322,
      "flos": 23288500650240.0,
      "grad_norm": 2.195101956816533,
      "language_loss": 0.86028963,
      "learning_rate": 9.613530278853919e-07,
      "loss": 0.88410133,
      "num_input_tokens_seen": 122264295,
      "step": 5685,
      "time_per_iteration": 2.6886448860168457
    },
    {
      "auxiliary_loss_clip": 0.01169826,
      "auxiliary_loss_mlp": 0.01021447,
      "balance_loss_clip": 1.04713392,
      "balance_loss_mlp": 1.01420498,
      "epoch": 0.6837010761738712,
      "flos": 21653416621440.0,
      "grad_norm": 2.4565649167369985,
      "language_loss": 0.74323547,
      "learning_rate": 9.60687414851879e-07,
      "loss": 0.76514816,
      "num_input_tokens_seen": 122285300,
      "step": 5686,
      "time_per_iteration": 2.7579874992370605
    },
    {
      "auxiliary_loss_clip": 0.01178026,
      "auxiliary_loss_mlp": 0.01027893,
      "balance_loss_clip": 1.04852295,
      "balance_loss_mlp": 1.02006662,
      "epoch": 0.6838213190645103,
      "flos": 17566387418880.0,
      "grad_norm": 3.887601662901379,
      "language_loss": 0.77988005,
      "learning_rate": 9.600219594812575e-07,
      "loss": 0.80193919,
      "num_input_tokens_seen": 122303240,
      "step": 5687,
      "time_per_iteration": 2.6296515464782715
    },
    {
      "auxiliary_loss_clip": 0.01167533,
      "auxiliary_loss_mlp": 0.01022738,
      "balance_loss_clip": 1.04763365,
      "balance_loss_mlp": 1.01574004,
      "epoch": 0.6839415619551494,
      "flos": 23112538899840.0,
      "grad_norm": 1.847503236005655,
      "language_loss": 0.73202652,
      "learning_rate": 9.593566618744786e-07,
      "loss": 0.7539292,
      "num_input_tokens_seen": 122323390,
      "step": 5688,
      "time_per_iteration": 2.7146570682525635
    },
    {
      "auxiliary_loss_clip": 0.01168846,
      "auxiliary_loss_mlp": 0.01026399,
      "balance_loss_clip": 1.04693675,
      "balance_loss_mlp": 1.01885295,
      "epoch": 0.6840618048457885,
      "flos": 22127868391680.0,
      "grad_norm": 1.9933496250844065,
      "language_loss": 0.73906529,
      "learning_rate": 9.58691522132466e-07,
      "loss": 0.76101768,
      "num_input_tokens_seen": 122342200,
      "step": 5689,
      "time_per_iteration": 2.5626513957977295
    },
    {
      "auxiliary_loss_clip": 0.01181515,
      "auxiliary_loss_mlp": 0.01023183,
      "balance_loss_clip": 1.0478549,
      "balance_loss_mlp": 1.01524985,
      "epoch": 0.6841820477364275,
      "flos": 22015898720640.0,
      "grad_norm": 2.372476119296249,
      "language_loss": 0.84724861,
      "learning_rate": 9.58026540356123e-07,
      "loss": 0.8692956,
      "num_input_tokens_seen": 122360465,
      "step": 5690,
      "time_per_iteration": 3.6078708171844482
    },
    {
      "auxiliary_loss_clip": 0.01173654,
      "auxiliary_loss_mlp": 0.01024283,
      "balance_loss_clip": 1.04526997,
      "balance_loss_mlp": 1.01680279,
      "epoch": 0.6843022906270667,
      "flos": 24900531125760.0,
      "grad_norm": 3.858046669607852,
      "language_loss": 0.86701167,
      "learning_rate": 9.573617166463246e-07,
      "loss": 0.888991,
      "num_input_tokens_seen": 122381680,
      "step": 5691,
      "time_per_iteration": 2.656188488006592
    },
    {
      "auxiliary_loss_clip": 0.01177713,
      "auxiliary_loss_mlp": 0.01023822,
      "balance_loss_clip": 1.04408526,
      "balance_loss_mlp": 1.01701212,
      "epoch": 0.6844225335177058,
      "flos": 19969924037760.0,
      "grad_norm": 2.4272692533161733,
      "language_loss": 0.60348022,
      "learning_rate": 9.56697051103924e-07,
      "loss": 0.62549555,
      "num_input_tokens_seen": 122399120,
      "step": 5692,
      "time_per_iteration": 2.6795451641082764
    },
    {
      "auxiliary_loss_clip": 0.01170541,
      "auxiliary_loss_mlp": 0.01023442,
      "balance_loss_clip": 1.04418349,
      "balance_loss_mlp": 1.01589048,
      "epoch": 0.6845427764083448,
      "flos": 25883334126720.0,
      "grad_norm": 3.620835253553041,
      "language_loss": 0.81469518,
      "learning_rate": 9.560325438297522e-07,
      "loss": 0.83663499,
      "num_input_tokens_seen": 122417430,
      "step": 5693,
      "time_per_iteration": 2.706895112991333
    },
    {
      "auxiliary_loss_clip": 0.0117603,
      "auxiliary_loss_mlp": 0.01023098,
      "balance_loss_clip": 1.04935408,
      "balance_loss_mlp": 1.01588035,
      "epoch": 0.684663019298984,
      "flos": 18880143356160.0,
      "grad_norm": 2.77194607715574,
      "language_loss": 0.86902487,
      "learning_rate": 9.553681949246127e-07,
      "loss": 0.89101619,
      "num_input_tokens_seen": 122435055,
      "step": 5694,
      "time_per_iteration": 2.6762442588806152
    },
    {
      "auxiliary_loss_clip": 0.01181968,
      "auxiliary_loss_mlp": 0.0102602,
      "balance_loss_clip": 1.0451808,
      "balance_loss_mlp": 1.017663,
      "epoch": 0.684783262189623,
      "flos": 54193725302400.0,
      "grad_norm": 2.5916818804248707,
      "language_loss": 0.75637388,
      "learning_rate": 9.547040044892886e-07,
      "loss": 0.77845383,
      "num_input_tokens_seen": 122462570,
      "step": 5695,
      "time_per_iteration": 3.9740922451019287
    },
    {
      "auxiliary_loss_clip": 0.01074893,
      "auxiliary_loss_mlp": 0.00999915,
      "balance_loss_clip": 1.0129962,
      "balance_loss_mlp": 0.9990803,
      "epoch": 0.6849035050802621,
      "flos": 63970264143360.0,
      "grad_norm": 0.8556101667345067,
      "language_loss": 0.6007421,
      "learning_rate": 9.540399726245354e-07,
      "loss": 0.62149012,
      "num_input_tokens_seen": 122519275,
      "step": 5696,
      "time_per_iteration": 3.0865402221679688
    },
    {
      "auxiliary_loss_clip": 0.01170597,
      "auxiliary_loss_mlp": 0.01026397,
      "balance_loss_clip": 1.04456246,
      "balance_loss_mlp": 1.01828456,
      "epoch": 0.6850237479709013,
      "flos": 25224121774080.0,
      "grad_norm": 2.37263762277494,
      "language_loss": 0.69223714,
      "learning_rate": 9.533760994310859e-07,
      "loss": 0.71420705,
      "num_input_tokens_seen": 122539675,
      "step": 5697,
      "time_per_iteration": 3.637530565261841
    },
    {
      "auxiliary_loss_clip": 0.01170086,
      "auxiliary_loss_mlp": 0.01023421,
      "balance_loss_clip": 1.04853773,
      "balance_loss_mlp": 1.01598215,
      "epoch": 0.6851439908615403,
      "flos": 19354128249600.0,
      "grad_norm": 2.418302346013526,
      "language_loss": 0.75165266,
      "learning_rate": 9.527123850096508e-07,
      "loss": 0.7735877,
      "num_input_tokens_seen": 122558035,
      "step": 5698,
      "time_per_iteration": 2.7628047466278076
    },
    {
      "auxiliary_loss_clip": 0.01175283,
      "auxiliary_loss_mlp": 0.01021116,
      "balance_loss_clip": 1.04548359,
      "balance_loss_mlp": 1.01396036,
      "epoch": 0.6852642337521794,
      "flos": 23182133500800.0,
      "grad_norm": 2.0484724955743236,
      "language_loss": 0.7173968,
      "learning_rate": 9.520488294609142e-07,
      "loss": 0.73936081,
      "num_input_tokens_seen": 122576815,
      "step": 5699,
      "time_per_iteration": 2.6540303230285645
    },
    {
      "auxiliary_loss_clip": 0.01096602,
      "auxiliary_loss_mlp": 0.01001026,
      "balance_loss_clip": 1.01753139,
      "balance_loss_mlp": 1.00020921,
      "epoch": 0.6853844766428185,
      "flos": 62647206583680.0,
      "grad_norm": 0.7441734539035115,
      "language_loss": 0.53789675,
      "learning_rate": 9.513854328855368e-07,
      "loss": 0.55887306,
      "num_input_tokens_seen": 122634690,
      "step": 5700,
      "time_per_iteration": 3.1619927883148193
    },
    {
      "auxiliary_loss_clip": 0.01167193,
      "auxiliary_loss_mlp": 0.01025025,
      "balance_loss_clip": 1.04852474,
      "balance_loss_mlp": 1.01776171,
      "epoch": 0.6855047195334576,
      "flos": 23437242869760.0,
      "grad_norm": 1.902451890852341,
      "language_loss": 0.81056774,
      "learning_rate": 9.507221953841558e-07,
      "loss": 0.83248985,
      "num_input_tokens_seen": 122652320,
      "step": 5701,
      "time_per_iteration": 3.5271565914154053
    },
    {
      "auxiliary_loss_clip": 0.01176937,
      "auxiliary_loss_mlp": 0.01024428,
      "balance_loss_clip": 1.05122793,
      "balance_loss_mlp": 1.01658976,
      "epoch": 0.6856249624240967,
      "flos": 20664831530880.0,
      "grad_norm": 1.9199545831913396,
      "language_loss": 0.77752876,
      "learning_rate": 9.500591170573824e-07,
      "loss": 0.79954237,
      "num_input_tokens_seen": 122672340,
      "step": 5702,
      "time_per_iteration": 2.642733097076416
    },
    {
      "auxiliary_loss_clip": 0.01177369,
      "auxiliary_loss_mlp": 0.0102462,
      "balance_loss_clip": 1.04247177,
      "balance_loss_mlp": 1.01708651,
      "epoch": 0.6857452053147358,
      "flos": 17087302794240.0,
      "grad_norm": 2.1575260430486347,
      "language_loss": 0.74070406,
      "learning_rate": 9.493961980058078e-07,
      "loss": 0.76272392,
      "num_input_tokens_seen": 122689935,
      "step": 5703,
      "time_per_iteration": 2.704704761505127
    },
    {
      "auxiliary_loss_clip": 0.01167168,
      "auxiliary_loss_mlp": 0.01023069,
      "balance_loss_clip": 1.03992343,
      "balance_loss_mlp": 1.01565456,
      "epoch": 0.6858654482053749,
      "flos": 30847266057600.0,
      "grad_norm": 2.8792344631818474,
      "language_loss": 0.67716575,
      "learning_rate": 9.48733438329993e-07,
      "loss": 0.69906807,
      "num_input_tokens_seen": 122710200,
      "step": 5704,
      "time_per_iteration": 2.786445379257202
    },
    {
      "auxiliary_loss_clip": 0.01168739,
      "auxiliary_loss_mlp": 0.01207959,
      "balance_loss_clip": 1.04912305,
      "balance_loss_mlp": 1.00109363,
      "epoch": 0.6859856910960139,
      "flos": 28877314510080.0,
      "grad_norm": 1.9074183951529415,
      "language_loss": 0.74544883,
      "learning_rate": 9.480708381304807e-07,
      "loss": 0.76921582,
      "num_input_tokens_seen": 122731495,
      "step": 5705,
      "time_per_iteration": 2.7283525466918945
    },
    {
      "auxiliary_loss_clip": 0.01177108,
      "auxiliary_loss_mlp": 0.01030516,
      "balance_loss_clip": 1.04495728,
      "balance_loss_mlp": 1.0227108,
      "epoch": 0.6861059339866531,
      "flos": 19354523299200.0,
      "grad_norm": 2.503117930142539,
      "language_loss": 0.83545184,
      "learning_rate": 9.474083975077858e-07,
      "loss": 0.85752809,
      "num_input_tokens_seen": 122748620,
      "step": 5706,
      "time_per_iteration": 2.6873300075531006
    },
    {
      "auxiliary_loss_clip": 0.01166055,
      "auxiliary_loss_mlp": 0.01020915,
      "balance_loss_clip": 1.04586565,
      "balance_loss_mlp": 1.01336038,
      "epoch": 0.6862261768772921,
      "flos": 22199976944640.0,
      "grad_norm": 2.4033084906038775,
      "language_loss": 0.8016299,
      "learning_rate": 9.467461165623994e-07,
      "loss": 0.82349968,
      "num_input_tokens_seen": 122767670,
      "step": 5707,
      "time_per_iteration": 2.7434322834014893
    },
    {
      "auxiliary_loss_clip": 0.01175887,
      "auxiliary_loss_mlp": 0.01021965,
      "balance_loss_clip": 1.0471487,
      "balance_loss_mlp": 1.01464009,
      "epoch": 0.6863464197679312,
      "flos": 26285677344000.0,
      "grad_norm": 2.053739500340016,
      "language_loss": 0.79650873,
      "learning_rate": 9.46083995394791e-07,
      "loss": 0.81848729,
      "num_input_tokens_seen": 122785480,
      "step": 5708,
      "time_per_iteration": 2.658914089202881
    },
    {
      "auxiliary_loss_clip": 0.01172985,
      "auxiliary_loss_mlp": 0.01207379,
      "balance_loss_clip": 1.04656601,
      "balance_loss_mlp": 1.00095356,
      "epoch": 0.6864666626585703,
      "flos": 37815228564480.0,
      "grad_norm": 1.9387359894778304,
      "language_loss": 0.63202626,
      "learning_rate": 9.454220341054012e-07,
      "loss": 0.65582991,
      "num_input_tokens_seen": 122810265,
      "step": 5709,
      "time_per_iteration": 2.769490957260132
    },
    {
      "auxiliary_loss_clip": 0.01174692,
      "auxiliary_loss_mlp": 0.0102696,
      "balance_loss_clip": 1.04252946,
      "balance_loss_mlp": 1.01896667,
      "epoch": 0.6865869055492094,
      "flos": 19391152193280.0,
      "grad_norm": 3.3378797147889014,
      "language_loss": 0.80885673,
      "learning_rate": 9.447602327946512e-07,
      "loss": 0.83087325,
      "num_input_tokens_seen": 122828905,
      "step": 5710,
      "time_per_iteration": 2.6978960037231445
    },
    {
      "auxiliary_loss_clip": 0.01174211,
      "auxiliary_loss_mlp": 0.01025379,
      "balance_loss_clip": 1.04510963,
      "balance_loss_mlp": 1.01764548,
      "epoch": 0.6867071484398485,
      "flos": 20375966355840.0,
      "grad_norm": 1.912048405963731,
      "language_loss": 0.76760328,
      "learning_rate": 9.440985915629338e-07,
      "loss": 0.78959918,
      "num_input_tokens_seen": 122846235,
      "step": 5711,
      "time_per_iteration": 2.6127498149871826
    },
    {
      "auxiliary_loss_clip": 0.01169653,
      "auxiliary_loss_mlp": 0.01022277,
      "balance_loss_clip": 1.05003691,
      "balance_loss_mlp": 1.01475501,
      "epoch": 0.6868273913304875,
      "flos": 15889143801600.0,
      "grad_norm": 2.097733571338672,
      "language_loss": 0.72954702,
      "learning_rate": 9.434371105106223e-07,
      "loss": 0.75146627,
      "num_input_tokens_seen": 122863835,
      "step": 5712,
      "time_per_iteration": 2.604764699935913
    },
    {
      "auxiliary_loss_clip": 0.01175383,
      "auxiliary_loss_mlp": 0.01024798,
      "balance_loss_clip": 1.04416108,
      "balance_loss_mlp": 1.01684999,
      "epoch": 0.6869476342211267,
      "flos": 24462492768000.0,
      "grad_norm": 2.313090179928704,
      "language_loss": 0.70840144,
      "learning_rate": 9.427757897380602e-07,
      "loss": 0.73040318,
      "num_input_tokens_seen": 122883235,
      "step": 5713,
      "time_per_iteration": 2.6810965538024902
    },
    {
      "auxiliary_loss_clip": 0.01174614,
      "auxiliary_loss_mlp": 0.01021197,
      "balance_loss_clip": 1.04464102,
      "balance_loss_mlp": 1.01303387,
      "epoch": 0.6870678771117658,
      "flos": 18442571875200.0,
      "grad_norm": 2.564322765298049,
      "language_loss": 0.84878284,
      "learning_rate": 9.421146293455695e-07,
      "loss": 0.87074089,
      "num_input_tokens_seen": 122898975,
      "step": 5714,
      "time_per_iteration": 2.824941635131836
    },
    {
      "auxiliary_loss_clip": 0.01171815,
      "auxiliary_loss_mlp": 0.01024339,
      "balance_loss_clip": 1.04346967,
      "balance_loss_mlp": 1.01685297,
      "epoch": 0.6871881200024048,
      "flos": 22200371994240.0,
      "grad_norm": 1.8577746730665998,
      "language_loss": 0.68510389,
      "learning_rate": 9.414536294334489e-07,
      "loss": 0.70706546,
      "num_input_tokens_seen": 122918995,
      "step": 5715,
      "time_per_iteration": 2.7117204666137695
    },
    {
      "auxiliary_loss_clip": 0.01175374,
      "auxiliary_loss_mlp": 0.01023807,
      "balance_loss_clip": 1.0418433,
      "balance_loss_mlp": 1.01599836,
      "epoch": 0.687308362893044,
      "flos": 22127724737280.0,
      "grad_norm": 2.109024897340779,
      "language_loss": 0.69894791,
      "learning_rate": 9.407927901019708e-07,
      "loss": 0.72093964,
      "num_input_tokens_seen": 122938125,
      "step": 5716,
      "time_per_iteration": 2.7119953632354736
    },
    {
      "auxiliary_loss_clip": 0.01172364,
      "auxiliary_loss_mlp": 0.01023969,
      "balance_loss_clip": 1.0462817,
      "balance_loss_mlp": 1.01659012,
      "epoch": 0.687428605783683,
      "flos": 25040546340480.0,
      "grad_norm": 2.6746090697252773,
      "language_loss": 0.77020741,
      "learning_rate": 9.401321114513854e-07,
      "loss": 0.79217076,
      "num_input_tokens_seen": 122957020,
      "step": 5717,
      "time_per_iteration": 3.653966188430786
    },
    {
      "auxiliary_loss_clip": 0.0117221,
      "auxiliary_loss_mlp": 0.01025754,
      "balance_loss_clip": 1.04926145,
      "balance_loss_mlp": 1.01780915,
      "epoch": 0.6875488486743221,
      "flos": 23770063313280.0,
      "grad_norm": 1.9551418227562152,
      "language_loss": 0.75490332,
      "learning_rate": 9.394715935819155e-07,
      "loss": 0.77688289,
      "num_input_tokens_seen": 122977410,
      "step": 5718,
      "time_per_iteration": 2.6731979846954346
    },
    {
      "auxiliary_loss_clip": 0.01176247,
      "auxiliary_loss_mlp": 0.01028833,
      "balance_loss_clip": 1.04716098,
      "balance_loss_mlp": 1.02135253,
      "epoch": 0.6876690915649613,
      "flos": 25516937445120.0,
      "grad_norm": 2.20155080204963,
      "language_loss": 0.62955469,
      "learning_rate": 9.388112365937608e-07,
      "loss": 0.65160549,
      "num_input_tokens_seen": 122996875,
      "step": 5719,
      "time_per_iteration": 2.7810230255126953
    },
    {
      "auxiliary_loss_clip": 0.01179381,
      "auxiliary_loss_mlp": 0.01023828,
      "balance_loss_clip": 1.04467201,
      "balance_loss_mlp": 1.01600158,
      "epoch": 0.6877893344556003,
      "flos": 19427996568960.0,
      "grad_norm": 2.5063069038868027,
      "language_loss": 0.82568514,
      "learning_rate": 9.381510405870985e-07,
      "loss": 0.84771723,
      "num_input_tokens_seen": 123015890,
      "step": 5720,
      "time_per_iteration": 2.6533560752868652
    },
    {
      "auxiliary_loss_clip": 0.01173325,
      "auxiliary_loss_mlp": 0.01024405,
      "balance_loss_clip": 1.04799092,
      "balance_loss_mlp": 1.01683509,
      "epoch": 0.6879095773462394,
      "flos": 18661303745280.0,
      "grad_norm": 1.9680571120351098,
      "language_loss": 0.7719804,
      "learning_rate": 9.374910056620791e-07,
      "loss": 0.79395765,
      "num_input_tokens_seen": 123034955,
      "step": 5721,
      "time_per_iteration": 2.6443233489990234
    },
    {
      "auxiliary_loss_clip": 0.01176456,
      "auxiliary_loss_mlp": 0.01028503,
      "balance_loss_clip": 1.04882264,
      "balance_loss_mlp": 1.02030194,
      "epoch": 0.6880298202368785,
      "flos": 20883132437760.0,
      "grad_norm": 2.80787905162408,
      "language_loss": 0.81054723,
      "learning_rate": 9.368311319188293e-07,
      "loss": 0.83259678,
      "num_input_tokens_seen": 123052770,
      "step": 5722,
      "time_per_iteration": 3.647432327270508
    },
    {
      "auxiliary_loss_clip": 0.01178191,
      "auxiliary_loss_mlp": 0.01022046,
      "balance_loss_clip": 1.04417038,
      "balance_loss_mlp": 1.01432765,
      "epoch": 0.6881500631275176,
      "flos": 30153292318080.0,
      "grad_norm": 7.467560038010713,
      "language_loss": 0.79372209,
      "learning_rate": 9.361714194574515e-07,
      "loss": 0.81572449,
      "num_input_tokens_seen": 123075105,
      "step": 5723,
      "time_per_iteration": 3.699906587600708
    },
    {
      "auxiliary_loss_clip": 0.01069082,
      "auxiliary_loss_mlp": 0.01002301,
      "balance_loss_clip": 1.01456428,
      "balance_loss_mlp": 1.00150228,
      "epoch": 0.6882703060181566,
      "flos": 66181537215360.0,
      "grad_norm": 0.7332554347814503,
      "language_loss": 0.58288544,
      "learning_rate": 9.355118683780228e-07,
      "loss": 0.60359925,
      "num_input_tokens_seen": 123145175,
      "step": 5724,
      "time_per_iteration": 3.2897770404815674
    },
    {
      "auxiliary_loss_clip": 0.0116858,
      "auxiliary_loss_mlp": 0.01024358,
      "balance_loss_clip": 1.04736614,
      "balance_loss_mlp": 1.01669836,
      "epoch": 0.6883905489087958,
      "flos": 18214646123520.0,
      "grad_norm": 2.3422917177911247,
      "language_loss": 0.79340863,
      "learning_rate": 9.348524787805987e-07,
      "loss": 0.81533802,
      "num_input_tokens_seen": 123160365,
      "step": 5725,
      "time_per_iteration": 2.6877241134643555
    },
    {
      "auxiliary_loss_clip": 0.01179445,
      "auxiliary_loss_mlp": 0.0102463,
      "balance_loss_clip": 1.04082656,
      "balance_loss_mlp": 1.0170995,
      "epoch": 0.6885107917994349,
      "flos": 14056262553600.0,
      "grad_norm": 3.7212790026324702,
      "language_loss": 0.85653728,
      "learning_rate": 9.341932507652053e-07,
      "loss": 0.87857801,
      "num_input_tokens_seen": 123174855,
      "step": 5726,
      "time_per_iteration": 2.6583878993988037
    },
    {
      "auxiliary_loss_clip": 0.01173558,
      "auxiliary_loss_mlp": 0.01028981,
      "balance_loss_clip": 1.04181778,
      "balance_loss_mlp": 1.02049291,
      "epoch": 0.6886310346900739,
      "flos": 28690722334080.0,
      "grad_norm": 1.7406439993773553,
      "language_loss": 0.78754854,
      "learning_rate": 9.335341844318489e-07,
      "loss": 0.80957395,
      "num_input_tokens_seen": 123194995,
      "step": 5727,
      "time_per_iteration": 2.713499069213867
    },
    {
      "auxiliary_loss_clip": 0.01174942,
      "auxiliary_loss_mlp": 0.01025254,
      "balance_loss_clip": 1.04554927,
      "balance_loss_mlp": 1.01753509,
      "epoch": 0.6887512775807131,
      "flos": 24535319592960.0,
      "grad_norm": 2.055766142918326,
      "language_loss": 0.73618722,
      "learning_rate": 9.328752798805091e-07,
      "loss": 0.75818914,
      "num_input_tokens_seen": 123213465,
      "step": 5728,
      "time_per_iteration": 3.5894970893859863
    },
    {
      "auxiliary_loss_clip": 0.01172861,
      "auxiliary_loss_mlp": 0.01026244,
      "balance_loss_clip": 1.04636145,
      "balance_loss_mlp": 1.01871586,
      "epoch": 0.6888715204713521,
      "flos": 22414363269120.0,
      "grad_norm": 3.469066403942271,
      "language_loss": 0.76248831,
      "learning_rate": 9.322165372111399e-07,
      "loss": 0.78447932,
      "num_input_tokens_seen": 123231610,
      "step": 5729,
      "time_per_iteration": 2.727604389190674
    },
    {
      "auxiliary_loss_clip": 0.01176312,
      "auxiliary_loss_mlp": 0.0102519,
      "balance_loss_clip": 1.0461787,
      "balance_loss_mlp": 1.01754284,
      "epoch": 0.6889917633619912,
      "flos": 22054323294720.0,
      "grad_norm": 8.904725999676641,
      "language_loss": 0.75554943,
      "learning_rate": 9.315579565236747e-07,
      "loss": 0.77756441,
      "num_input_tokens_seen": 123250715,
      "step": 5730,
      "time_per_iteration": 2.9273295402526855
    },
    {
      "auxiliary_loss_clip": 0.01171638,
      "auxiliary_loss_mlp": 0.01026509,
      "balance_loss_clip": 1.0473237,
      "balance_loss_mlp": 1.01790178,
      "epoch": 0.6891120062526304,
      "flos": 23949724164480.0,
      "grad_norm": 1.757706248825808,
      "language_loss": 0.74100631,
      "learning_rate": 9.308995379180162e-07,
      "loss": 0.76298773,
      "num_input_tokens_seen": 123270270,
      "step": 5731,
      "time_per_iteration": 2.830173969268799
    },
    {
      "auxiliary_loss_clip": 0.01076618,
      "auxiliary_loss_mlp": 0.0100213,
      "balance_loss_clip": 1.01425838,
      "balance_loss_mlp": 1.00138509,
      "epoch": 0.6892322491432694,
      "flos": 64117354337280.0,
      "grad_norm": 0.7532165361242815,
      "language_loss": 0.59495234,
      "learning_rate": 9.302412814940488e-07,
      "loss": 0.61573994,
      "num_input_tokens_seen": 123333045,
      "step": 5732,
      "time_per_iteration": 3.279832601547241
    },
    {
      "auxiliary_loss_clip": 0.01173405,
      "auxiliary_loss_mlp": 0.01024564,
      "balance_loss_clip": 1.04288328,
      "balance_loss_mlp": 1.01652956,
      "epoch": 0.6893524920339085,
      "flos": 23002436736000.0,
      "grad_norm": 2.1394395612089983,
      "language_loss": 0.71415353,
      "learning_rate": 9.295831873516276e-07,
      "loss": 0.73613322,
      "num_input_tokens_seen": 123352320,
      "step": 5733,
      "time_per_iteration": 2.7402048110961914
    },
    {
      "auxiliary_loss_clip": 0.01169515,
      "auxiliary_loss_mlp": 0.01024291,
      "balance_loss_clip": 1.04913998,
      "balance_loss_mlp": 1.01635218,
      "epoch": 0.6894727349245476,
      "flos": 21396260177280.0,
      "grad_norm": 1.6581477576438142,
      "language_loss": 0.7643773,
      "learning_rate": 9.289252555905873e-07,
      "loss": 0.78631532,
      "num_input_tokens_seen": 123372400,
      "step": 5734,
      "time_per_iteration": 2.8004627227783203
    },
    {
      "auxiliary_loss_clip": 0.0117495,
      "auxiliary_loss_mlp": 0.01026774,
      "balance_loss_clip": 1.04963696,
      "balance_loss_mlp": 1.0189476,
      "epoch": 0.6895929778151867,
      "flos": 19865316654720.0,
      "grad_norm": 3.6945579485397695,
      "language_loss": 0.75644195,
      "learning_rate": 9.282674863107334e-07,
      "loss": 0.77845919,
      "num_input_tokens_seen": 123390215,
      "step": 5735,
      "time_per_iteration": 2.7622158527374268
    },
    {
      "auxiliary_loss_clip": 0.01167472,
      "auxiliary_loss_mlp": 0.01026237,
      "balance_loss_clip": 1.04646182,
      "balance_loss_mlp": 1.01856613,
      "epoch": 0.6897132207058257,
      "flos": 18179166464640.0,
      "grad_norm": 2.173416974069886,
      "language_loss": 0.75677425,
      "learning_rate": 9.276098796118488e-07,
      "loss": 0.77871132,
      "num_input_tokens_seen": 123406700,
      "step": 5736,
      "time_per_iteration": 2.7717597484588623
    },
    {
      "auxiliary_loss_clip": 0.01174703,
      "auxiliary_loss_mlp": 0.01024105,
      "balance_loss_clip": 1.04671144,
      "balance_loss_mlp": 1.01659524,
      "epoch": 0.6898334635964649,
      "flos": 32561641359360.0,
      "grad_norm": 1.8611946613452843,
      "language_loss": 0.66276181,
      "learning_rate": 9.269524355936938e-07,
      "loss": 0.68474984,
      "num_input_tokens_seen": 123429880,
      "step": 5737,
      "time_per_iteration": 2.8348162174224854
    },
    {
      "auxiliary_loss_clip": 0.01167042,
      "auxiliary_loss_mlp": 0.01022673,
      "balance_loss_clip": 1.04157865,
      "balance_loss_mlp": 1.01532352,
      "epoch": 0.689953706487104,
      "flos": 22819004956800.0,
      "grad_norm": 1.7165466752773795,
      "language_loss": 0.85267591,
      "learning_rate": 9.262951543560002e-07,
      "loss": 0.87457311,
      "num_input_tokens_seen": 123449105,
      "step": 5738,
      "time_per_iteration": 2.9213573932647705
    },
    {
      "auxiliary_loss_clip": 0.01173907,
      "auxiliary_loss_mlp": 0.01031454,
      "balance_loss_clip": 1.04787278,
      "balance_loss_mlp": 1.0236156,
      "epoch": 0.690073949377743,
      "flos": 18515362786560.0,
      "grad_norm": 2.3758582314702528,
      "language_loss": 0.86178672,
      "learning_rate": 9.256380359984795e-07,
      "loss": 0.88384032,
      "num_input_tokens_seen": 123466215,
      "step": 5739,
      "time_per_iteration": 2.7325785160064697
    },
    {
      "auxiliary_loss_clip": 0.01185049,
      "auxiliary_loss_mlp": 0.01026674,
      "balance_loss_clip": 1.03887868,
      "balance_loss_mlp": 1.01918769,
      "epoch": 0.6901941922683821,
      "flos": 34857194716800.0,
      "grad_norm": 1.7707658933511796,
      "language_loss": 0.74775857,
      "learning_rate": 9.249810806208139e-07,
      "loss": 0.76987582,
      "num_input_tokens_seen": 123485480,
      "step": 5740,
      "time_per_iteration": 2.9772539138793945
    },
    {
      "auxiliary_loss_clip": 0.01175703,
      "auxiliary_loss_mlp": 0.01207665,
      "balance_loss_clip": 1.03717685,
      "balance_loss_mlp": 1.00109875,
      "epoch": 0.6903144351590212,
      "flos": 16253672976000.0,
      "grad_norm": 2.892246053800956,
      "language_loss": 0.80366445,
      "learning_rate": 9.243242883226627e-07,
      "loss": 0.8274982,
      "num_input_tokens_seen": 123504575,
      "step": 5741,
      "time_per_iteration": 2.822448492050171
    },
    {
      "auxiliary_loss_clip": 0.01175078,
      "auxiliary_loss_mlp": 0.0102572,
      "balance_loss_clip": 1.04351318,
      "balance_loss_mlp": 1.0172503,
      "epoch": 0.6904346780496603,
      "flos": 28035137255040.0,
      "grad_norm": 2.072231755744014,
      "language_loss": 0.69406664,
      "learning_rate": 9.236676592036628e-07,
      "loss": 0.71607459,
      "num_input_tokens_seen": 123524250,
      "step": 5742,
      "time_per_iteration": 2.7866265773773193
    },
    {
      "auxiliary_loss_clip": 0.0117065,
      "auxiliary_loss_mlp": 0.0102428,
      "balance_loss_clip": 1.0464623,
      "balance_loss_mlp": 1.016716,
      "epoch": 0.6905549209402994,
      "flos": 23624266008960.0,
      "grad_norm": 1.8251382320763807,
      "language_loss": 0.73462409,
      "learning_rate": 9.230111933634228e-07,
      "loss": 0.75657338,
      "num_input_tokens_seen": 123545845,
      "step": 5743,
      "time_per_iteration": 2.777632474899292
    },
    {
      "auxiliary_loss_clip": 0.01174761,
      "auxiliary_loss_mlp": 0.01020477,
      "balance_loss_clip": 1.04825974,
      "balance_loss_mlp": 1.01311302,
      "epoch": 0.6906751638309385,
      "flos": 23114945111040.0,
      "grad_norm": 1.5887872464607367,
      "language_loss": 0.8100605,
      "learning_rate": 9.223548909015288e-07,
      "loss": 0.83201289,
      "num_input_tokens_seen": 123567535,
      "step": 5744,
      "time_per_iteration": 3.6727850437164307
    },
    {
      "auxiliary_loss_clip": 0.01173302,
      "auxiliary_loss_mlp": 0.0102789,
      "balance_loss_clip": 1.04089856,
      "balance_loss_mlp": 1.02029335,
      "epoch": 0.6907954067215776,
      "flos": 27305468375040.0,
      "grad_norm": 1.8945787712188245,
      "language_loss": 0.71981639,
      "learning_rate": 9.216987519175407e-07,
      "loss": 0.74182832,
      "num_input_tokens_seen": 123587710,
      "step": 5745,
      "time_per_iteration": 2.8832595348358154
    },
    {
      "auxiliary_loss_clip": 0.01163288,
      "auxiliary_loss_mlp": 0.01021972,
      "balance_loss_clip": 1.04418957,
      "balance_loss_mlp": 1.01450336,
      "epoch": 0.6909156496122166,
      "flos": 21689399070720.0,
      "grad_norm": 1.7439474430589421,
      "language_loss": 0.68413788,
      "learning_rate": 9.210427765109942e-07,
      "loss": 0.70599049,
      "num_input_tokens_seen": 123607385,
      "step": 5746,
      "time_per_iteration": 2.752570867538452
    },
    {
      "auxiliary_loss_clip": 0.0117452,
      "auxiliary_loss_mlp": 0.01026787,
      "balance_loss_clip": 1.04325831,
      "balance_loss_mlp": 1.01832902,
      "epoch": 0.6910358925028558,
      "flos": 22561453463040.0,
      "grad_norm": 2.925076191066301,
      "language_loss": 0.81497121,
      "learning_rate": 9.20386964781402e-07,
      "loss": 0.8369844,
      "num_input_tokens_seen": 123625405,
      "step": 5747,
      "time_per_iteration": 2.8653953075408936
    },
    {
      "auxiliary_loss_clip": 0.01171804,
      "auxiliary_loss_mlp": 0.01024557,
      "balance_loss_clip": 1.04518628,
      "balance_loss_mlp": 1.01704049,
      "epoch": 0.6911561353934949,
      "flos": 22054107813120.0,
      "grad_norm": 1.9720308681400194,
      "language_loss": 0.8425259,
      "learning_rate": 9.197313168282472e-07,
      "loss": 0.8644895,
      "num_input_tokens_seen": 123642850,
      "step": 5748,
      "time_per_iteration": 2.826261043548584
    },
    {
      "auxiliary_loss_clip": 0.01165757,
      "auxiliary_loss_mlp": 0.010247,
      "balance_loss_clip": 1.04223108,
      "balance_loss_mlp": 1.01689136,
      "epoch": 0.6912763782841339,
      "flos": 24206557386240.0,
      "grad_norm": 2.02365548024813,
      "language_loss": 0.72018766,
      "learning_rate": 9.190758327509935e-07,
      "loss": 0.74209225,
      "num_input_tokens_seen": 123661595,
      "step": 5749,
      "time_per_iteration": 3.6684505939483643
    },
    {
      "auxiliary_loss_clip": 0.01097777,
      "auxiliary_loss_mlp": 0.01198715,
      "balance_loss_clip": 1.01399612,
      "balance_loss_mlp": 1.00018716,
      "epoch": 0.6913966211747731,
      "flos": 52329641091840.0,
      "grad_norm": 0.9412335287229472,
      "language_loss": 0.64422911,
      "learning_rate": 9.184205126490767e-07,
      "loss": 0.66719401,
      "num_input_tokens_seen": 123710490,
      "step": 5750,
      "time_per_iteration": 3.9790337085723877
    },
    {
      "auxiliary_loss_clip": 0.01087972,
      "auxiliary_loss_mlp": 0.01198867,
      "balance_loss_clip": 1.01305389,
      "balance_loss_mlp": 1.00011849,
      "epoch": 0.6915168640654121,
      "flos": 66741274851840.0,
      "grad_norm": 1.1077176232524166,
      "language_loss": 0.59663361,
      "learning_rate": 9.177653566219075e-07,
      "loss": 0.61950201,
      "num_input_tokens_seen": 123765215,
      "step": 5751,
      "time_per_iteration": 3.2429866790771484
    },
    {
      "auxiliary_loss_clip": 0.01180662,
      "auxiliary_loss_mlp": 0.01026739,
      "balance_loss_clip": 1.04272103,
      "balance_loss_mlp": 1.01918757,
      "epoch": 0.6916371069560512,
      "flos": 18296523175680.0,
      "grad_norm": 2.845121337323171,
      "language_loss": 0.76276416,
      "learning_rate": 9.171103647688744e-07,
      "loss": 0.7848382,
      "num_input_tokens_seen": 123783955,
      "step": 5752,
      "time_per_iteration": 2.8291943073272705
    },
    {
      "auxiliary_loss_clip": 0.01174198,
      "auxiliary_loss_mlp": 0.010246,
      "balance_loss_clip": 1.03756642,
      "balance_loss_mlp": 1.01774001,
      "epoch": 0.6917573498466904,
      "flos": 19645794685440.0,
      "grad_norm": 2.3327124507303516,
      "language_loss": 0.69159555,
      "learning_rate": 9.164555371893367e-07,
      "loss": 0.71358353,
      "num_input_tokens_seen": 123803885,
      "step": 5753,
      "time_per_iteration": 2.918839931488037
    },
    {
      "auxiliary_loss_clip": 0.01172375,
      "auxiliary_loss_mlp": 0.01207795,
      "balance_loss_clip": 1.04740667,
      "balance_loss_mlp": 1.0009793,
      "epoch": 0.6918775927373294,
      "flos": 14210319985920.0,
      "grad_norm": 1.85882428649007,
      "language_loss": 0.74925685,
      "learning_rate": 9.158008739826333e-07,
      "loss": 0.77305853,
      "num_input_tokens_seen": 123821485,
      "step": 5754,
      "time_per_iteration": 3.7031686305999756
    },
    {
      "auxiliary_loss_clip": 0.01173133,
      "auxiliary_loss_mlp": 0.01024242,
      "balance_loss_clip": 1.04721808,
      "balance_loss_mlp": 1.016523,
      "epoch": 0.6919978356279685,
      "flos": 23985455218560.0,
      "grad_norm": 1.8395190561289345,
      "language_loss": 0.86898911,
      "learning_rate": 9.151463752480744e-07,
      "loss": 0.89096284,
      "num_input_tokens_seen": 123840215,
      "step": 5755,
      "time_per_iteration": 2.7948641777038574
    },
    {
      "auxiliary_loss_clip": 0.01164999,
      "auxiliary_loss_mlp": 0.01026389,
      "balance_loss_clip": 1.04027843,
      "balance_loss_mlp": 1.01877189,
      "epoch": 0.6921180785186076,
      "flos": 23622937205760.0,
      "grad_norm": 1.5739426013647195,
      "language_loss": 0.8032636,
      "learning_rate": 9.144920410849493e-07,
      "loss": 0.82517749,
      "num_input_tokens_seen": 123861450,
      "step": 5756,
      "time_per_iteration": 2.851851224899292
    },
    {
      "auxiliary_loss_clip": 0.01178895,
      "auxiliary_loss_mlp": 0.01027538,
      "balance_loss_clip": 1.04500341,
      "balance_loss_mlp": 1.01971769,
      "epoch": 0.6922383214092467,
      "flos": 21142623265920.0,
      "grad_norm": 2.4210680403231537,
      "language_loss": 0.80707431,
      "learning_rate": 9.138378715925176e-07,
      "loss": 0.82913864,
      "num_input_tokens_seen": 123880545,
      "step": 5757,
      "time_per_iteration": 2.8382315635681152
    },
    {
      "auxiliary_loss_clip": 0.0116757,
      "auxiliary_loss_mlp": 0.01024,
      "balance_loss_clip": 1.042081,
      "balance_loss_mlp": 1.01618564,
      "epoch": 0.6923585642998857,
      "flos": 21470667200640.0,
      "grad_norm": 1.9016042079165156,
      "language_loss": 0.81398135,
      "learning_rate": 9.131838668700167e-07,
      "loss": 0.83589709,
      "num_input_tokens_seen": 123900615,
      "step": 5758,
      "time_per_iteration": 2.9880547523498535
    },
    {
      "auxiliary_loss_clip": 0.01176719,
      "auxiliary_loss_mlp": 0.01023827,
      "balance_loss_clip": 1.04246545,
      "balance_loss_mlp": 1.01629305,
      "epoch": 0.6924788071905249,
      "flos": 21105204272640.0,
      "grad_norm": 1.9741325006433386,
      "language_loss": 0.86550194,
      "learning_rate": 9.125300270166598e-07,
      "loss": 0.88750744,
      "num_input_tokens_seen": 123921220,
      "step": 5759,
      "time_per_iteration": 2.96343731880188
    },
    {
      "auxiliary_loss_clip": 0.01185668,
      "auxiliary_loss_mlp": 0.0102201,
      "balance_loss_clip": 1.0445168,
      "balance_loss_mlp": 1.01400542,
      "epoch": 0.692599050081164,
      "flos": 26250018117120.0,
      "grad_norm": 1.9410735337781067,
      "language_loss": 0.85693187,
      "learning_rate": 9.118763521316324e-07,
      "loss": 0.87900865,
      "num_input_tokens_seen": 123941795,
      "step": 5760,
      "time_per_iteration": 2.841160297393799
    },
    {
      "auxiliary_loss_clip": 0.01167928,
      "auxiliary_loss_mlp": 0.01208374,
      "balance_loss_clip": 1.04532969,
      "balance_loss_mlp": 1.00091624,
      "epoch": 0.692719292971803,
      "flos": 20885215426560.0,
      "grad_norm": 1.834613766734604,
      "language_loss": 0.76144087,
      "learning_rate": 9.112228423140987e-07,
      "loss": 0.78520387,
      "num_input_tokens_seen": 123960715,
      "step": 5761,
      "time_per_iteration": 2.8055460453033447
    },
    {
      "auxiliary_loss_clip": 0.01180595,
      "auxiliary_loss_mlp": 0.01029233,
      "balance_loss_clip": 1.04621589,
      "balance_loss_mlp": 1.02103686,
      "epoch": 0.6928395358624422,
      "flos": 25921938268800.0,
      "grad_norm": 2.3253491137426763,
      "language_loss": 0.86514014,
      "learning_rate": 9.105694976631932e-07,
      "loss": 0.88723844,
      "num_input_tokens_seen": 123978625,
      "step": 5762,
      "time_per_iteration": 2.724992275238037
    },
    {
      "auxiliary_loss_clip": 0.01171051,
      "auxiliary_loss_mlp": 0.01028206,
      "balance_loss_clip": 1.04760885,
      "balance_loss_mlp": 1.02007604,
      "epoch": 0.6929597787530812,
      "flos": 23586559706880.0,
      "grad_norm": 2.3319298963491435,
      "language_loss": 0.7305007,
      "learning_rate": 9.099163182780283e-07,
      "loss": 0.75249326,
      "num_input_tokens_seen": 123996780,
      "step": 5763,
      "time_per_iteration": 2.75357723236084
    },
    {
      "auxiliary_loss_clip": 0.01169988,
      "auxiliary_loss_mlp": 0.01024648,
      "balance_loss_clip": 1.04440737,
      "balance_loss_mlp": 1.01658368,
      "epoch": 0.6930800216437203,
      "flos": 18255656476800.0,
      "grad_norm": 3.0475792799266053,
      "language_loss": 0.49377334,
      "learning_rate": 9.092633042576916e-07,
      "loss": 0.51571971,
      "num_input_tokens_seen": 124014045,
      "step": 5764,
      "time_per_iteration": 2.686694622039795
    },
    {
      "auxiliary_loss_clip": 0.0117149,
      "auxiliary_loss_mlp": 0.01025396,
      "balance_loss_clip": 1.04517043,
      "balance_loss_mlp": 1.01766539,
      "epoch": 0.6932002645343595,
      "flos": 29168621809920.0,
      "grad_norm": 1.8418394488516854,
      "language_loss": 0.56499493,
      "learning_rate": 9.086104557012446e-07,
      "loss": 0.58696377,
      "num_input_tokens_seen": 124034615,
      "step": 5765,
      "time_per_iteration": 3.001025676727295
    },
    {
      "auxiliary_loss_clip": 0.01164101,
      "auxiliary_loss_mlp": 0.01023251,
      "balance_loss_clip": 1.04586434,
      "balance_loss_mlp": 1.01557362,
      "epoch": 0.6933205074249985,
      "flos": 23842746483840.0,
      "grad_norm": 2.4258471991716593,
      "language_loss": 0.65697807,
      "learning_rate": 9.079577727077239e-07,
      "loss": 0.6788516,
      "num_input_tokens_seen": 124053445,
      "step": 5766,
      "time_per_iteration": 2.722304582595825
    },
    {
      "auxiliary_loss_clip": 0.01174861,
      "auxiliary_loss_mlp": 0.01029901,
      "balance_loss_clip": 1.04876804,
      "balance_loss_mlp": 1.02193832,
      "epoch": 0.6934407503156376,
      "flos": 24166696268160.0,
      "grad_norm": 3.2526959506238917,
      "language_loss": 0.71994984,
      "learning_rate": 9.073052553761404e-07,
      "loss": 0.74199748,
      "num_input_tokens_seen": 124072810,
      "step": 5767,
      "time_per_iteration": 2.7590115070343018
    },
    {
      "auxiliary_loss_clip": 0.01183002,
      "auxiliary_loss_mlp": 0.01025959,
      "balance_loss_clip": 1.04274249,
      "balance_loss_mlp": 1.01769173,
      "epoch": 0.6935609932062767,
      "flos": 20631327120000.0,
      "grad_norm": 1.7794216914368959,
      "language_loss": 0.78254032,
      "learning_rate": 9.066529038054805e-07,
      "loss": 0.80462992,
      "num_input_tokens_seen": 124092875,
      "step": 5768,
      "time_per_iteration": 2.804453134536743
    },
    {
      "auxiliary_loss_clip": 0.01172162,
      "auxiliary_loss_mlp": 0.01023603,
      "balance_loss_clip": 1.04496777,
      "balance_loss_mlp": 1.01638484,
      "epoch": 0.6936812360969158,
      "flos": 18254184019200.0,
      "grad_norm": 1.6810189903762738,
      "language_loss": 0.74006116,
      "learning_rate": 9.060007180947071e-07,
      "loss": 0.7620188,
      "num_input_tokens_seen": 124110930,
      "step": 5769,
      "time_per_iteration": 2.76631498336792
    },
    {
      "auxiliary_loss_clip": 0.01179038,
      "auxiliary_loss_mlp": 0.01027819,
      "balance_loss_clip": 1.03841424,
      "balance_loss_mlp": 1.01951623,
      "epoch": 0.6938014789875548,
      "flos": 31317336368640.0,
      "grad_norm": 2.0593696095916134,
      "language_loss": 0.73318124,
      "learning_rate": 9.053486983427534e-07,
      "loss": 0.75524974,
      "num_input_tokens_seen": 124132180,
      "step": 5770,
      "time_per_iteration": 3.7462728023529053
    },
    {
      "auxiliary_loss_clip": 0.01176958,
      "auxiliary_loss_mlp": 0.01027996,
      "balance_loss_clip": 1.04290044,
      "balance_loss_mlp": 1.02021146,
      "epoch": 0.6939217218781939,
      "flos": 17528429721600.0,
      "grad_norm": 2.542681256071468,
      "language_loss": 0.71014535,
      "learning_rate": 9.046968446485326e-07,
      "loss": 0.73219484,
      "num_input_tokens_seen": 124150585,
      "step": 5771,
      "time_per_iteration": 2.6202409267425537
    },
    {
      "auxiliary_loss_clip": 0.01176271,
      "auxiliary_loss_mlp": 0.01030143,
      "balance_loss_clip": 1.04872584,
      "balance_loss_mlp": 1.02161932,
      "epoch": 0.6940419647688331,
      "flos": 18551776199040.0,
      "grad_norm": 2.372658337422011,
      "language_loss": 0.70741194,
      "learning_rate": 9.040451571109295e-07,
      "loss": 0.72947603,
      "num_input_tokens_seen": 124166205,
      "step": 5772,
      "time_per_iteration": 2.589150905609131
    },
    {
      "auxiliary_loss_clip": 0.01084361,
      "auxiliary_loss_mlp": 0.01002352,
      "balance_loss_clip": 1.0162077,
      "balance_loss_mlp": 1.00144613,
      "epoch": 0.6941622076594721,
      "flos": 66926286829440.0,
      "grad_norm": 0.828242414941286,
      "language_loss": 0.60397661,
      "learning_rate": 9.033936358288042e-07,
      "loss": 0.62484372,
      "num_input_tokens_seen": 124219940,
      "step": 5773,
      "time_per_iteration": 3.1347479820251465
    },
    {
      "auxiliary_loss_clip": 0.01171814,
      "auxiliary_loss_mlp": 0.01018774,
      "balance_loss_clip": 1.04910278,
      "balance_loss_mlp": 1.01097143,
      "epoch": 0.6942824505501112,
      "flos": 26578062051840.0,
      "grad_norm": 2.280566279361226,
      "language_loss": 0.82292861,
      "learning_rate": 9.027422809009937e-07,
      "loss": 0.84483445,
      "num_input_tokens_seen": 124239885,
      "step": 5774,
      "time_per_iteration": 2.6237921714782715
    },
    {
      "auxiliary_loss_clip": 0.01173684,
      "auxiliary_loss_mlp": 0.01022154,
      "balance_loss_clip": 1.04512274,
      "balance_loss_mlp": 1.01438141,
      "epoch": 0.6944026934407503,
      "flos": 21248308056960.0,
      "grad_norm": 9.46407320671221,
      "language_loss": 0.8337338,
      "learning_rate": 9.020910924263054e-07,
      "loss": 0.85569209,
      "num_input_tokens_seen": 124258410,
      "step": 5775,
      "time_per_iteration": 3.4999120235443115
    },
    {
      "auxiliary_loss_clip": 0.01083108,
      "auxiliary_loss_mlp": 0.0099972,
      "balance_loss_clip": 1.0159955,
      "balance_loss_mlp": 0.99875468,
      "epoch": 0.6945229363313894,
      "flos": 70677191537280.0,
      "grad_norm": 0.8201180842342632,
      "language_loss": 0.5811311,
      "learning_rate": 9.014400705035261e-07,
      "loss": 0.60195935,
      "num_input_tokens_seen": 124315315,
      "step": 5776,
      "time_per_iteration": 3.3039650917053223
    },
    {
      "auxiliary_loss_clip": 0.01168357,
      "auxiliary_loss_mlp": 0.01022714,
      "balance_loss_clip": 1.04977334,
      "balance_loss_mlp": 1.01524854,
      "epoch": 0.6946431792220285,
      "flos": 18952934267520.0,
      "grad_norm": 2.1217347881248854,
      "language_loss": 0.77040493,
      "learning_rate": 9.00789215231414e-07,
      "loss": 0.7923156,
      "num_input_tokens_seen": 124333710,
      "step": 5777,
      "time_per_iteration": 3.5472769737243652
    },
    {
      "auxiliary_loss_clip": 0.01175877,
      "auxiliary_loss_mlp": 0.01208544,
      "balance_loss_clip": 1.04034078,
      "balance_loss_mlp": 1.00123215,
      "epoch": 0.6947634221126676,
      "flos": 20338834671360.0,
      "grad_norm": 1.8466310754328734,
      "language_loss": 0.82102239,
      "learning_rate": 9.001385267087056e-07,
      "loss": 0.84486657,
      "num_input_tokens_seen": 124352855,
      "step": 5778,
      "time_per_iteration": 2.8060619831085205
    },
    {
      "auxiliary_loss_clip": 0.01174678,
      "auxiliary_loss_mlp": 0.0102383,
      "balance_loss_clip": 1.04724193,
      "balance_loss_mlp": 1.01619446,
      "epoch": 0.6948836650033067,
      "flos": 21833723917440.0,
      "grad_norm": 1.7975379543702243,
      "language_loss": 0.70703936,
      "learning_rate": 8.994880050341072e-07,
      "loss": 0.72902447,
      "num_input_tokens_seen": 124372960,
      "step": 5779,
      "time_per_iteration": 2.7002336978912354
    },
    {
      "auxiliary_loss_clip": 0.01169756,
      "auxiliary_loss_mlp": 0.01032291,
      "balance_loss_clip": 1.04571652,
      "balance_loss_mlp": 1.02460778,
      "epoch": 0.6950039078939457,
      "flos": 23657519024640.0,
      "grad_norm": 1.9141705130635405,
      "language_loss": 0.7744137,
      "learning_rate": 8.988376503063026e-07,
      "loss": 0.79643416,
      "num_input_tokens_seen": 124394220,
      "step": 5780,
      "time_per_iteration": 2.860628128051758
    },
    {
      "auxiliary_loss_clip": 0.01189611,
      "auxiliary_loss_mlp": 0.01024731,
      "balance_loss_clip": 1.04381919,
      "balance_loss_mlp": 1.01641035,
      "epoch": 0.6951241507845849,
      "flos": 21792462168960.0,
      "grad_norm": 2.5303515971626607,
      "language_loss": 0.81501865,
      "learning_rate": 8.981874626239521e-07,
      "loss": 0.83716202,
      "num_input_tokens_seen": 124412795,
      "step": 5781,
      "time_per_iteration": 3.704972505569458
    },
    {
      "auxiliary_loss_clip": 0.01174964,
      "auxiliary_loss_mlp": 0.01032724,
      "balance_loss_clip": 1.04986644,
      "balance_loss_mlp": 1.02452862,
      "epoch": 0.695244393675224,
      "flos": 14647568244480.0,
      "grad_norm": 2.1280916245532295,
      "language_loss": 0.88199401,
      "learning_rate": 8.975374420856872e-07,
      "loss": 0.90407085,
      "num_input_tokens_seen": 124429690,
      "step": 5782,
      "time_per_iteration": 2.860825300216675
    },
    {
      "auxiliary_loss_clip": 0.01166666,
      "auxiliary_loss_mlp": 0.01023673,
      "balance_loss_clip": 1.03992486,
      "balance_loss_mlp": 1.01614833,
      "epoch": 0.695364636565863,
      "flos": 16873203778560.0,
      "grad_norm": 2.5362067406732085,
      "language_loss": 0.72810286,
      "learning_rate": 8.968875887901157e-07,
      "loss": 0.75000632,
      "num_input_tokens_seen": 124447070,
      "step": 5783,
      "time_per_iteration": 2.8857107162475586
    },
    {
      "auxiliary_loss_clip": 0.01173838,
      "auxiliary_loss_mlp": 0.0102541,
      "balance_loss_clip": 1.04244781,
      "balance_loss_mlp": 1.01732135,
      "epoch": 0.6954848794565022,
      "flos": 19354523299200.0,
      "grad_norm": 2.7030973510545446,
      "language_loss": 0.62747586,
      "learning_rate": 8.9623790283582e-07,
      "loss": 0.6494683,
      "num_input_tokens_seen": 124464950,
      "step": 5784,
      "time_per_iteration": 2.7718160152435303
    },
    {
      "auxiliary_loss_clip": 0.01181236,
      "auxiliary_loss_mlp": 0.01029118,
      "balance_loss_clip": 1.04597139,
      "balance_loss_mlp": 1.02089834,
      "epoch": 0.6956051223471412,
      "flos": 18990209606400.0,
      "grad_norm": 2.1493846872217337,
      "language_loss": 0.76321411,
      "learning_rate": 8.955883843213561e-07,
      "loss": 0.78531766,
      "num_input_tokens_seen": 124483965,
      "step": 5785,
      "time_per_iteration": 2.8181047439575195
    },
    {
      "auxiliary_loss_clip": 0.01179196,
      "auxiliary_loss_mlp": 0.01029081,
      "balance_loss_clip": 1.04768896,
      "balance_loss_mlp": 1.02074862,
      "epoch": 0.6957253652377803,
      "flos": 16107229226880.0,
      "grad_norm": 1.990674384633361,
      "language_loss": 0.86964417,
      "learning_rate": 8.949390333452569e-07,
      "loss": 0.89172691,
      "num_input_tokens_seen": 124501910,
      "step": 5786,
      "time_per_iteration": 2.7140660285949707
    },
    {
      "auxiliary_loss_clip": 0.01168832,
      "auxiliary_loss_mlp": 0.01026097,
      "balance_loss_clip": 1.04912543,
      "balance_loss_mlp": 1.01822352,
      "epoch": 0.6958456081284194,
      "flos": 29388646569600.0,
      "grad_norm": 1.6473239101117254,
      "language_loss": 0.67623919,
      "learning_rate": 8.942898500060279e-07,
      "loss": 0.69818848,
      "num_input_tokens_seen": 124521625,
      "step": 5787,
      "time_per_iteration": 2.7897353172302246
    },
    {
      "auxiliary_loss_clip": 0.01187424,
      "auxiliary_loss_mlp": 0.01025267,
      "balance_loss_clip": 1.04365778,
      "balance_loss_mlp": 1.01704144,
      "epoch": 0.6959658510190585,
      "flos": 25154850395520.0,
      "grad_norm": 3.5126162024279988,
      "language_loss": 0.71867841,
      "learning_rate": 8.936408344021493e-07,
      "loss": 0.74080539,
      "num_input_tokens_seen": 124538540,
      "step": 5788,
      "time_per_iteration": 2.8828506469726562
    },
    {
      "auxiliary_loss_clip": 0.01185205,
      "auxiliary_loss_mlp": 0.01028603,
      "balance_loss_clip": 1.04865563,
      "balance_loss_mlp": 1.01952505,
      "epoch": 0.6960860939096976,
      "flos": 42814388759040.0,
      "grad_norm": 2.3481225462778084,
      "language_loss": 0.71019971,
      "learning_rate": 8.929919866320765e-07,
      "loss": 0.73233783,
      "num_input_tokens_seen": 124559355,
      "step": 5789,
      "time_per_iteration": 2.918940305709839
    },
    {
      "auxiliary_loss_clip": 0.01182644,
      "auxiliary_loss_mlp": 0.01208727,
      "balance_loss_clip": 1.04199755,
      "balance_loss_mlp": 1.00091028,
      "epoch": 0.6962063368003367,
      "flos": 17566566986880.0,
      "grad_norm": 1.886923976011268,
      "language_loss": 0.81639099,
      "learning_rate": 8.923433067942385e-07,
      "loss": 0.84030473,
      "num_input_tokens_seen": 124577920,
      "step": 5790,
      "time_per_iteration": 2.752708673477173
    },
    {
      "auxiliary_loss_clip": 0.01185892,
      "auxiliary_loss_mlp": 0.01029974,
      "balance_loss_clip": 1.04554415,
      "balance_loss_mlp": 1.02232659,
      "epoch": 0.6963265796909758,
      "flos": 21251648021760.0,
      "grad_norm": 1.8730975605200155,
      "language_loss": 0.68802595,
      "learning_rate": 8.916947949870417e-07,
      "loss": 0.71018463,
      "num_input_tokens_seen": 124597585,
      "step": 5791,
      "time_per_iteration": 2.7034823894500732
    },
    {
      "auxiliary_loss_clip": 0.01076106,
      "auxiliary_loss_mlp": 0.01001064,
      "balance_loss_clip": 1.01400161,
      "balance_loss_mlp": 1.00015759,
      "epoch": 0.6964468225816148,
      "flos": 68828295801600.0,
      "grad_norm": 0.7431759162034768,
      "language_loss": 0.58110249,
      "learning_rate": 8.910464513088615e-07,
      "loss": 0.60187423,
      "num_input_tokens_seen": 124661625,
      "step": 5792,
      "time_per_iteration": 3.257622003555298
    },
    {
      "auxiliary_loss_clip": 0.01169595,
      "auxiliary_loss_mlp": 0.0102487,
      "balance_loss_clip": 1.04428041,
      "balance_loss_mlp": 1.0161674,
      "epoch": 0.696567065472254,
      "flos": 18950887192320.0,
      "grad_norm": 2.0043625044757047,
      "language_loss": 0.78273261,
      "learning_rate": 8.903982758580542e-07,
      "loss": 0.80467725,
      "num_input_tokens_seen": 124680565,
      "step": 5793,
      "time_per_iteration": 2.6840097904205322
    },
    {
      "auxiliary_loss_clip": 0.01170653,
      "auxiliary_loss_mlp": 0.01031852,
      "balance_loss_clip": 1.04473186,
      "balance_loss_mlp": 1.02381682,
      "epoch": 0.696687308362893,
      "flos": 22856675345280.0,
      "grad_norm": 2.0197411519438604,
      "language_loss": 0.80413306,
      "learning_rate": 8.897502687329457e-07,
      "loss": 0.82615805,
      "num_input_tokens_seen": 124700365,
      "step": 5794,
      "time_per_iteration": 2.7442264556884766
    },
    {
      "auxiliary_loss_clip": 0.01173456,
      "auxiliary_loss_mlp": 0.0102564,
      "balance_loss_clip": 1.04308891,
      "balance_loss_mlp": 1.0183028,
      "epoch": 0.6968075512535321,
      "flos": 24972926987520.0,
      "grad_norm": 2.114292839132203,
      "language_loss": 0.79847157,
      "learning_rate": 8.891024300318382e-07,
      "loss": 0.82046258,
      "num_input_tokens_seen": 124718935,
      "step": 5795,
      "time_per_iteration": 2.793395757675171
    },
    {
      "auxiliary_loss_clip": 0.01167013,
      "auxiliary_loss_mlp": 0.01024344,
      "balance_loss_clip": 1.04078567,
      "balance_loss_mlp": 1.01710153,
      "epoch": 0.6969277941441713,
      "flos": 21030438113280.0,
      "grad_norm": 1.6234861635035687,
      "language_loss": 0.75871778,
      "learning_rate": 8.884547598530103e-07,
      "loss": 0.7806313,
      "num_input_tokens_seen": 124739505,
      "step": 5796,
      "time_per_iteration": 3.6976184844970703
    },
    {
      "auxiliary_loss_clip": 0.01172419,
      "auxiliary_loss_mlp": 0.01028546,
      "balance_loss_clip": 1.03700483,
      "balance_loss_mlp": 1.02056527,
      "epoch": 0.6970480370348103,
      "flos": 21579404647680.0,
      "grad_norm": 1.9801252889831107,
      "language_loss": 0.75653642,
      "learning_rate": 8.8780725829471e-07,
      "loss": 0.77854604,
      "num_input_tokens_seen": 124757410,
      "step": 5797,
      "time_per_iteration": 2.873678684234619
    },
    {
      "auxiliary_loss_clip": 0.0117047,
      "auxiliary_loss_mlp": 0.01027238,
      "balance_loss_clip": 1.04762673,
      "balance_loss_mlp": 1.01904821,
      "epoch": 0.6971682799254494,
      "flos": 22419175691520.0,
      "grad_norm": 2.523647330708066,
      "language_loss": 0.78381521,
      "learning_rate": 8.87159925455165e-07,
      "loss": 0.80579233,
      "num_input_tokens_seen": 124777240,
      "step": 5798,
      "time_per_iteration": 2.7098922729492188
    },
    {
      "auxiliary_loss_clip": 0.01171654,
      "auxiliary_loss_mlp": 0.01024576,
      "balance_loss_clip": 1.0424633,
      "balance_loss_mlp": 1.01709545,
      "epoch": 0.6972885228160886,
      "flos": 20005834659840.0,
      "grad_norm": 2.0667758091815287,
      "language_loss": 0.73197532,
      "learning_rate": 8.865127614325738e-07,
      "loss": 0.7539376,
      "num_input_tokens_seen": 124795670,
      "step": 5799,
      "time_per_iteration": 2.720775842666626
    },
    {
      "auxiliary_loss_clip": 0.01166669,
      "auxiliary_loss_mlp": 0.010312,
      "balance_loss_clip": 1.04231834,
      "balance_loss_mlp": 1.02270055,
      "epoch": 0.6974087657067276,
      "flos": 37853437656960.0,
      "grad_norm": 2.4448226832518745,
      "language_loss": 0.66673338,
      "learning_rate": 8.85865766325113e-07,
      "loss": 0.68871212,
      "num_input_tokens_seen": 124819600,
      "step": 5800,
      "time_per_iteration": 2.801257610321045
    },
    {
      "auxiliary_loss_clip": 0.01172068,
      "auxiliary_loss_mlp": 0.01027656,
      "balance_loss_clip": 1.04500365,
      "balance_loss_mlp": 1.01996708,
      "epoch": 0.6975290085973667,
      "flos": 29489267543040.0,
      "grad_norm": 3.1333270855359467,
      "language_loss": 0.72543144,
      "learning_rate": 8.852189402309287e-07,
      "loss": 0.74742872,
      "num_input_tokens_seen": 124838785,
      "step": 5801,
      "time_per_iteration": 3.6120998859405518
    },
    {
      "auxiliary_loss_clip": 0.01171922,
      "auxiliary_loss_mlp": 0.01025975,
      "balance_loss_clip": 1.04745221,
      "balance_loss_mlp": 1.01858413,
      "epoch": 0.6976492514880057,
      "flos": 12895630295040.0,
      "grad_norm": 2.381397187745272,
      "language_loss": 0.73831844,
      "learning_rate": 8.845722832481441e-07,
      "loss": 0.76029736,
      "num_input_tokens_seen": 124854215,
      "step": 5802,
      "time_per_iteration": 2.6877217292785645
    },
    {
      "auxiliary_loss_clip": 0.01170964,
      "auxiliary_loss_mlp": 0.01027892,
      "balance_loss_clip": 1.0462271,
      "balance_loss_mlp": 1.02021813,
      "epoch": 0.6977694943786449,
      "flos": 24352929308160.0,
      "grad_norm": 2.014988757077649,
      "language_loss": 0.77676916,
      "learning_rate": 8.83925795474858e-07,
      "loss": 0.79875773,
      "num_input_tokens_seen": 124874340,
      "step": 5803,
      "time_per_iteration": 3.730559825897217
    },
    {
      "auxiliary_loss_clip": 0.01173579,
      "auxiliary_loss_mlp": 0.01023401,
      "balance_loss_clip": 1.0447824,
      "balance_loss_mlp": 1.01548886,
      "epoch": 0.6978897372692839,
      "flos": 29898470257920.0,
      "grad_norm": 2.796095463901373,
      "language_loss": 0.59826672,
      "learning_rate": 8.832794770091414e-07,
      "loss": 0.62023658,
      "num_input_tokens_seen": 124895175,
      "step": 5804,
      "time_per_iteration": 2.742208480834961
    },
    {
      "auxiliary_loss_clip": 0.01177161,
      "auxiliary_loss_mlp": 0.01025114,
      "balance_loss_clip": 1.04372191,
      "balance_loss_mlp": 1.01737452,
      "epoch": 0.698009980159923,
      "flos": 21761579450880.0,
      "grad_norm": 2.0399268736838474,
      "language_loss": 0.8262105,
      "learning_rate": 8.826333279490401e-07,
      "loss": 0.84823322,
      "num_input_tokens_seen": 124915810,
      "step": 5805,
      "time_per_iteration": 2.750983715057373
    },
    {
      "auxiliary_loss_clip": 0.01178393,
      "auxiliary_loss_mlp": 0.01025898,
      "balance_loss_clip": 1.04711056,
      "balance_loss_mlp": 1.01870382,
      "epoch": 0.6981302230505622,
      "flos": 19857164267520.0,
      "grad_norm": 2.202671103298434,
      "language_loss": 0.68152523,
      "learning_rate": 8.819873483925748e-07,
      "loss": 0.7035681,
      "num_input_tokens_seen": 124932930,
      "step": 5806,
      "time_per_iteration": 2.6668899059295654
    },
    {
      "auxiliary_loss_clip": 0.01181218,
      "auxiliary_loss_mlp": 0.01207847,
      "balance_loss_clip": 1.04596853,
      "balance_loss_mlp": 1.00095081,
      "epoch": 0.6982504659412012,
      "flos": 22198648141440.0,
      "grad_norm": 2.2026898993138393,
      "language_loss": 0.74628603,
      "learning_rate": 8.81341538437739e-07,
      "loss": 0.77017671,
      "num_input_tokens_seen": 124951220,
      "step": 5807,
      "time_per_iteration": 2.757390260696411
    },
    {
      "auxiliary_loss_clip": 0.01176142,
      "auxiliary_loss_mlp": 0.01019843,
      "balance_loss_clip": 1.04166508,
      "balance_loss_mlp": 1.01208794,
      "epoch": 0.6983707088318403,
      "flos": 35588479708800.0,
      "grad_norm": 1.6474707905748462,
      "language_loss": 0.6820451,
      "learning_rate": 8.80695898182503e-07,
      "loss": 0.704005,
      "num_input_tokens_seen": 124972200,
      "step": 5808,
      "time_per_iteration": 3.711009979248047
    },
    {
      "auxiliary_loss_clip": 0.01069617,
      "auxiliary_loss_mlp": 0.01002146,
      "balance_loss_clip": 1.01609278,
      "balance_loss_mlp": 1.00119257,
      "epoch": 0.6984909517224794,
      "flos": 65440052760960.0,
      "grad_norm": 0.8286868484234744,
      "language_loss": 0.65022981,
      "learning_rate": 8.800504277248093e-07,
      "loss": 0.67094743,
      "num_input_tokens_seen": 125036950,
      "step": 5809,
      "time_per_iteration": 3.2049691677093506
    },
    {
      "auxiliary_loss_clip": 0.01174546,
      "auxiliary_loss_mlp": 0.01207833,
      "balance_loss_clip": 1.04893088,
      "balance_loss_mlp": 1.00104356,
      "epoch": 0.6986111946131185,
      "flos": 18546927863040.0,
      "grad_norm": 1.7600043579792983,
      "language_loss": 0.75350571,
      "learning_rate": 8.794051271625753e-07,
      "loss": 0.7773295,
      "num_input_tokens_seen": 125054585,
      "step": 5810,
      "time_per_iteration": 2.8115270137786865
    },
    {
      "auxiliary_loss_clip": 0.01174453,
      "auxiliary_loss_mlp": 0.01022421,
      "balance_loss_clip": 1.04663134,
      "balance_loss_mlp": 1.01494062,
      "epoch": 0.6987314375037575,
      "flos": 23039173370880.0,
      "grad_norm": 2.284199974330538,
      "language_loss": 0.83318937,
      "learning_rate": 8.787599965936925e-07,
      "loss": 0.85515815,
      "num_input_tokens_seen": 125075515,
      "step": 5811,
      "time_per_iteration": 2.7878952026367188
    },
    {
      "auxiliary_loss_clip": 0.01170974,
      "auxiliary_loss_mlp": 0.01025492,
      "balance_loss_clip": 1.04364204,
      "balance_loss_mlp": 1.01822305,
      "epoch": 0.6988516803943967,
      "flos": 38400393029760.0,
      "grad_norm": 2.451176961918209,
      "language_loss": 0.71987063,
      "learning_rate": 8.781150361160261e-07,
      "loss": 0.7418353,
      "num_input_tokens_seen": 125097425,
      "step": 5812,
      "time_per_iteration": 3.0582056045532227
    },
    {
      "auxiliary_loss_clip": 0.01179672,
      "auxiliary_loss_mlp": 0.01023874,
      "balance_loss_clip": 1.04371989,
      "balance_loss_mlp": 1.01605368,
      "epoch": 0.6989719232850358,
      "flos": 24096993926400.0,
      "grad_norm": 1.8642094650308632,
      "language_loss": 0.73673767,
      "learning_rate": 8.774702458274181e-07,
      "loss": 0.75877309,
      "num_input_tokens_seen": 125117830,
      "step": 5813,
      "time_per_iteration": 2.904447078704834
    },
    {
      "auxiliary_loss_clip": 0.01170901,
      "auxiliary_loss_mlp": 0.0102489,
      "balance_loss_clip": 1.04659319,
      "balance_loss_mlp": 1.01690888,
      "epoch": 0.6990921661756748,
      "flos": 14866838818560.0,
      "grad_norm": 2.705648873189314,
      "language_loss": 0.70944953,
      "learning_rate": 8.768256258256799e-07,
      "loss": 0.7314074,
      "num_input_tokens_seen": 125134455,
      "step": 5814,
      "time_per_iteration": 2.8459713459014893
    },
    {
      "auxiliary_loss_clip": 0.01176459,
      "auxiliary_loss_mlp": 0.0102325,
      "balance_loss_clip": 1.04818392,
      "balance_loss_mlp": 1.01535249,
      "epoch": 0.699212409066314,
      "flos": 20193719725440.0,
      "grad_norm": 1.7903026453638002,
      "language_loss": 0.73926497,
      "learning_rate": 8.76181176208602e-07,
      "loss": 0.76126206,
      "num_input_tokens_seen": 125152555,
      "step": 5815,
      "time_per_iteration": 2.807215690612793
    },
    {
      "auxiliary_loss_clip": 0.01168012,
      "auxiliary_loss_mlp": 0.01031937,
      "balance_loss_clip": 1.0387342,
      "balance_loss_mlp": 1.02360988,
      "epoch": 0.699332651956953,
      "flos": 19427888828160.0,
      "grad_norm": 1.8510744931709031,
      "language_loss": 0.7340399,
      "learning_rate": 8.755368970739461e-07,
      "loss": 0.75603938,
      "num_input_tokens_seen": 125171915,
      "step": 5816,
      "time_per_iteration": 2.932560920715332
    },
    {
      "auxiliary_loss_clip": 0.01180206,
      "auxiliary_loss_mlp": 0.01025625,
      "balance_loss_clip": 1.0414778,
      "balance_loss_mlp": 1.01711345,
      "epoch": 0.6994528948475921,
      "flos": 16143714466560.0,
      "grad_norm": 2.957986731200695,
      "language_loss": 0.61729527,
      "learning_rate": 8.748927885194479e-07,
      "loss": 0.63935363,
      "num_input_tokens_seen": 125190220,
      "step": 5817,
      "time_per_iteration": 2.9585649967193604
    },
    {
      "auxiliary_loss_clip": 0.01082888,
      "auxiliary_loss_mlp": 0.01000337,
      "balance_loss_clip": 1.01803136,
      "balance_loss_mlp": 0.99953824,
      "epoch": 0.6995731377382313,
      "flos": 64952420699520.0,
      "grad_norm": 0.7875730761246895,
      "language_loss": 0.57320523,
      "learning_rate": 8.742488506428209e-07,
      "loss": 0.59403753,
      "num_input_tokens_seen": 125249310,
      "step": 5818,
      "time_per_iteration": 3.31010103225708
    },
    {
      "auxiliary_loss_clip": 0.01176568,
      "auxiliary_loss_mlp": 0.01207826,
      "balance_loss_clip": 1.04483843,
      "balance_loss_mlp": 1.00080395,
      "epoch": 0.6996933806288703,
      "flos": 24900136076160.0,
      "grad_norm": 1.9135600778412747,
      "language_loss": 0.78183031,
      "learning_rate": 8.736050835417466e-07,
      "loss": 0.80567431,
      "num_input_tokens_seen": 125269350,
      "step": 5819,
      "time_per_iteration": 2.798159599304199
    },
    {
      "auxiliary_loss_clip": 0.01175511,
      "auxiliary_loss_mlp": 0.01025996,
      "balance_loss_clip": 1.04779792,
      "balance_loss_mlp": 1.01797938,
      "epoch": 0.6998136235195094,
      "flos": 20777806782720.0,
      "grad_norm": 1.9749748557104574,
      "language_loss": 0.61644304,
      "learning_rate": 8.729614873138862e-07,
      "loss": 0.63845813,
      "num_input_tokens_seen": 125286985,
      "step": 5820,
      "time_per_iteration": 2.8521416187286377
    },
    {
      "auxiliary_loss_clip": 0.01187831,
      "auxiliary_loss_mlp": 0.01024695,
      "balance_loss_clip": 1.04619098,
      "balance_loss_mlp": 1.01642799,
      "epoch": 0.6999338664101485,
      "flos": 23733470332800.0,
      "grad_norm": 2.696679430042357,
      "language_loss": 0.77777058,
      "learning_rate": 8.723180620568716e-07,
      "loss": 0.79989588,
      "num_input_tokens_seen": 125306240,
      "step": 5821,
      "time_per_iteration": 2.8991148471832275
    },
    {
      "auxiliary_loss_clip": 0.011767,
      "auxiliary_loss_mlp": 0.0102386,
      "balance_loss_clip": 1.04384756,
      "balance_loss_mlp": 1.01622462,
      "epoch": 0.7000541093007876,
      "flos": 19864598382720.0,
      "grad_norm": 2.0591172411769714,
      "language_loss": 0.85250592,
      "learning_rate": 8.716748078683116e-07,
      "loss": 0.87451148,
      "num_input_tokens_seen": 125323015,
      "step": 5822,
      "time_per_iteration": 3.731799840927124
    },
    {
      "auxiliary_loss_clip": 0.01171817,
      "auxiliary_loss_mlp": 0.01026718,
      "balance_loss_clip": 1.03640914,
      "balance_loss_mlp": 1.01807547,
      "epoch": 0.7001743521914267,
      "flos": 29679056029440.0,
      "grad_norm": 4.702115625606211,
      "language_loss": 0.68976068,
      "learning_rate": 8.710317248457855e-07,
      "loss": 0.7117461,
      "num_input_tokens_seen": 125342630,
      "step": 5823,
      "time_per_iteration": 2.9903676509857178
    },
    {
      "auxiliary_loss_clip": 0.0117006,
      "auxiliary_loss_mlp": 0.01028169,
      "balance_loss_clip": 1.04523206,
      "balance_loss_mlp": 1.02050066,
      "epoch": 0.7002945950820658,
      "flos": 27489762080640.0,
      "grad_norm": 2.2180707983298786,
      "language_loss": 0.72336358,
      "learning_rate": 8.703888130868482e-07,
      "loss": 0.74534583,
      "num_input_tokens_seen": 125364480,
      "step": 5824,
      "time_per_iteration": 2.9405643939971924
    },
    {
      "auxiliary_loss_clip": 0.01177626,
      "auxiliary_loss_mlp": 0.01024699,
      "balance_loss_clip": 1.04397249,
      "balance_loss_mlp": 1.01737642,
      "epoch": 0.7004148379727049,
      "flos": 22158463800960.0,
      "grad_norm": 2.2002452853653613,
      "language_loss": 0.8214404,
      "learning_rate": 8.697460726890307e-07,
      "loss": 0.84346366,
      "num_input_tokens_seen": 125381625,
      "step": 5825,
      "time_per_iteration": 2.842783212661743
    },
    {
      "auxiliary_loss_clip": 0.01175376,
      "auxiliary_loss_mlp": 0.01207975,
      "balance_loss_clip": 1.04097986,
      "balance_loss_mlp": 1.00086653,
      "epoch": 0.7005350808633439,
      "flos": 19423758764160.0,
      "grad_norm": 2.261157471479151,
      "language_loss": 0.90684777,
      "learning_rate": 8.691035037498354e-07,
      "loss": 0.93068129,
      "num_input_tokens_seen": 125397615,
      "step": 5826,
      "time_per_iteration": 2.974330425262451
    },
    {
      "auxiliary_loss_clip": 0.01170972,
      "auxiliary_loss_mlp": 0.01023417,
      "balance_loss_clip": 1.042629,
      "balance_loss_mlp": 1.01591909,
      "epoch": 0.7006553237539831,
      "flos": 23476708938240.0,
      "grad_norm": 1.9136838913349685,
      "language_loss": 0.72355199,
      "learning_rate": 8.684611063667391e-07,
      "loss": 0.74549592,
      "num_input_tokens_seen": 125418080,
      "step": 5827,
      "time_per_iteration": 2.740628242492676
    },
    {
      "auxiliary_loss_clip": 0.01169965,
      "auxiliary_loss_mlp": 0.01023422,
      "balance_loss_clip": 1.04417586,
      "balance_loss_mlp": 1.0164181,
      "epoch": 0.7007755666446221,
      "flos": 31212872640000.0,
      "grad_norm": 2.0323023888869205,
      "language_loss": 0.77244127,
      "learning_rate": 8.678188806371935e-07,
      "loss": 0.79437512,
      "num_input_tokens_seen": 125440115,
      "step": 5828,
      "time_per_iteration": 3.716289758682251
    },
    {
      "auxiliary_loss_clip": 0.01170449,
      "auxiliary_loss_mlp": 0.01023493,
      "balance_loss_clip": 1.04524088,
      "balance_loss_mlp": 1.01704931,
      "epoch": 0.7008958095352612,
      "flos": 18149899858560.0,
      "grad_norm": 1.6671860176973394,
      "language_loss": 0.85297889,
      "learning_rate": 8.671768266586228e-07,
      "loss": 0.87491828,
      "num_input_tokens_seen": 125458240,
      "step": 5829,
      "time_per_iteration": 2.7399044036865234
    },
    {
      "auxiliary_loss_clip": 0.01173575,
      "auxiliary_loss_mlp": 0.01027274,
      "balance_loss_clip": 1.04210663,
      "balance_loss_mlp": 1.01959074,
      "epoch": 0.7010160524259004,
      "flos": 27452307173760.0,
      "grad_norm": 1.9047537076220629,
      "language_loss": 0.77993393,
      "learning_rate": 8.665349445284275e-07,
      "loss": 0.80194241,
      "num_input_tokens_seen": 125477980,
      "step": 5830,
      "time_per_iteration": 2.8016934394836426
    },
    {
      "auxiliary_loss_clip": 0.01171353,
      "auxiliary_loss_mlp": 0.01021812,
      "balance_loss_clip": 1.04304647,
      "balance_loss_mlp": 1.01422095,
      "epoch": 0.7011362953165394,
      "flos": 23842064125440.0,
      "grad_norm": 1.4492971052278183,
      "language_loss": 0.81006354,
      "learning_rate": 8.658932343439799e-07,
      "loss": 0.83199513,
      "num_input_tokens_seen": 125497765,
      "step": 5831,
      "time_per_iteration": 3.6037566661834717
    },
    {
      "auxiliary_loss_clip": 0.01169702,
      "auxiliary_loss_mlp": 0.01029067,
      "balance_loss_clip": 1.04782999,
      "balance_loss_mlp": 1.02071607,
      "epoch": 0.7012565382071785,
      "flos": 24823430582400.0,
      "grad_norm": 2.5475536975639534,
      "language_loss": 0.77729988,
      "learning_rate": 8.65251696202627e-07,
      "loss": 0.79928756,
      "num_input_tokens_seen": 125514145,
      "step": 5832,
      "time_per_iteration": 2.718411684036255
    },
    {
      "auxiliary_loss_clip": 0.01178165,
      "auxiliary_loss_mlp": 0.0102684,
      "balance_loss_clip": 1.04465306,
      "balance_loss_mlp": 1.01925814,
      "epoch": 0.7013767810978175,
      "flos": 21397445326080.0,
      "grad_norm": 2.417674839315718,
      "language_loss": 0.87855023,
      "learning_rate": 8.646103302016896e-07,
      "loss": 0.90060031,
      "num_input_tokens_seen": 125533115,
      "step": 5833,
      "time_per_iteration": 2.7344751358032227
    },
    {
      "auxiliary_loss_clip": 0.01189395,
      "auxiliary_loss_mlp": 0.01022578,
      "balance_loss_clip": 1.04233861,
      "balance_loss_mlp": 1.01456988,
      "epoch": 0.7014970239884567,
      "flos": 16687150306560.0,
      "grad_norm": 1.880198681174516,
      "language_loss": 0.8873291,
      "learning_rate": 8.639691364384614e-07,
      "loss": 0.9094488,
      "num_input_tokens_seen": 125550740,
      "step": 5834,
      "time_per_iteration": 3.7291629314422607
    },
    {
      "auxiliary_loss_clip": 0.01173649,
      "auxiliary_loss_mlp": 0.01026614,
      "balance_loss_clip": 1.04426658,
      "balance_loss_mlp": 1.01862144,
      "epoch": 0.7016172668790958,
      "flos": 12568268718720.0,
      "grad_norm": 1.9773381632620728,
      "language_loss": 0.73145109,
      "learning_rate": 8.633281150102136e-07,
      "loss": 0.75345373,
      "num_input_tokens_seen": 125567590,
      "step": 5835,
      "time_per_iteration": 2.8671956062316895
    },
    {
      "auxiliary_loss_clip": 0.01172498,
      "auxiliary_loss_mlp": 0.01024302,
      "balance_loss_clip": 1.04536152,
      "balance_loss_mlp": 1.01697373,
      "epoch": 0.7017375097697348,
      "flos": 17452729808640.0,
      "grad_norm": 2.9561791096846743,
      "language_loss": 0.67742413,
      "learning_rate": 8.626872660141855e-07,
      "loss": 0.69939208,
      "num_input_tokens_seen": 125585500,
      "step": 5836,
      "time_per_iteration": 2.7617104053497314
    },
    {
      "auxiliary_loss_clip": 0.0117971,
      "auxiliary_loss_mlp": 0.01026312,
      "balance_loss_clip": 1.04401302,
      "balance_loss_mlp": 1.01861072,
      "epoch": 0.701857752660374,
      "flos": 18513028402560.0,
      "grad_norm": 1.73041910662498,
      "language_loss": 0.75052667,
      "learning_rate": 8.620465895475957e-07,
      "loss": 0.77258682,
      "num_input_tokens_seen": 125603720,
      "step": 5837,
      "time_per_iteration": 2.7438907623291016
    },
    {
      "auxiliary_loss_clip": 0.01174734,
      "auxiliary_loss_mlp": 0.01027707,
      "balance_loss_clip": 1.04274595,
      "balance_loss_mlp": 1.01997089,
      "epoch": 0.701977995551013,
      "flos": 24425971614720.0,
      "grad_norm": 1.4833028004424487,
      "language_loss": 0.75191796,
      "learning_rate": 8.614060857076333e-07,
      "loss": 0.77394235,
      "num_input_tokens_seen": 125624390,
      "step": 5838,
      "time_per_iteration": 2.821814775466919
    },
    {
      "auxiliary_loss_clip": 0.0116992,
      "auxiliary_loss_mlp": 0.01029961,
      "balance_loss_clip": 1.04408765,
      "balance_loss_mlp": 1.02187824,
      "epoch": 0.7020982384416521,
      "flos": 23002759958400.0,
      "grad_norm": 3.2579984167437317,
      "language_loss": 0.74444044,
      "learning_rate": 8.60765754591462e-07,
      "loss": 0.76643926,
      "num_input_tokens_seen": 125644085,
      "step": 5839,
      "time_per_iteration": 2.7503669261932373
    },
    {
      "auxiliary_loss_clip": 0.0116691,
      "auxiliary_loss_mlp": 0.01020849,
      "balance_loss_clip": 1.04692364,
      "balance_loss_mlp": 1.01333284,
      "epoch": 0.7022184813322913,
      "flos": 20449080489600.0,
      "grad_norm": 2.014059967236674,
      "language_loss": 0.73104912,
      "learning_rate": 8.601255962962211e-07,
      "loss": 0.75292671,
      "num_input_tokens_seen": 125663095,
      "step": 5840,
      "time_per_iteration": 2.6543123722076416
    },
    {
      "auxiliary_loss_clip": 0.01184596,
      "auxiliary_loss_mlp": 0.01028163,
      "balance_loss_clip": 1.05112791,
      "balance_loss_mlp": 1.01941276,
      "epoch": 0.7023387242229303,
      "flos": 19790514581760.0,
      "grad_norm": 2.443871207605121,
      "language_loss": 0.72284925,
      "learning_rate": 8.594856109190194e-07,
      "loss": 0.74497682,
      "num_input_tokens_seen": 125680125,
      "step": 5841,
      "time_per_iteration": 2.728081703186035
    },
    {
      "auxiliary_loss_clip": 0.01168166,
      "auxiliary_loss_mlp": 0.01022407,
      "balance_loss_clip": 1.04714108,
      "balance_loss_mlp": 1.01455104,
      "epoch": 0.7024589671135694,
      "flos": 33259278286080.0,
      "grad_norm": 1.8683943353713495,
      "language_loss": 0.69471693,
      "learning_rate": 8.588457985569446e-07,
      "loss": 0.71662271,
      "num_input_tokens_seen": 125703035,
      "step": 5842,
      "time_per_iteration": 2.791759490966797
    },
    {
      "auxiliary_loss_clip": 0.0117204,
      "auxiliary_loss_mlp": 0.01026968,
      "balance_loss_clip": 1.04856861,
      "balance_loss_mlp": 1.01902819,
      "epoch": 0.7025792100042085,
      "flos": 19098982967040.0,
      "grad_norm": 2.3368690459504387,
      "language_loss": 0.71828437,
      "learning_rate": 8.582061593070542e-07,
      "loss": 0.74027443,
      "num_input_tokens_seen": 125723765,
      "step": 5843,
      "time_per_iteration": 2.6841185092926025
    },
    {
      "auxiliary_loss_clip": 0.01169955,
      "auxiliary_loss_mlp": 0.01208028,
      "balance_loss_clip": 1.04799402,
      "balance_loss_mlp": 1.00086832,
      "epoch": 0.7026994528948476,
      "flos": 18952611045120.0,
      "grad_norm": 2.482292047355347,
      "language_loss": 0.76814544,
      "learning_rate": 8.57566693266383e-07,
      "loss": 0.79192531,
      "num_input_tokens_seen": 125741455,
      "step": 5844,
      "time_per_iteration": 2.677258014678955
    },
    {
      "auxiliary_loss_clip": 0.011781,
      "auxiliary_loss_mlp": 0.01208567,
      "balance_loss_clip": 1.04382467,
      "balance_loss_mlp": 1.0008378,
      "epoch": 0.7028196957854866,
      "flos": 19536662188800.0,
      "grad_norm": 1.997661233834702,
      "language_loss": 0.6972338,
      "learning_rate": 8.569274005319354e-07,
      "loss": 0.72110045,
      "num_input_tokens_seen": 125759855,
      "step": 5845,
      "time_per_iteration": 2.665543794631958
    },
    {
      "auxiliary_loss_clip": 0.01167329,
      "auxiliary_loss_mlp": 0.0102442,
      "balance_loss_clip": 1.04502559,
      "balance_loss_mlp": 1.01653087,
      "epoch": 0.7029399386761258,
      "flos": 20845318394880.0,
      "grad_norm": 1.8058663606149583,
      "language_loss": 0.79667389,
      "learning_rate": 8.562882812006913e-07,
      "loss": 0.81859136,
      "num_input_tokens_seen": 125777345,
      "step": 5846,
      "time_per_iteration": 2.714085340499878
    },
    {
      "auxiliary_loss_clip": 0.01167414,
      "auxiliary_loss_mlp": 0.01027662,
      "balance_loss_clip": 1.04729247,
      "balance_loss_mlp": 1.01976395,
      "epoch": 0.7030601815667649,
      "flos": 22055005653120.0,
      "grad_norm": 1.8781172536532063,
      "language_loss": 0.77913642,
      "learning_rate": 8.556493353696066e-07,
      "loss": 0.80108726,
      "num_input_tokens_seen": 125796345,
      "step": 5847,
      "time_per_iteration": 2.848475694656372
    },
    {
      "auxiliary_loss_clip": 0.01176579,
      "auxiliary_loss_mlp": 0.01208654,
      "balance_loss_clip": 1.05007958,
      "balance_loss_mlp": 1.00095904,
      "epoch": 0.7031804244574039,
      "flos": 27198742089600.0,
      "grad_norm": 2.1788575463891244,
      "language_loss": 0.67994857,
      "learning_rate": 8.550105631356077e-07,
      "loss": 0.70380092,
      "num_input_tokens_seen": 125816070,
      "step": 5848,
      "time_per_iteration": 2.8472113609313965
    },
    {
      "auxiliary_loss_clip": 0.01171026,
      "auxiliary_loss_mlp": 0.01027182,
      "balance_loss_clip": 1.04056835,
      "balance_loss_mlp": 1.01890886,
      "epoch": 0.7033006673480431,
      "flos": 22379853277440.0,
      "grad_norm": 1.9694816753954263,
      "language_loss": 0.76955175,
      "learning_rate": 8.543719645955961e-07,
      "loss": 0.79153383,
      "num_input_tokens_seen": 125834400,
      "step": 5849,
      "time_per_iteration": 3.843299627304077
    },
    {
      "auxiliary_loss_clip": 0.01176101,
      "auxiliary_loss_mlp": 0.01024587,
      "balance_loss_clip": 1.04520202,
      "balance_loss_mlp": 1.01693368,
      "epoch": 0.7034209102386821,
      "flos": 24715986024960.0,
      "grad_norm": 1.9775604539807896,
      "language_loss": 0.74683106,
      "learning_rate": 8.537335398464467e-07,
      "loss": 0.76883793,
      "num_input_tokens_seen": 125854720,
      "step": 5850,
      "time_per_iteration": 2.835986852645874
    },
    {
      "auxiliary_loss_clip": 0.01174979,
      "auxiliary_loss_mlp": 0.01029072,
      "balance_loss_clip": 1.04333401,
      "balance_loss_mlp": 1.02121019,
      "epoch": 0.7035411531293212,
      "flos": 22556174163840.0,
      "grad_norm": 3.229142339995015,
      "language_loss": 0.85487795,
      "learning_rate": 8.53095288985007e-07,
      "loss": 0.87691844,
      "num_input_tokens_seen": 125868455,
      "step": 5851,
      "time_per_iteration": 2.8089821338653564
    },
    {
      "auxiliary_loss_clip": 0.01166712,
      "auxiliary_loss_mlp": 0.01021544,
      "balance_loss_clip": 1.0476265,
      "balance_loss_mlp": 1.01397371,
      "epoch": 0.7036613960199604,
      "flos": 22674967418880.0,
      "grad_norm": 1.7197463465763738,
      "language_loss": 0.8244893,
      "learning_rate": 8.524572121081009e-07,
      "loss": 0.84637189,
      "num_input_tokens_seen": 125888555,
      "step": 5852,
      "time_per_iteration": 2.623202323913574
    },
    {
      "auxiliary_loss_clip": 0.0117634,
      "auxiliary_loss_mlp": 0.01025294,
      "balance_loss_clip": 1.04641879,
      "balance_loss_mlp": 1.01772404,
      "epoch": 0.7037816389105994,
      "flos": 22492146170880.0,
      "grad_norm": 2.045197746146452,
      "language_loss": 0.62579274,
      "learning_rate": 8.518193093125232e-07,
      "loss": 0.64780903,
      "num_input_tokens_seen": 125907610,
      "step": 5853,
      "time_per_iteration": 2.74267840385437
    },
    {
      "auxiliary_loss_clip": 0.01179234,
      "auxiliary_loss_mlp": 0.01027121,
      "balance_loss_clip": 1.04673314,
      "balance_loss_mlp": 1.01977754,
      "epoch": 0.7039018818012385,
      "flos": 27087490690560.0,
      "grad_norm": 1.9898931342977515,
      "language_loss": 0.81172889,
      "learning_rate": 8.511815806950436e-07,
      "loss": 0.83379245,
      "num_input_tokens_seen": 125928640,
      "step": 5854,
      "time_per_iteration": 3.8803560733795166
    },
    {
      "auxiliary_loss_clip": 0.01168621,
      "auxiliary_loss_mlp": 0.01023247,
      "balance_loss_clip": 1.04257441,
      "balance_loss_mlp": 1.01565957,
      "epoch": 0.7040221246918776,
      "flos": 17749819198080.0,
      "grad_norm": 1.9284830558990995,
      "language_loss": 0.78063017,
      "learning_rate": 8.505440263524044e-07,
      "loss": 0.80254889,
      "num_input_tokens_seen": 125947485,
      "step": 5855,
      "time_per_iteration": 2.6945302486419678
    },
    {
      "auxiliary_loss_clip": 0.01173032,
      "auxiliary_loss_mlp": 0.01024874,
      "balance_loss_clip": 1.04416132,
      "balance_loss_mlp": 1.01644039,
      "epoch": 0.7041423675825167,
      "flos": 16279851012480.0,
      "grad_norm": 5.800391478373263,
      "language_loss": 0.88232404,
      "learning_rate": 8.49906646381322e-07,
      "loss": 0.90430313,
      "num_input_tokens_seen": 125960320,
      "step": 5856,
      "time_per_iteration": 2.8301281929016113
    },
    {
      "auxiliary_loss_clip": 0.01179,
      "auxiliary_loss_mlp": 0.01022848,
      "balance_loss_clip": 1.04591799,
      "balance_loss_mlp": 1.01569581,
      "epoch": 0.7042626104731557,
      "flos": 25483181639040.0,
      "grad_norm": 1.8276345815800248,
      "language_loss": 0.72145033,
      "learning_rate": 8.492694408784884e-07,
      "loss": 0.74346888,
      "num_input_tokens_seen": 125980575,
      "step": 5857,
      "time_per_iteration": 3.7355797290802
    },
    {
      "auxiliary_loss_clip": 0.01176094,
      "auxiliary_loss_mlp": 0.01024422,
      "balance_loss_clip": 1.04718328,
      "balance_loss_mlp": 1.01697135,
      "epoch": 0.7043828533637949,
      "flos": 17857622891520.0,
      "grad_norm": 3.766813139605459,
      "language_loss": 0.6177755,
      "learning_rate": 8.486324099405642e-07,
      "loss": 0.63978064,
      "num_input_tokens_seen": 125997420,
      "step": 5858,
      "time_per_iteration": 2.7209312915802
    },
    {
      "auxiliary_loss_clip": 0.01169916,
      "auxiliary_loss_mlp": 0.01025126,
      "balance_loss_clip": 1.04521167,
      "balance_loss_mlp": 1.01780081,
      "epoch": 0.704503096254434,
      "flos": 29494259533440.0,
      "grad_norm": 1.62376557277304,
      "language_loss": 0.75105047,
      "learning_rate": 8.479955536641887e-07,
      "loss": 0.7730009,
      "num_input_tokens_seen": 126018915,
      "step": 5859,
      "time_per_iteration": 2.987940788269043
    },
    {
      "auxiliary_loss_clip": 0.01165957,
      "auxiliary_loss_mlp": 0.01025427,
      "balance_loss_clip": 1.03971148,
      "balance_loss_mlp": 1.01799405,
      "epoch": 0.704623339145073,
      "flos": 30920739327360.0,
      "grad_norm": 2.1802857235129656,
      "language_loss": 0.66514134,
      "learning_rate": 8.473588721459716e-07,
      "loss": 0.68705523,
      "num_input_tokens_seen": 126038825,
      "step": 5860,
      "time_per_iteration": 2.733267307281494
    },
    {
      "auxiliary_loss_clip": 0.01177021,
      "auxiliary_loss_mlp": 0.01035218,
      "balance_loss_clip": 1.05027032,
      "balance_loss_mlp": 1.02642632,
      "epoch": 0.7047435820357122,
      "flos": 23914747296000.0,
      "grad_norm": 2.181542213836085,
      "language_loss": 0.70401752,
      "learning_rate": 8.467223654824967e-07,
      "loss": 0.72613996,
      "num_input_tokens_seen": 126058280,
      "step": 5861,
      "time_per_iteration": 3.6512303352355957
    },
    {
      "auxiliary_loss_clip": 0.0116533,
      "auxiliary_loss_mlp": 0.01026103,
      "balance_loss_clip": 1.04509401,
      "balance_loss_mlp": 1.01830673,
      "epoch": 0.7048638249263512,
      "flos": 46494010926720.0,
      "grad_norm": 2.569167488693971,
      "language_loss": 0.62486315,
      "learning_rate": 8.460860337703233e-07,
      "loss": 0.64677751,
      "num_input_tokens_seen": 126078885,
      "step": 5862,
      "time_per_iteration": 2.8893141746520996
    },
    {
      "auxiliary_loss_clip": 0.0116498,
      "auxiliary_loss_mlp": 0.0102975,
      "balance_loss_clip": 1.04114151,
      "balance_loss_mlp": 1.02126563,
      "epoch": 0.7049840678169903,
      "flos": 21689219502720.0,
      "grad_norm": 2.226836358921413,
      "language_loss": 0.70606083,
      "learning_rate": 8.454498771059797e-07,
      "loss": 0.72800815,
      "num_input_tokens_seen": 126098260,
      "step": 5863,
      "time_per_iteration": 2.8234777450561523
    },
    {
      "auxiliary_loss_clip": 0.01173274,
      "auxiliary_loss_mlp": 0.01024161,
      "balance_loss_clip": 1.04083455,
      "balance_loss_mlp": 1.01587558,
      "epoch": 0.7051043107076294,
      "flos": 18405081054720.0,
      "grad_norm": 2.194264718918664,
      "language_loss": 0.83488989,
      "learning_rate": 8.448138955859725e-07,
      "loss": 0.85686427,
      "num_input_tokens_seen": 126114845,
      "step": 5864,
      "time_per_iteration": 2.761582612991333
    },
    {
      "auxiliary_loss_clip": 0.01176653,
      "auxiliary_loss_mlp": 0.01025406,
      "balance_loss_clip": 1.04595447,
      "balance_loss_mlp": 1.01749003,
      "epoch": 0.7052245535982685,
      "flos": 19319043640320.0,
      "grad_norm": 1.9505374757171619,
      "language_loss": 0.9044441,
      "learning_rate": 8.44178089306778e-07,
      "loss": 0.92646468,
      "num_input_tokens_seen": 126132780,
      "step": 5865,
      "time_per_iteration": 2.784038782119751
    },
    {
      "auxiliary_loss_clip": 0.0116914,
      "auxiliary_loss_mlp": 0.01022334,
      "balance_loss_clip": 1.04826248,
      "balance_loss_mlp": 1.01493132,
      "epoch": 0.7053447964889076,
      "flos": 19062138591360.0,
      "grad_norm": 2.1366620520500628,
      "language_loss": 0.766366,
      "learning_rate": 8.4354245836485e-07,
      "loss": 0.78828079,
      "num_input_tokens_seen": 126151225,
      "step": 5866,
      "time_per_iteration": 2.7289693355560303
    },
    {
      "auxiliary_loss_clip": 0.01179264,
      "auxiliary_loss_mlp": 0.01028723,
      "balance_loss_clip": 1.04457259,
      "balance_loss_mlp": 1.0197053,
      "epoch": 0.7054650393795466,
      "flos": 27379228953600.0,
      "grad_norm": 1.5671857718414215,
      "language_loss": 0.7294749,
      "learning_rate": 8.429070028566108e-07,
      "loss": 0.75155473,
      "num_input_tokens_seen": 126172535,
      "step": 5867,
      "time_per_iteration": 2.7587027549743652
    },
    {
      "auxiliary_loss_clip": 0.0117032,
      "auxiliary_loss_mlp": 0.0102959,
      "balance_loss_clip": 1.0465498,
      "balance_loss_mlp": 1.02148414,
      "epoch": 0.7055852822701858,
      "flos": 16102201322880.0,
      "grad_norm": 2.2125989461453672,
      "language_loss": 0.74727523,
      "learning_rate": 8.422717228784586e-07,
      "loss": 0.76927435,
      "num_input_tokens_seen": 126189410,
      "step": 5868,
      "time_per_iteration": 2.7228593826293945
    },
    {
      "auxiliary_loss_clip": 0.01179513,
      "auxiliary_loss_mlp": 0.01026265,
      "balance_loss_clip": 1.04537928,
      "balance_loss_mlp": 1.01812935,
      "epoch": 0.7057055251608249,
      "flos": 11692299744000.0,
      "grad_norm": 2.106924628228815,
      "language_loss": 0.69626677,
      "learning_rate": 8.416366185267663e-07,
      "loss": 0.71832454,
      "num_input_tokens_seen": 126206910,
      "step": 5869,
      "time_per_iteration": 2.6998043060302734
    },
    {
      "auxiliary_loss_clip": 0.01170264,
      "auxiliary_loss_mlp": 0.01025177,
      "balance_loss_clip": 1.04375815,
      "balance_loss_mlp": 1.01751137,
      "epoch": 0.7058257680514639,
      "flos": 22711560399360.0,
      "grad_norm": 1.9237062253024442,
      "language_loss": 0.77691889,
      "learning_rate": 8.410016898978778e-07,
      "loss": 0.79887331,
      "num_input_tokens_seen": 126224385,
      "step": 5870,
      "time_per_iteration": 2.641090154647827
    },
    {
      "auxiliary_loss_clip": 0.01176923,
      "auxiliary_loss_mlp": 0.01022778,
      "balance_loss_clip": 1.04220545,
      "balance_loss_mlp": 1.01521158,
      "epoch": 0.7059460109421031,
      "flos": 17529543043200.0,
      "grad_norm": 1.749045294043927,
      "language_loss": 0.79111207,
      "learning_rate": 8.403669370881115e-07,
      "loss": 0.81310904,
      "num_input_tokens_seen": 126243120,
      "step": 5871,
      "time_per_iteration": 2.660618782043457
    },
    {
      "auxiliary_loss_clip": 0.01170828,
      "auxiliary_loss_mlp": 0.01027369,
      "balance_loss_clip": 1.04915667,
      "balance_loss_mlp": 1.02024341,
      "epoch": 0.7060662538327421,
      "flos": 23544687427200.0,
      "grad_norm": 1.9876492389455838,
      "language_loss": 0.78177464,
      "learning_rate": 8.397323601937587e-07,
      "loss": 0.80375671,
      "num_input_tokens_seen": 126263020,
      "step": 5872,
      "time_per_iteration": 2.624544620513916
    },
    {
      "auxiliary_loss_clip": 0.01173385,
      "auxiliary_loss_mlp": 0.01026442,
      "balance_loss_clip": 1.0438714,
      "balance_loss_mlp": 1.01906598,
      "epoch": 0.7061864967233812,
      "flos": 30260736875520.0,
      "grad_norm": 2.3239585837473857,
      "language_loss": 0.77316409,
      "learning_rate": 8.390979593110838e-07,
      "loss": 0.79516232,
      "num_input_tokens_seen": 126285150,
      "step": 5873,
      "time_per_iteration": 2.706451892852783
    },
    {
      "auxiliary_loss_clip": 0.01181288,
      "auxiliary_loss_mlp": 0.01024776,
      "balance_loss_clip": 1.04855204,
      "balance_loss_mlp": 1.01641321,
      "epoch": 0.7063067396140204,
      "flos": 20701460424960.0,
      "grad_norm": 2.6637969285117933,
      "language_loss": 0.81823534,
      "learning_rate": 8.384637345363262e-07,
      "loss": 0.84029603,
      "num_input_tokens_seen": 126304340,
      "step": 5874,
      "time_per_iteration": 2.6407358646392822
    },
    {
      "auxiliary_loss_clip": 0.0116814,
      "auxiliary_loss_mlp": 0.0102461,
      "balance_loss_clip": 1.04225945,
      "balance_loss_mlp": 1.01676583,
      "epoch": 0.7064269825046594,
      "flos": 32266168081920.0,
      "grad_norm": 1.8725076306127364,
      "language_loss": 0.76861227,
      "learning_rate": 8.378296859656964e-07,
      "loss": 0.79053974,
      "num_input_tokens_seen": 126325495,
      "step": 5875,
      "time_per_iteration": 3.694545030593872
    },
    {
      "auxiliary_loss_clip": 0.01175029,
      "auxiliary_loss_mlp": 0.01030086,
      "balance_loss_clip": 1.04440665,
      "balance_loss_mlp": 1.0221169,
      "epoch": 0.7065472253952985,
      "flos": 30227124723840.0,
      "grad_norm": 4.297901679728801,
      "language_loss": 0.67954218,
      "learning_rate": 8.371958136953792e-07,
      "loss": 0.7015934,
      "num_input_tokens_seen": 126345525,
      "step": 5876,
      "time_per_iteration": 2.6635963916778564
    },
    {
      "auxiliary_loss_clip": 0.01181637,
      "auxiliary_loss_mlp": 0.01029699,
      "balance_loss_clip": 1.04268074,
      "balance_loss_mlp": 1.02112532,
      "epoch": 0.7066674682859376,
      "flos": 16216720859520.0,
      "grad_norm": 2.4543492886669123,
      "language_loss": 0.66218245,
      "learning_rate": 8.365621178215326e-07,
      "loss": 0.68429583,
      "num_input_tokens_seen": 126361995,
      "step": 5877,
      "time_per_iteration": 2.7472665309906006
    },
    {
      "auxiliary_loss_clip": 0.01167437,
      "auxiliary_loss_mlp": 0.01027499,
      "balance_loss_clip": 1.04576921,
      "balance_loss_mlp": 1.02024508,
      "epoch": 0.7067877111765767,
      "flos": 14830461319680.0,
      "grad_norm": 1.8739003834186099,
      "language_loss": 0.75170481,
      "learning_rate": 8.359285984402871e-07,
      "loss": 0.77365422,
      "num_input_tokens_seen": 126379260,
      "step": 5878,
      "time_per_iteration": 2.8091118335723877
    },
    {
      "auxiliary_loss_clip": 0.01167675,
      "auxiliary_loss_mlp": 0.01025617,
      "balance_loss_clip": 1.04415488,
      "balance_loss_mlp": 1.01825523,
      "epoch": 0.7069079540672157,
      "flos": 25440196037760.0,
      "grad_norm": 2.2560136542758236,
      "language_loss": 0.74124694,
      "learning_rate": 8.352952556477489e-07,
      "loss": 0.76317978,
      "num_input_tokens_seen": 126397170,
      "step": 5879,
      "time_per_iteration": 2.6962473392486572
    },
    {
      "auxiliary_loss_clip": 0.01169983,
      "auxiliary_loss_mlp": 0.01024854,
      "balance_loss_clip": 1.04615605,
      "balance_loss_mlp": 1.01715863,
      "epoch": 0.7070281969578549,
      "flos": 24607751368320.0,
      "grad_norm": 1.859164709194372,
      "language_loss": 0.76758981,
      "learning_rate": 8.34662089539993e-07,
      "loss": 0.7895382,
      "num_input_tokens_seen": 126416680,
      "step": 5880,
      "time_per_iteration": 2.7071261405944824
    },
    {
      "auxiliary_loss_clip": 0.01165586,
      "auxiliary_loss_mlp": 0.01023461,
      "balance_loss_clip": 1.04686248,
      "balance_loss_mlp": 1.01589119,
      "epoch": 0.707148439848494,
      "flos": 26724469887360.0,
      "grad_norm": 2.1035035611815682,
      "language_loss": 0.7900666,
      "learning_rate": 8.340291002130722e-07,
      "loss": 0.81195706,
      "num_input_tokens_seen": 126435870,
      "step": 5881,
      "time_per_iteration": 3.747410535812378
    },
    {
      "auxiliary_loss_clip": 0.01171101,
      "auxiliary_loss_mlp": 0.01027217,
      "balance_loss_clip": 1.04810297,
      "balance_loss_mlp": 1.01884818,
      "epoch": 0.707268682739133,
      "flos": 15085750256640.0,
      "grad_norm": 4.517400411439231,
      "language_loss": 0.79727346,
      "learning_rate": 8.3339628776301e-07,
      "loss": 0.8192566,
      "num_input_tokens_seen": 126454010,
      "step": 5882,
      "time_per_iteration": 2.6250617504119873
    },
    {
      "auxiliary_loss_clip": 0.01167115,
      "auxiliary_loss_mlp": 0.0102891,
      "balance_loss_clip": 1.04605675,
      "balance_loss_mlp": 1.02139735,
      "epoch": 0.7073889256297722,
      "flos": 34313148345600.0,
      "grad_norm": 1.813809971845574,
      "language_loss": 0.56663173,
      "learning_rate": 8.327636522858033e-07,
      "loss": 0.58859199,
      "num_input_tokens_seen": 126473615,
      "step": 5883,
      "time_per_iteration": 2.716097354888916
    },
    {
      "auxiliary_loss_clip": 0.0118268,
      "auxiliary_loss_mlp": 0.01029386,
      "balance_loss_clip": 1.04641831,
      "balance_loss_mlp": 1.0213213,
      "epoch": 0.7075091685204112,
      "flos": 20083940784000.0,
      "grad_norm": 2.0571445632289045,
      "language_loss": 0.77313936,
      "learning_rate": 8.321311938774225e-07,
      "loss": 0.79525995,
      "num_input_tokens_seen": 126492705,
      "step": 5884,
      "time_per_iteration": 3.6111135482788086
    },
    {
      "auxiliary_loss_clip": 0.01171892,
      "auxiliary_loss_mlp": 0.01027036,
      "balance_loss_clip": 1.04778564,
      "balance_loss_mlp": 1.01888847,
      "epoch": 0.7076294114110503,
      "flos": 20777124424320.0,
      "grad_norm": 2.07287721191385,
      "language_loss": 0.78791034,
      "learning_rate": 8.314989126338104e-07,
      "loss": 0.80989963,
      "num_input_tokens_seen": 126512715,
      "step": 5885,
      "time_per_iteration": 2.611677646636963
    },
    {
      "auxiliary_loss_clip": 0.01174308,
      "auxiliary_loss_mlp": 0.01024302,
      "balance_loss_clip": 1.04594922,
      "balance_loss_mlp": 1.01674128,
      "epoch": 0.7077496543016895,
      "flos": 17967689141760.0,
      "grad_norm": 9.080301604666342,
      "language_loss": 0.84355217,
      "learning_rate": 8.308668086508847e-07,
      "loss": 0.86553824,
      "num_input_tokens_seen": 126530795,
      "step": 5886,
      "time_per_iteration": 2.720761775970459
    },
    {
      "auxiliary_loss_clip": 0.01181206,
      "auxiliary_loss_mlp": 0.01020627,
      "balance_loss_clip": 1.04214978,
      "balance_loss_mlp": 1.01210928,
      "epoch": 0.7078698971923285,
      "flos": 45478098564480.0,
      "grad_norm": 15.036811430854758,
      "language_loss": 0.73760694,
      "learning_rate": 8.302348820245342e-07,
      "loss": 0.75962526,
      "num_input_tokens_seen": 126553360,
      "step": 5887,
      "time_per_iteration": 2.916933059692383
    },
    {
      "auxiliary_loss_clip": 0.01178273,
      "auxiliary_loss_mlp": 0.01029659,
      "balance_loss_clip": 1.04268765,
      "balance_loss_mlp": 1.02126122,
      "epoch": 0.7079901400829676,
      "flos": 26943704547840.0,
      "grad_norm": 2.3380967203374308,
      "language_loss": 0.70425731,
      "learning_rate": 8.296031328506232e-07,
      "loss": 0.7263366,
      "num_input_tokens_seen": 126573110,
      "step": 5888,
      "time_per_iteration": 3.582292079925537
    },
    {
      "auxiliary_loss_clip": 0.01176535,
      "auxiliary_loss_mlp": 0.01025423,
      "balance_loss_clip": 1.04619455,
      "balance_loss_mlp": 1.01745987,
      "epoch": 0.7081103829736067,
      "flos": 24423206267520.0,
      "grad_norm": 1.8510764197953764,
      "language_loss": 0.76100409,
      "learning_rate": 8.289715612249857e-07,
      "loss": 0.78302366,
      "num_input_tokens_seen": 126593725,
      "step": 5889,
      "time_per_iteration": 2.6903131008148193
    },
    {
      "auxiliary_loss_clip": 0.01171011,
      "auxiliary_loss_mlp": 0.01028096,
      "balance_loss_clip": 1.04480612,
      "balance_loss_mlp": 1.02008462,
      "epoch": 0.7082306258642458,
      "flos": 18543300589440.0,
      "grad_norm": 9.219691149975613,
      "language_loss": 0.7793076,
      "learning_rate": 8.283401672434305e-07,
      "loss": 0.80129868,
      "num_input_tokens_seen": 126608950,
      "step": 5890,
      "time_per_iteration": 2.632415771484375
    },
    {
      "auxiliary_loss_clip": 0.01171538,
      "auxiliary_loss_mlp": 0.01026462,
      "balance_loss_clip": 1.04598475,
      "balance_loss_mlp": 1.01873755,
      "epoch": 0.7083508687548848,
      "flos": 23477534951040.0,
      "grad_norm": 2.69462912287188,
      "language_loss": 0.70317292,
      "learning_rate": 8.277089510017412e-07,
      "loss": 0.72515291,
      "num_input_tokens_seen": 126629755,
      "step": 5891,
      "time_per_iteration": 2.6904304027557373
    },
    {
      "auxiliary_loss_clip": 0.01173884,
      "auxiliary_loss_mlp": 0.01022213,
      "balance_loss_clip": 1.04760015,
      "balance_loss_mlp": 1.01462531,
      "epoch": 0.708471111645524,
      "flos": 22419463000320.0,
      "grad_norm": 1.9322909820183918,
      "language_loss": 0.8214314,
      "learning_rate": 8.270779125956719e-07,
      "loss": 0.84339231,
      "num_input_tokens_seen": 126650135,
      "step": 5892,
      "time_per_iteration": 2.8569283485412598
    },
    {
      "auxiliary_loss_clip": 0.01174137,
      "auxiliary_loss_mlp": 0.01024618,
      "balance_loss_clip": 1.04127097,
      "balance_loss_mlp": 1.01697373,
      "epoch": 0.7085913545361631,
      "flos": 20922885815040.0,
      "grad_norm": 1.959429709146144,
      "language_loss": 0.80011404,
      "learning_rate": 8.264470521209505e-07,
      "loss": 0.82210159,
      "num_input_tokens_seen": 126668500,
      "step": 5893,
      "time_per_iteration": 2.7242934703826904
    },
    {
      "auxiliary_loss_clip": 0.0116219,
      "auxiliary_loss_mlp": 0.01027096,
      "balance_loss_clip": 1.04334879,
      "balance_loss_mlp": 1.01920104,
      "epoch": 0.7087115974268021,
      "flos": 15012384727680.0,
      "grad_norm": 3.157379476700109,
      "language_loss": 0.76783651,
      "learning_rate": 8.258163696732785e-07,
      "loss": 0.78972936,
      "num_input_tokens_seen": 126686090,
      "step": 5894,
      "time_per_iteration": 2.710247039794922
    },
    {
      "auxiliary_loss_clip": 0.01166969,
      "auxiliary_loss_mlp": 0.01024998,
      "balance_loss_clip": 1.04549587,
      "balance_loss_mlp": 1.01739264,
      "epoch": 0.7088318403174413,
      "flos": 21539040739200.0,
      "grad_norm": 2.995859081178366,
      "language_loss": 0.77190906,
      "learning_rate": 8.251858653483288e-07,
      "loss": 0.79382873,
      "num_input_tokens_seen": 126704255,
      "step": 5895,
      "time_per_iteration": 2.639288902282715
    },
    {
      "auxiliary_loss_clip": 0.01171665,
      "auxiliary_loss_mlp": 0.01027372,
      "balance_loss_clip": 1.04735792,
      "balance_loss_mlp": 1.01950407,
      "epoch": 0.7089520832080803,
      "flos": 15516785462400.0,
      "grad_norm": 2.6209907652983184,
      "language_loss": 0.85788918,
      "learning_rate": 8.245555392417501e-07,
      "loss": 0.87987959,
      "num_input_tokens_seen": 126718910,
      "step": 5896,
      "time_per_iteration": 2.6768884658813477
    },
    {
      "auxiliary_loss_clip": 0.01166069,
      "auxiliary_loss_mlp": 0.01023584,
      "balance_loss_clip": 1.03903782,
      "balance_loss_mlp": 1.01531696,
      "epoch": 0.7090723260987194,
      "flos": 20412667077120.0,
      "grad_norm": 1.9944666920048806,
      "language_loss": 0.79039913,
      "learning_rate": 8.239253914491613e-07,
      "loss": 0.81229568,
      "num_input_tokens_seen": 126737235,
      "step": 5897,
      "time_per_iteration": 2.7149457931518555
    },
    {
      "auxiliary_loss_clip": 0.0117408,
      "auxiliary_loss_mlp": 0.01022532,
      "balance_loss_clip": 1.04433727,
      "balance_loss_mlp": 1.01490855,
      "epoch": 0.7091925689893585,
      "flos": 25668337271040.0,
      "grad_norm": 2.176721998268799,
      "language_loss": 0.75083315,
      "learning_rate": 8.232954220661556e-07,
      "loss": 0.77279925,
      "num_input_tokens_seen": 126759970,
      "step": 5898,
      "time_per_iteration": 2.721867322921753
    },
    {
      "auxiliary_loss_clip": 0.0117098,
      "auxiliary_loss_mlp": 0.01026157,
      "balance_loss_clip": 1.05097306,
      "balance_loss_mlp": 1.01898694,
      "epoch": 0.7093128118799976,
      "flos": 24206629213440.0,
      "grad_norm": 2.554079329268396,
      "language_loss": 0.70237023,
      "learning_rate": 8.226656311882989e-07,
      "loss": 0.72434157,
      "num_input_tokens_seen": 126779280,
      "step": 5899,
      "time_per_iteration": 2.5956029891967773
    },
    {
      "auxiliary_loss_clip": 0.01167978,
      "auxiliary_loss_mlp": 0.01026022,
      "balance_loss_clip": 1.04619265,
      "balance_loss_mlp": 1.01859832,
      "epoch": 0.7094330547706367,
      "flos": 16646786398080.0,
      "grad_norm": 2.5916650615934302,
      "language_loss": 0.7733016,
      "learning_rate": 8.22036018911129e-07,
      "loss": 0.79524159,
      "num_input_tokens_seen": 126797310,
      "step": 5900,
      "time_per_iteration": 2.665771722793579
    },
    {
      "auxiliary_loss_clip": 0.01173387,
      "auxiliary_loss_mlp": 0.01027719,
      "balance_loss_clip": 1.04810834,
      "balance_loss_mlp": 1.01970863,
      "epoch": 0.7095532976612757,
      "flos": 16283370545280.0,
      "grad_norm": 2.608623920167299,
      "language_loss": 0.80857033,
      "learning_rate": 8.214065853301599e-07,
      "loss": 0.83058143,
      "num_input_tokens_seen": 126812840,
      "step": 5901,
      "time_per_iteration": 2.536862373352051
    },
    {
      "auxiliary_loss_clip": 0.01074724,
      "auxiliary_loss_mlp": 0.01002727,
      "balance_loss_clip": 1.01297379,
      "balance_loss_mlp": 1.00194001,
      "epoch": 0.7096735405519149,
      "flos": 70722080559360.0,
      "grad_norm": 0.8084070513830425,
      "language_loss": 0.58152711,
      "learning_rate": 8.207773305408734e-07,
      "loss": 0.60230166,
      "num_input_tokens_seen": 126880060,
      "step": 5902,
      "time_per_iteration": 4.200056314468384
    },
    {
      "auxiliary_loss_clip": 0.01187641,
      "auxiliary_loss_mlp": 0.01026359,
      "balance_loss_clip": 1.04229927,
      "balance_loss_mlp": 1.01772261,
      "epoch": 0.709793783442554,
      "flos": 23621500661760.0,
      "grad_norm": 2.358700921722429,
      "language_loss": 0.80134845,
      "learning_rate": 8.201482546387288e-07,
      "loss": 0.82348841,
      "num_input_tokens_seen": 126899535,
      "step": 5903,
      "time_per_iteration": 2.783055305480957
    },
    {
      "auxiliary_loss_clip": 0.01170694,
      "auxiliary_loss_mlp": 0.01023575,
      "balance_loss_clip": 1.04712582,
      "balance_loss_mlp": 1.0164578,
      "epoch": 0.709914026333193,
      "flos": 25993472204160.0,
      "grad_norm": 1.638582305090425,
      "language_loss": 0.9195137,
      "learning_rate": 8.195193577191553e-07,
      "loss": 0.94145644,
      "num_input_tokens_seen": 126921365,
      "step": 5904,
      "time_per_iteration": 2.6563661098480225
    },
    {
      "auxiliary_loss_clip": 0.01180741,
      "auxiliary_loss_mlp": 0.01208305,
      "balance_loss_clip": 1.04473484,
      "balance_loss_mlp": 1.00083232,
      "epoch": 0.7100342692238322,
      "flos": 24861531934080.0,
      "grad_norm": 1.7133203744664396,
      "language_loss": 0.8474068,
      "learning_rate": 8.188906398775579e-07,
      "loss": 0.87129724,
      "num_input_tokens_seen": 126941910,
      "step": 5905,
      "time_per_iteration": 2.736497163772583
    },
    {
      "auxiliary_loss_clip": 0.01169317,
      "auxiliary_loss_mlp": 0.01208682,
      "balance_loss_clip": 1.04682314,
      "balance_loss_mlp": 1.00090766,
      "epoch": 0.7101545121144712,
      "flos": 24932203943040.0,
      "grad_norm": 2.0872212724169454,
      "language_loss": 0.68825245,
      "learning_rate": 8.18262101209311e-07,
      "loss": 0.71203244,
      "num_input_tokens_seen": 126961120,
      "step": 5906,
      "time_per_iteration": 2.6615118980407715
    },
    {
      "auxiliary_loss_clip": 0.01175614,
      "auxiliary_loss_mlp": 0.01024484,
      "balance_loss_clip": 1.04652071,
      "balance_loss_mlp": 1.01680732,
      "epoch": 0.7102747550051103,
      "flos": 23768842250880.0,
      "grad_norm": 2.264217050729355,
      "language_loss": 0.70024723,
      "learning_rate": 8.176337418097626e-07,
      "loss": 0.7222482,
      "num_input_tokens_seen": 126981590,
      "step": 5907,
      "time_per_iteration": 2.651383876800537
    },
    {
      "auxiliary_loss_clip": 0.01169565,
      "auxiliary_loss_mlp": 0.01208218,
      "balance_loss_clip": 1.04656339,
      "balance_loss_mlp": 1.00082886,
      "epoch": 0.7103949978957494,
      "flos": 15303907509120.0,
      "grad_norm": 2.1673396850281876,
      "language_loss": 0.80038917,
      "learning_rate": 8.170055617742364e-07,
      "loss": 0.82416701,
      "num_input_tokens_seen": 126998870,
      "step": 5908,
      "time_per_iteration": 3.581831932067871
    },
    {
      "auxiliary_loss_clip": 0.01167375,
      "auxiliary_loss_mlp": 0.01027355,
      "balance_loss_clip": 1.04197764,
      "balance_loss_mlp": 1.01929629,
      "epoch": 0.7105152407863885,
      "flos": 22638805401600.0,
      "grad_norm": 1.9296725571388176,
      "language_loss": 0.70797384,
      "learning_rate": 8.163775611980252e-07,
      "loss": 0.7299211,
      "num_input_tokens_seen": 127017980,
      "step": 5909,
      "time_per_iteration": 2.6736526489257812
    },
    {
      "auxiliary_loss_clip": 0.01176383,
      "auxiliary_loss_mlp": 0.01024718,
      "balance_loss_clip": 1.04721856,
      "balance_loss_mlp": 1.01730895,
      "epoch": 0.7106354836770276,
      "flos": 17238594879360.0,
      "grad_norm": 1.5997359706320107,
      "language_loss": 0.78605914,
      "learning_rate": 8.157497401763982e-07,
      "loss": 0.80807006,
      "num_input_tokens_seen": 127035645,
      "step": 5910,
      "time_per_iteration": 2.6642987728118896
    },
    {
      "auxiliary_loss_clip": 0.01171364,
      "auxiliary_loss_mlp": 0.01025237,
      "balance_loss_clip": 1.04700732,
      "balance_loss_mlp": 1.01723838,
      "epoch": 0.7107557265676667,
      "flos": 20193647898240.0,
      "grad_norm": 1.9194924675532659,
      "language_loss": 0.77636337,
      "learning_rate": 8.151220988045935e-07,
      "loss": 0.79832941,
      "num_input_tokens_seen": 127054900,
      "step": 5911,
      "time_per_iteration": 3.528519630432129
    },
    {
      "auxiliary_loss_clip": 0.01169793,
      "auxiliary_loss_mlp": 0.01025174,
      "balance_loss_clip": 1.04594767,
      "balance_loss_mlp": 1.01782513,
      "epoch": 0.7108759694583058,
      "flos": 21507080613120.0,
      "grad_norm": 1.715709502442378,
      "language_loss": 0.82741868,
      "learning_rate": 8.144946371778234e-07,
      "loss": 0.84936833,
      "num_input_tokens_seen": 127075010,
      "step": 5912,
      "time_per_iteration": 2.818488836288452
    },
    {
      "auxiliary_loss_clip": 0.01175714,
      "auxiliary_loss_mlp": 0.01208901,
      "balance_loss_clip": 1.04743457,
      "balance_loss_mlp": 1.00088978,
      "epoch": 0.7109962123489448,
      "flos": 24061909317120.0,
      "grad_norm": 1.7420777349858025,
      "language_loss": 0.78364426,
      "learning_rate": 8.138673553912751e-07,
      "loss": 0.80749035,
      "num_input_tokens_seen": 127095570,
      "step": 5913,
      "time_per_iteration": 2.7498779296875
    },
    {
      "auxiliary_loss_clip": 0.01179923,
      "auxiliary_loss_mlp": 0.01027387,
      "balance_loss_clip": 1.04308486,
      "balance_loss_mlp": 1.01895261,
      "epoch": 0.711116455239584,
      "flos": 30480474326400.0,
      "grad_norm": 2.993538018323359,
      "language_loss": 0.56882906,
      "learning_rate": 8.132402535401059e-07,
      "loss": 0.59090215,
      "num_input_tokens_seen": 127116825,
      "step": 5914,
      "time_per_iteration": 3.7091784477233887
    },
    {
      "auxiliary_loss_clip": 0.01167587,
      "auxiliary_loss_mlp": 0.01024458,
      "balance_loss_clip": 1.04646254,
      "balance_loss_mlp": 1.01647735,
      "epoch": 0.711236698130223,
      "flos": 25045610158080.0,
      "grad_norm": 1.8016873562288909,
      "language_loss": 0.74535549,
      "learning_rate": 8.126133317194465e-07,
      "loss": 0.76727593,
      "num_input_tokens_seen": 127137015,
      "step": 5915,
      "time_per_iteration": 2.681628942489624
    },
    {
      "auxiliary_loss_clip": 0.01190419,
      "auxiliary_loss_mlp": 0.01028354,
      "balance_loss_clip": 1.03912663,
      "balance_loss_mlp": 1.01992583,
      "epoch": 0.7113569410208621,
      "flos": 24206701040640.0,
      "grad_norm": 2.474973536489571,
      "language_loss": 0.74421751,
      "learning_rate": 8.11986590024401e-07,
      "loss": 0.76640522,
      "num_input_tokens_seen": 127156755,
      "step": 5916,
      "time_per_iteration": 2.7063424587249756
    },
    {
      "auxiliary_loss_clip": 0.01179024,
      "auxiliary_loss_mlp": 0.01031406,
      "balance_loss_clip": 1.04848576,
      "balance_loss_mlp": 1.02278697,
      "epoch": 0.7114771839115013,
      "flos": 35439306526080.0,
      "grad_norm": 1.7089973499465807,
      "language_loss": 0.6884883,
      "learning_rate": 8.113600285500442e-07,
      "loss": 0.71059257,
      "num_input_tokens_seen": 127176965,
      "step": 5917,
      "time_per_iteration": 2.7552309036254883
    },
    {
      "auxiliary_loss_clip": 0.01169837,
      "auxiliary_loss_mlp": 0.01021945,
      "balance_loss_clip": 1.04727602,
      "balance_loss_mlp": 1.01439357,
      "epoch": 0.7115974268021403,
      "flos": 21099458096640.0,
      "grad_norm": 1.799911843163258,
      "language_loss": 0.74571049,
      "learning_rate": 8.107336473914268e-07,
      "loss": 0.76762831,
      "num_input_tokens_seen": 127195595,
      "step": 5918,
      "time_per_iteration": 2.6176812648773193
    },
    {
      "auxiliary_loss_clip": 0.01078073,
      "auxiliary_loss_mlp": 0.01003418,
      "balance_loss_clip": 1.01294088,
      "balance_loss_mlp": 1.00263083,
      "epoch": 0.7117176696927794,
      "flos": 56752866616320.0,
      "grad_norm": 0.7732914346351453,
      "language_loss": 0.55778944,
      "learning_rate": 8.101074466435694e-07,
      "loss": 0.57860434,
      "num_input_tokens_seen": 127255070,
      "step": 5919,
      "time_per_iteration": 3.111694097518921
    },
    {
      "auxiliary_loss_clip": 0.01166158,
      "auxiliary_loss_mlp": 0.01028181,
      "balance_loss_clip": 1.04512393,
      "balance_loss_mlp": 1.02049196,
      "epoch": 0.7118379125834186,
      "flos": 15925269905280.0,
      "grad_norm": 1.6851399231241748,
      "language_loss": 0.68076545,
      "learning_rate": 8.094814264014662e-07,
      "loss": 0.70270884,
      "num_input_tokens_seen": 127273825,
      "step": 5920,
      "time_per_iteration": 2.6329305171966553
    },
    {
      "auxiliary_loss_clip": 0.01173331,
      "auxiliary_loss_mlp": 0.01027221,
      "balance_loss_clip": 1.04809284,
      "balance_loss_mlp": 1.01867414,
      "epoch": 0.7119581554740576,
      "flos": 20193360589440.0,
      "grad_norm": 2.8056119634945222,
      "language_loss": 0.81653255,
      "learning_rate": 8.088555867600844e-07,
      "loss": 0.83853805,
      "num_input_tokens_seen": 127289990,
      "step": 5921,
      "time_per_iteration": 2.588752269744873
    },
    {
      "auxiliary_loss_clip": 0.01175743,
      "auxiliary_loss_mlp": 0.01025306,
      "balance_loss_clip": 1.04352891,
      "balance_loss_mlp": 1.01785588,
      "epoch": 0.7120783983646967,
      "flos": 34715383822080.0,
      "grad_norm": 5.92705623425558,
      "language_loss": 0.60408425,
      "learning_rate": 8.08229927814362e-07,
      "loss": 0.62609476,
      "num_input_tokens_seen": 127312880,
      "step": 5922,
      "time_per_iteration": 2.830186605453491
    },
    {
      "auxiliary_loss_clip": 0.01175969,
      "auxiliary_loss_mlp": 0.01022686,
      "balance_loss_clip": 1.04283857,
      "balance_loss_mlp": 1.01486921,
      "epoch": 0.7121986412553358,
      "flos": 26359114700160.0,
      "grad_norm": 1.6748873491307255,
      "language_loss": 0.64948398,
      "learning_rate": 8.076044496592134e-07,
      "loss": 0.67147052,
      "num_input_tokens_seen": 127334730,
      "step": 5923,
      "time_per_iteration": 2.733210325241089
    },
    {
      "auxiliary_loss_clip": 0.01175638,
      "auxiliary_loss_mlp": 0.01025998,
      "balance_loss_clip": 1.04624915,
      "balance_loss_mlp": 1.01875329,
      "epoch": 0.7123188841459749,
      "flos": 11145344371200.0,
      "grad_norm": 2.818014400393316,
      "language_loss": 0.77578425,
      "learning_rate": 8.069791523895204e-07,
      "loss": 0.79780066,
      "num_input_tokens_seen": 127351180,
      "step": 5924,
      "time_per_iteration": 2.6683573722839355
    },
    {
      "auxiliary_loss_clip": 0.01165928,
      "auxiliary_loss_mlp": 0.01027181,
      "balance_loss_clip": 1.03979325,
      "balance_loss_mlp": 1.01950359,
      "epoch": 0.7124391270366139,
      "flos": 20811670329600.0,
      "grad_norm": 3.7061738905030777,
      "language_loss": 0.7757408,
      "learning_rate": 8.063540361001422e-07,
      "loss": 0.79767185,
      "num_input_tokens_seen": 127369750,
      "step": 5925,
      "time_per_iteration": 2.7397871017456055
    },
    {
      "auxiliary_loss_clip": 0.01171503,
      "auxiliary_loss_mlp": 0.01023879,
      "balance_loss_clip": 1.04246604,
      "balance_loss_mlp": 1.01564729,
      "epoch": 0.7125593699272531,
      "flos": 17603734584960.0,
      "grad_norm": 13.923633897123848,
      "language_loss": 0.79931921,
      "learning_rate": 8.057291008859069e-07,
      "loss": 0.82127303,
      "num_input_tokens_seen": 127387910,
      "step": 5926,
      "time_per_iteration": 2.6797144412994385
    },
    {
      "auxiliary_loss_clip": 0.01167381,
      "auxiliary_loss_mlp": 0.01026751,
      "balance_loss_clip": 1.04440808,
      "balance_loss_mlp": 1.01912403,
      "epoch": 0.7126796128178922,
      "flos": 28654057526400.0,
      "grad_norm": 1.9991416432089877,
      "language_loss": 0.6802187,
      "learning_rate": 8.051043468416187e-07,
      "loss": 0.70216,
      "num_input_tokens_seen": 127409160,
      "step": 5927,
      "time_per_iteration": 2.6652448177337646
    },
    {
      "auxiliary_loss_clip": 0.0116888,
      "auxiliary_loss_mlp": 0.0102291,
      "balance_loss_clip": 1.04932785,
      "balance_loss_mlp": 1.015167,
      "epoch": 0.7127998557085312,
      "flos": 16034438315520.0,
      "grad_norm": 1.952196075985017,
      "language_loss": 0.82564688,
      "learning_rate": 8.044797740620506e-07,
      "loss": 0.84756476,
      "num_input_tokens_seen": 127427765,
      "step": 5928,
      "time_per_iteration": 2.626521587371826
    },
    {
      "auxiliary_loss_clip": 0.01173596,
      "auxiliary_loss_mlp": 0.01028581,
      "balance_loss_clip": 1.04324484,
      "balance_loss_mlp": 1.02127302,
      "epoch": 0.7129200985991703,
      "flos": 23403271582080.0,
      "grad_norm": 2.442420957113023,
      "language_loss": 0.79286373,
      "learning_rate": 8.038553826419494e-07,
      "loss": 0.8148855,
      "num_input_tokens_seen": 127446475,
      "step": 5929,
      "time_per_iteration": 3.7815585136413574
    },
    {
      "auxiliary_loss_clip": 0.01167401,
      "auxiliary_loss_mlp": 0.01022477,
      "balance_loss_clip": 1.04611421,
      "balance_loss_mlp": 1.0146327,
      "epoch": 0.7130403414898094,
      "flos": 21397445326080.0,
      "grad_norm": 1.9991868630900673,
      "language_loss": 0.80708134,
      "learning_rate": 8.032311726760364e-07,
      "loss": 0.82898009,
      "num_input_tokens_seen": 127467695,
      "step": 5930,
      "time_per_iteration": 2.622978687286377
    },
    {
      "auxiliary_loss_clip": 0.01167208,
      "auxiliary_loss_mlp": 0.01022246,
      "balance_loss_clip": 1.04242933,
      "balance_loss_mlp": 1.01369858,
      "epoch": 0.7131605843804485,
      "flos": 74739045306240.0,
      "grad_norm": 1.7504127997975492,
      "language_loss": 0.69055593,
      "learning_rate": 8.026071442590022e-07,
      "loss": 0.71245044,
      "num_input_tokens_seen": 127494590,
      "step": 5931,
      "time_per_iteration": 3.0663819313049316
    },
    {
      "auxiliary_loss_clip": 0.01172601,
      "auxiliary_loss_mlp": 0.01022963,
      "balance_loss_clip": 1.04936254,
      "balance_loss_mlp": 1.01541734,
      "epoch": 0.7132808272710875,
      "flos": 18368739469440.0,
      "grad_norm": 2.075225783559101,
      "language_loss": 0.80595779,
      "learning_rate": 8.019832974855134e-07,
      "loss": 0.8279134,
      "num_input_tokens_seen": 127512550,
      "step": 5932,
      "time_per_iteration": 2.6541125774383545
    },
    {
      "auxiliary_loss_clip": 0.01174891,
      "auxiliary_loss_mlp": 0.01025707,
      "balance_loss_clip": 1.04460716,
      "balance_loss_mlp": 1.01750553,
      "epoch": 0.7134010701617267,
      "flos": 23253380127360.0,
      "grad_norm": 3.5192019602561637,
      "language_loss": 0.82866198,
      "learning_rate": 8.013596324502052e-07,
      "loss": 0.85066795,
      "num_input_tokens_seen": 127531015,
      "step": 5933,
      "time_per_iteration": 2.6957526206970215
    },
    {
      "auxiliary_loss_clip": 0.01166395,
      "auxiliary_loss_mlp": 0.01023233,
      "balance_loss_clip": 1.04739952,
      "balance_loss_mlp": 1.01568747,
      "epoch": 0.7135213130523658,
      "flos": 23653137565440.0,
      "grad_norm": 1.9629986256655982,
      "language_loss": 0.7883153,
      "learning_rate": 8.007361492476872e-07,
      "loss": 0.81021154,
      "num_input_tokens_seen": 127550340,
      "step": 5934,
      "time_per_iteration": 2.599485158920288
    },
    {
      "auxiliary_loss_clip": 0.01186342,
      "auxiliary_loss_mlp": 0.01027932,
      "balance_loss_clip": 1.04448223,
      "balance_loss_mlp": 1.01976657,
      "epoch": 0.7136415559430048,
      "flos": 24790644443520.0,
      "grad_norm": 1.5313946260080171,
      "language_loss": 0.79101408,
      "learning_rate": 8.001128479725426e-07,
      "loss": 0.81315678,
      "num_input_tokens_seen": 127572245,
      "step": 5935,
      "time_per_iteration": 3.6233396530151367
    },
    {
      "auxiliary_loss_clip": 0.01170617,
      "auxiliary_loss_mlp": 0.01022177,
      "balance_loss_clip": 1.03844655,
      "balance_loss_mlp": 1.01414871,
      "epoch": 0.713761798833644,
      "flos": 18296954138880.0,
      "grad_norm": 1.6430911575673484,
      "language_loss": 0.80815953,
      "learning_rate": 7.994897287193248e-07,
      "loss": 0.83008742,
      "num_input_tokens_seen": 127591625,
      "step": 5936,
      "time_per_iteration": 2.7787022590637207
    },
    {
      "auxiliary_loss_clip": 0.01173768,
      "auxiliary_loss_mlp": 0.01028637,
      "balance_loss_clip": 1.04586709,
      "balance_loss_mlp": 1.0204649,
      "epoch": 0.713882041724283,
      "flos": 15558262692480.0,
      "grad_norm": 3.231444592001822,
      "language_loss": 0.83708233,
      "learning_rate": 7.988667915825605e-07,
      "loss": 0.8591063,
      "num_input_tokens_seen": 127608690,
      "step": 5937,
      "time_per_iteration": 3.5658113956451416
    },
    {
      "auxiliary_loss_clip": 0.01175323,
      "auxiliary_loss_mlp": 0.01025139,
      "balance_loss_clip": 1.04555869,
      "balance_loss_mlp": 1.01713967,
      "epoch": 0.7140022846149221,
      "flos": 24061011477120.0,
      "grad_norm": 2.154199807173768,
      "language_loss": 0.75646657,
      "learning_rate": 7.982440366567491e-07,
      "loss": 0.77847123,
      "num_input_tokens_seen": 127627180,
      "step": 5938,
      "time_per_iteration": 2.790025234222412
    },
    {
      "auxiliary_loss_clip": 0.01165859,
      "auxiliary_loss_mlp": 0.01023924,
      "balance_loss_clip": 1.04481936,
      "balance_loss_mlp": 1.01632452,
      "epoch": 0.7141225275055613,
      "flos": 27891710248320.0,
      "grad_norm": 1.6419702312018103,
      "language_loss": 0.75124449,
      "learning_rate": 7.97621464036361e-07,
      "loss": 0.77314234,
      "num_input_tokens_seen": 127648940,
      "step": 5939,
      "time_per_iteration": 2.6686692237854004
    },
    {
      "auxiliary_loss_clip": 0.01174824,
      "auxiliary_loss_mlp": 0.01026523,
      "balance_loss_clip": 1.04700029,
      "balance_loss_mlp": 1.0181489,
      "epoch": 0.7142427703962003,
      "flos": 19682603147520.0,
      "grad_norm": 2.0029423696795607,
      "language_loss": 0.68004739,
      "learning_rate": 7.969990738158417e-07,
      "loss": 0.70206082,
      "num_input_tokens_seen": 127667350,
      "step": 5940,
      "time_per_iteration": 2.5833632946014404
    },
    {
      "auxiliary_loss_clip": 0.01174777,
      "auxiliary_loss_mlp": 0.01027606,
      "balance_loss_clip": 1.04838991,
      "balance_loss_mlp": 1.01942241,
      "epoch": 0.7143630132868394,
      "flos": 21032377447680.0,
      "grad_norm": 3.928113408186821,
      "language_loss": 0.85152721,
      "learning_rate": 7.963768660896062e-07,
      "loss": 0.87355101,
      "num_input_tokens_seen": 127685760,
      "step": 5941,
      "time_per_iteration": 2.589567184448242
    },
    {
      "auxiliary_loss_clip": 0.01174432,
      "auxiliary_loss_mlp": 0.01025186,
      "balance_loss_clip": 1.04662323,
      "balance_loss_mlp": 1.01678205,
      "epoch": 0.7144832561774785,
      "flos": 24129923719680.0,
      "grad_norm": 2.40384493453909,
      "language_loss": 0.82313931,
      "learning_rate": 7.957548409520432e-07,
      "loss": 0.84513545,
      "num_input_tokens_seen": 127704985,
      "step": 5942,
      "time_per_iteration": 3.5636050701141357
    },
    {
      "auxiliary_loss_clip": 0.01176853,
      "auxiliary_loss_mlp": 0.01021355,
      "balance_loss_clip": 1.0425036,
      "balance_loss_mlp": 1.01390409,
      "epoch": 0.7146034990681176,
      "flos": 16325817442560.0,
      "grad_norm": 2.0796158338845676,
      "language_loss": 0.8454625,
      "learning_rate": 7.951329984975135e-07,
      "loss": 0.86744463,
      "num_input_tokens_seen": 127721925,
      "step": 5943,
      "time_per_iteration": 2.657011032104492
    },
    {
      "auxiliary_loss_clip": 0.01093179,
      "auxiliary_loss_mlp": 0.0100128,
      "balance_loss_clip": 1.01541901,
      "balance_loss_mlp": 1.00045145,
      "epoch": 0.7147237419587567,
      "flos": 69627164232960.0,
      "grad_norm": 0.7362534410315467,
      "language_loss": 0.54183489,
      "learning_rate": 7.94511338820349e-07,
      "loss": 0.56277949,
      "num_input_tokens_seen": 127784230,
      "step": 5944,
      "time_per_iteration": 3.2271082401275635
    },
    {
      "auxiliary_loss_clip": 0.01176482,
      "auxiliary_loss_mlp": 0.01208607,
      "balance_loss_clip": 1.04531765,
      "balance_loss_mlp": 1.00063205,
      "epoch": 0.7148439848493958,
      "flos": 22266806198400.0,
      "grad_norm": 6.280934685525179,
      "language_loss": 0.78671789,
      "learning_rate": 7.938898620148575e-07,
      "loss": 0.81056881,
      "num_input_tokens_seen": 127801990,
      "step": 5945,
      "time_per_iteration": 2.6964519023895264
    },
    {
      "auxiliary_loss_clip": 0.01173415,
      "auxiliary_loss_mlp": 0.01027002,
      "balance_loss_clip": 1.04514503,
      "balance_loss_mlp": 1.01928926,
      "epoch": 0.7149642277400349,
      "flos": 17931383470080.0,
      "grad_norm": 1.9183656155101778,
      "language_loss": 0.71070898,
      "learning_rate": 7.932685681753135e-07,
      "loss": 0.7327131,
      "num_input_tokens_seen": 127819270,
      "step": 5946,
      "time_per_iteration": 2.67276930809021
    },
    {
      "auxiliary_loss_clip": 0.01166532,
      "auxiliary_loss_mlp": 0.01021359,
      "balance_loss_clip": 1.04793644,
      "balance_loss_mlp": 1.01422465,
      "epoch": 0.7150844706306739,
      "flos": 31681937370240.0,
      "grad_norm": 2.2886208513502537,
      "language_loss": 0.62991202,
      "learning_rate": 7.92647457395969e-07,
      "loss": 0.65179092,
      "num_input_tokens_seen": 127841095,
      "step": 5947,
      "time_per_iteration": 2.6896209716796875
    },
    {
      "auxiliary_loss_clip": 0.01189553,
      "auxiliary_loss_mlp": 0.01025878,
      "balance_loss_clip": 1.03924847,
      "balance_loss_mlp": 1.01784301,
      "epoch": 0.7152047135213131,
      "flos": 10926217451520.0,
      "grad_norm": 2.1505888318129687,
      "language_loss": 0.7441026,
      "learning_rate": 7.920265297710444e-07,
      "loss": 0.76625687,
      "num_input_tokens_seen": 127858485,
      "step": 5948,
      "time_per_iteration": 2.6656758785247803
    },
    {
      "auxiliary_loss_clip": 0.01173859,
      "auxiliary_loss_mlp": 0.01028806,
      "balance_loss_clip": 1.04819143,
      "balance_loss_mlp": 1.02099192,
      "epoch": 0.7153249564119522,
      "flos": 20995640812800.0,
      "grad_norm": 1.8259112750540998,
      "language_loss": 0.73309958,
      "learning_rate": 7.914057853947363e-07,
      "loss": 0.75512624,
      "num_input_tokens_seen": 127877665,
      "step": 5949,
      "time_per_iteration": 2.6355440616607666
    },
    {
      "auxiliary_loss_clip": 0.01174059,
      "auxiliary_loss_mlp": 0.01027885,
      "balance_loss_clip": 1.04273915,
      "balance_loss_mlp": 1.01980865,
      "epoch": 0.7154451993025912,
      "flos": 24243114453120.0,
      "grad_norm": 3.396439532402694,
      "language_loss": 0.62945247,
      "learning_rate": 7.907852243612089e-07,
      "loss": 0.65147185,
      "num_input_tokens_seen": 127898070,
      "step": 5950,
      "time_per_iteration": 2.766681671142578
    },
    {
      "auxiliary_loss_clip": 0.01172879,
      "auxiliary_loss_mlp": 0.01023534,
      "balance_loss_clip": 1.04453695,
      "balance_loss_mlp": 1.0158006,
      "epoch": 0.7155654421932304,
      "flos": 23330947547520.0,
      "grad_norm": 1.917063224922864,
      "language_loss": 0.72566384,
      "learning_rate": 7.901648467646009e-07,
      "loss": 0.74762797,
      "num_input_tokens_seen": 127917010,
      "step": 5951,
      "time_per_iteration": 2.684842586517334
    },
    {
      "auxiliary_loss_clip": 0.01171356,
      "auxiliary_loss_mlp": 0.01027456,
      "balance_loss_clip": 1.04906225,
      "balance_loss_mlp": 1.01985097,
      "epoch": 0.7156856850838694,
      "flos": 22711883621760.0,
      "grad_norm": 1.7500300348470335,
      "language_loss": 0.72292221,
      "learning_rate": 7.895446526990244e-07,
      "loss": 0.74491036,
      "num_input_tokens_seen": 127937025,
      "step": 5952,
      "time_per_iteration": 2.6722569465637207
    },
    {
      "auxiliary_loss_clip": 0.01188573,
      "auxiliary_loss_mlp": 0.01024566,
      "balance_loss_clip": 1.04376233,
      "balance_loss_mlp": 1.01673961,
      "epoch": 0.7158059279745085,
      "flos": 19865424395520.0,
      "grad_norm": 1.738762313344546,
      "language_loss": 0.75678426,
      "learning_rate": 7.889246422585609e-07,
      "loss": 0.77891564,
      "num_input_tokens_seen": 127956410,
      "step": 5953,
      "time_per_iteration": 2.6844658851623535
    },
    {
      "auxiliary_loss_clip": 0.01171509,
      "auxiliary_loss_mlp": 0.01027557,
      "balance_loss_clip": 1.05006719,
      "balance_loss_mlp": 1.02021384,
      "epoch": 0.7159261708651476,
      "flos": 24134772055680.0,
      "grad_norm": 1.8041902862429466,
      "language_loss": 0.73660302,
      "learning_rate": 7.883048155372675e-07,
      "loss": 0.75859362,
      "num_input_tokens_seen": 127974925,
      "step": 5954,
      "time_per_iteration": 2.6669368743896484
    },
    {
      "auxiliary_loss_clip": 0.01180724,
      "auxiliary_loss_mlp": 0.01021898,
      "balance_loss_clip": 1.0464375,
      "balance_loss_mlp": 1.01411378,
      "epoch": 0.7160464137557867,
      "flos": 16983198201600.0,
      "grad_norm": 2.153273823903448,
      "language_loss": 0.71697676,
      "learning_rate": 7.876851726291698e-07,
      "loss": 0.73900294,
      "num_input_tokens_seen": 127993225,
      "step": 5955,
      "time_per_iteration": 2.608048915863037
    },
    {
      "auxiliary_loss_clip": 0.0117969,
      "auxiliary_loss_mlp": 0.01023668,
      "balance_loss_clip": 1.04291594,
      "balance_loss_mlp": 1.01622939,
      "epoch": 0.7161666566464258,
      "flos": 25228251838080.0,
      "grad_norm": 1.9033611539913184,
      "language_loss": 0.78572434,
      "learning_rate": 7.870657136282666e-07,
      "loss": 0.80775791,
      "num_input_tokens_seen": 128012085,
      "step": 5956,
      "time_per_iteration": 3.610860586166382
    },
    {
      "auxiliary_loss_clip": 0.01168541,
      "auxiliary_loss_mlp": 0.01029676,
      "balance_loss_clip": 1.04639125,
      "balance_loss_mlp": 1.02168345,
      "epoch": 0.7162868995370649,
      "flos": 26468390851200.0,
      "grad_norm": 1.8589931551625378,
      "language_loss": 0.81960702,
      "learning_rate": 7.86446438628531e-07,
      "loss": 0.84158921,
      "num_input_tokens_seen": 128033155,
      "step": 5957,
      "time_per_iteration": 2.680034875869751
    },
    {
      "auxiliary_loss_clip": 0.01068118,
      "auxiliary_loss_mlp": 0.01002227,
      "balance_loss_clip": 1.01402974,
      "balance_loss_mlp": 1.00139236,
      "epoch": 0.716407142427704,
      "flos": 69998912040960.0,
      "grad_norm": 0.7667684508597383,
      "language_loss": 0.56851816,
      "learning_rate": 7.858273477239059e-07,
      "loss": 0.58922166,
      "num_input_tokens_seen": 128101575,
      "step": 5958,
      "time_per_iteration": 3.185379981994629
    },
    {
      "auxiliary_loss_clip": 0.01164816,
      "auxiliary_loss_mlp": 0.01027814,
      "balance_loss_clip": 1.04049015,
      "balance_loss_mlp": 1.01954663,
      "epoch": 0.716527385318343,
      "flos": 20740459616640.0,
      "grad_norm": 1.819378185243125,
      "language_loss": 0.7145201,
      "learning_rate": 7.852084410083067e-07,
      "loss": 0.73644638,
      "num_input_tokens_seen": 128120395,
      "step": 5959,
      "time_per_iteration": 2.6972272396087646
    },
    {
      "auxiliary_loss_clip": 0.01170582,
      "auxiliary_loss_mlp": 0.01024128,
      "balance_loss_clip": 1.04572082,
      "balance_loss_mlp": 1.0169251,
      "epoch": 0.7166476282089821,
      "flos": 25371966153600.0,
      "grad_norm": 1.9004797978482002,
      "language_loss": 0.63955098,
      "learning_rate": 7.84589718575621e-07,
      "loss": 0.66149807,
      "num_input_tokens_seen": 128140840,
      "step": 5960,
      "time_per_iteration": 2.6545169353485107
    },
    {
      "auxiliary_loss_clip": 0.01174562,
      "auxiliary_loss_mlp": 0.01025135,
      "balance_loss_clip": 1.04109359,
      "balance_loss_mlp": 1.01711869,
      "epoch": 0.7167678710996213,
      "flos": 24133730561280.0,
      "grad_norm": 1.9944238528549134,
      "language_loss": 0.69243771,
      "learning_rate": 7.83971180519708e-07,
      "loss": 0.71443468,
      "num_input_tokens_seen": 128159695,
      "step": 5961,
      "time_per_iteration": 2.6691250801086426
    },
    {
      "auxiliary_loss_clip": 0.01173147,
      "auxiliary_loss_mlp": 0.01021051,
      "balance_loss_clip": 1.05032325,
      "balance_loss_mlp": 1.01293254,
      "epoch": 0.7168881139902603,
      "flos": 30226586019840.0,
      "grad_norm": 2.8853819893026533,
      "language_loss": 0.76349002,
      "learning_rate": 7.833528269344008e-07,
      "loss": 0.78543198,
      "num_input_tokens_seen": 128179600,
      "step": 5962,
      "time_per_iteration": 3.567293643951416
    },
    {
      "auxiliary_loss_clip": 0.01175902,
      "auxiliary_loss_mlp": 0.01025515,
      "balance_loss_clip": 1.04478443,
      "balance_loss_mlp": 1.01711655,
      "epoch": 0.7170083568808994,
      "flos": 14606414236800.0,
      "grad_norm": 2.0661045337891664,
      "language_loss": 0.77032542,
      "learning_rate": 7.827346579135023e-07,
      "loss": 0.79233962,
      "num_input_tokens_seen": 128196940,
      "step": 5963,
      "time_per_iteration": 2.7116010189056396
    },
    {
      "auxiliary_loss_clip": 0.01171036,
      "auxiliary_loss_mlp": 0.01023266,
      "balance_loss_clip": 1.04262972,
      "balance_loss_mlp": 1.01508188,
      "epoch": 0.7171285997715385,
      "flos": 23331091201920.0,
      "grad_norm": 1.9711806271142036,
      "language_loss": 0.83153486,
      "learning_rate": 7.821166735507885e-07,
      "loss": 0.85347784,
      "num_input_tokens_seen": 128215970,
      "step": 5964,
      "time_per_iteration": 3.59159779548645
    },
    {
      "auxiliary_loss_clip": 0.01169251,
      "auxiliary_loss_mlp": 0.0102402,
      "balance_loss_clip": 1.04816461,
      "balance_loss_mlp": 1.01621819,
      "epoch": 0.7172488426621776,
      "flos": 16543543731840.0,
      "grad_norm": 1.6638004769588361,
      "language_loss": 0.68570095,
      "learning_rate": 7.81498873940007e-07,
      "loss": 0.70763361,
      "num_input_tokens_seen": 128233185,
      "step": 5965,
      "time_per_iteration": 2.5658698081970215
    },
    {
      "auxiliary_loss_clip": 0.01177004,
      "auxiliary_loss_mlp": 0.01020105,
      "balance_loss_clip": 1.04497099,
      "balance_loss_mlp": 1.01194501,
      "epoch": 0.7173690855528166,
      "flos": 26541612725760.0,
      "grad_norm": 2.345428496410574,
      "language_loss": 0.77654409,
      "learning_rate": 7.808812591748768e-07,
      "loss": 0.7985152,
      "num_input_tokens_seen": 128253565,
      "step": 5966,
      "time_per_iteration": 2.6428966522216797
    },
    {
      "auxiliary_loss_clip": 0.01174225,
      "auxiliary_loss_mlp": 0.01022236,
      "balance_loss_clip": 1.04374111,
      "balance_loss_mlp": 1.01413023,
      "epoch": 0.7174893284434558,
      "flos": 22784099915520.0,
      "grad_norm": 2.1610549759989675,
      "language_loss": 0.65292311,
      "learning_rate": 7.802638293490915e-07,
      "loss": 0.67488766,
      "num_input_tokens_seen": 128273210,
      "step": 5967,
      "time_per_iteration": 2.6294031143188477
    },
    {
      "auxiliary_loss_clip": 0.01177864,
      "auxiliary_loss_mlp": 0.01024582,
      "balance_loss_clip": 1.04465079,
      "balance_loss_mlp": 1.01710737,
      "epoch": 0.7176095713340949,
      "flos": 23293564467840.0,
      "grad_norm": 2.2769772150002314,
      "language_loss": 0.77412164,
      "learning_rate": 7.796465845563123e-07,
      "loss": 0.79614615,
      "num_input_tokens_seen": 128292085,
      "step": 5968,
      "time_per_iteration": 3.6313886642456055
    },
    {
      "auxiliary_loss_clip": 0.01168207,
      "auxiliary_loss_mlp": 0.01208593,
      "balance_loss_clip": 1.04379368,
      "balance_loss_mlp": 1.0007273,
      "epoch": 0.7177298142247339,
      "flos": 25591631777280.0,
      "grad_norm": 2.1190171056564275,
      "language_loss": 0.79510498,
      "learning_rate": 7.790295248901766e-07,
      "loss": 0.81887299,
      "num_input_tokens_seen": 128313215,
      "step": 5969,
      "time_per_iteration": 2.722926616668701
    },
    {
      "auxiliary_loss_clip": 0.01172037,
      "auxiliary_loss_mlp": 0.01023454,
      "balance_loss_clip": 1.04678082,
      "balance_loss_mlp": 1.01563358,
      "epoch": 0.7178500571153731,
      "flos": 31652778504960.0,
      "grad_norm": 1.755409664287914,
      "language_loss": 0.62270474,
      "learning_rate": 7.784126504442902e-07,
      "loss": 0.64465964,
      "num_input_tokens_seen": 128336445,
      "step": 5970,
      "time_per_iteration": 2.7728583812713623
    },
    {
      "auxiliary_loss_clip": 0.01169546,
      "auxiliary_loss_mlp": 0.0102219,
      "balance_loss_clip": 1.04379165,
      "balance_loss_mlp": 1.01420283,
      "epoch": 0.7179703000060121,
      "flos": 19427242383360.0,
      "grad_norm": 1.377922804695639,
      "language_loss": 0.67890179,
      "learning_rate": 7.777959613122351e-07,
      "loss": 0.70081913,
      "num_input_tokens_seen": 128356270,
      "step": 5971,
      "time_per_iteration": 2.704385757446289
    },
    {
      "auxiliary_loss_clip": 0.01165609,
      "auxiliary_loss_mlp": 0.01028612,
      "balance_loss_clip": 1.04365659,
      "balance_loss_mlp": 1.02087545,
      "epoch": 0.7180905428966512,
      "flos": 28839249072000.0,
      "grad_norm": 1.9773987732869729,
      "language_loss": 0.77971601,
      "learning_rate": 7.771794575875604e-07,
      "loss": 0.80165827,
      "num_input_tokens_seen": 128378140,
      "step": 5972,
      "time_per_iteration": 2.6415863037109375
    },
    {
      "auxiliary_loss_clip": 0.01172362,
      "auxiliary_loss_mlp": 0.01031381,
      "balance_loss_clip": 1.04789853,
      "balance_loss_mlp": 1.02309036,
      "epoch": 0.7182107857872904,
      "flos": 20047563285120.0,
      "grad_norm": 3.3729969200977616,
      "language_loss": 0.77820891,
      "learning_rate": 7.765631393637888e-07,
      "loss": 0.80024636,
      "num_input_tokens_seen": 128396335,
      "step": 5973,
      "time_per_iteration": 2.6115810871124268
    },
    {
      "auxiliary_loss_clip": 0.01168255,
      "auxiliary_loss_mlp": 0.01024572,
      "balance_loss_clip": 1.04418957,
      "balance_loss_mlp": 1.01632285,
      "epoch": 0.7183310286779294,
      "flos": 22747686503040.0,
      "grad_norm": 2.5684340020001875,
      "language_loss": 0.4942959,
      "learning_rate": 7.75947006734417e-07,
      "loss": 0.51622415,
      "num_input_tokens_seen": 128414115,
      "step": 5974,
      "time_per_iteration": 2.5611941814422607
    },
    {
      "auxiliary_loss_clip": 0.01168572,
      "auxiliary_loss_mlp": 0.01027921,
      "balance_loss_clip": 1.04684782,
      "balance_loss_mlp": 1.01983225,
      "epoch": 0.7184512715685685,
      "flos": 17158262112000.0,
      "grad_norm": 2.47195959091163,
      "language_loss": 0.83003205,
      "learning_rate": 7.753310597929101e-07,
      "loss": 0.85199696,
      "num_input_tokens_seen": 128430755,
      "step": 5975,
      "time_per_iteration": 2.572134017944336
    },
    {
      "auxiliary_loss_clip": 0.01068892,
      "auxiliary_loss_mlp": 0.01002969,
      "balance_loss_clip": 1.01485777,
      "balance_loss_mlp": 1.00211036,
      "epoch": 0.7185715144592076,
      "flos": 65509611448320.0,
      "grad_norm": 0.7535403571698569,
      "language_loss": 0.55143648,
      "learning_rate": 7.747152986327095e-07,
      "loss": 0.57215506,
      "num_input_tokens_seen": 128491300,
      "step": 5976,
      "time_per_iteration": 3.065023183822632
    },
    {
      "auxiliary_loss_clip": 0.01182756,
      "auxiliary_loss_mlp": 0.01026336,
      "balance_loss_clip": 1.04269528,
      "balance_loss_mlp": 1.01884413,
      "epoch": 0.7186917573498467,
      "flos": 16180522928640.0,
      "grad_norm": 1.8895047545962727,
      "language_loss": 0.68044335,
      "learning_rate": 7.740997233472228e-07,
      "loss": 0.70253426,
      "num_input_tokens_seen": 128508920,
      "step": 5977,
      "time_per_iteration": 2.7217891216278076
    },
    {
      "auxiliary_loss_clip": 0.01173918,
      "auxiliary_loss_mlp": 0.01021879,
      "balance_loss_clip": 1.0445255,
      "balance_loss_mlp": 1.01487494,
      "epoch": 0.7188120002404857,
      "flos": 29242274647680.0,
      "grad_norm": 2.4958607455389092,
      "language_loss": 0.70362675,
      "learning_rate": 7.734843340298329e-07,
      "loss": 0.72558475,
      "num_input_tokens_seen": 128528745,
      "step": 5978,
      "time_per_iteration": 2.7261857986450195
    },
    {
      "auxiliary_loss_clip": 0.01179028,
      "auxiliary_loss_mlp": 0.01031155,
      "balance_loss_clip": 1.04424047,
      "balance_loss_mlp": 1.02266109,
      "epoch": 0.7189322431311249,
      "flos": 33401161008000.0,
      "grad_norm": 3.4224578552528553,
      "language_loss": 0.74746537,
      "learning_rate": 7.72869130773895e-07,
      "loss": 0.76956719,
      "num_input_tokens_seen": 128549345,
      "step": 5979,
      "time_per_iteration": 2.807544469833374
    },
    {
      "auxiliary_loss_clip": 0.01077332,
      "auxiliary_loss_mlp": 0.01000775,
      "balance_loss_clip": 1.01572514,
      "balance_loss_mlp": 0.99998254,
      "epoch": 0.719052486021764,
      "flos": 61351263792000.0,
      "grad_norm": 1.0446082098203422,
      "language_loss": 0.59276831,
      "learning_rate": 7.722541136727343e-07,
      "loss": 0.61354935,
      "num_input_tokens_seen": 128605360,
      "step": 5980,
      "time_per_iteration": 3.0535647869110107
    },
    {
      "auxiliary_loss_clip": 0.01173089,
      "auxiliary_loss_mlp": 0.01023364,
      "balance_loss_clip": 1.04802406,
      "balance_loss_mlp": 1.01554334,
      "epoch": 0.719172728912403,
      "flos": 15596795007360.0,
      "grad_norm": 2.3511837701616507,
      "language_loss": 0.8092317,
      "learning_rate": 7.716392828196483e-07,
      "loss": 0.83119625,
      "num_input_tokens_seen": 128623160,
      "step": 5981,
      "time_per_iteration": 2.6171863079071045
    },
    {
      "auxiliary_loss_clip": 0.01172755,
      "auxiliary_loss_mlp": 0.01028385,
      "balance_loss_clip": 1.04754496,
      "balance_loss_mlp": 1.02075541,
      "epoch": 0.7192929718030422,
      "flos": 15553162961280.0,
      "grad_norm": 2.4663599297506686,
      "language_loss": 0.77455729,
      "learning_rate": 7.710246383079064e-07,
      "loss": 0.79656869,
      "num_input_tokens_seen": 128638545,
      "step": 5982,
      "time_per_iteration": 3.5442054271698
    },
    {
      "auxiliary_loss_clip": 0.01175912,
      "auxiliary_loss_mlp": 0.01023196,
      "balance_loss_clip": 1.04268074,
      "balance_loss_mlp": 1.01529241,
      "epoch": 0.7194132146936812,
      "flos": 21862487733120.0,
      "grad_norm": 3.1828030649546717,
      "language_loss": 0.92544949,
      "learning_rate": 7.704101802307492e-07,
      "loss": 0.94744056,
      "num_input_tokens_seen": 128650845,
      "step": 5983,
      "time_per_iteration": 2.700925350189209
    },
    {
      "auxiliary_loss_clip": 0.0117142,
      "auxiliary_loss_mlp": 0.0103244,
      "balance_loss_clip": 1.04456663,
      "balance_loss_mlp": 1.02392256,
      "epoch": 0.7195334575843203,
      "flos": 27338900958720.0,
      "grad_norm": 2.5267113776127594,
      "language_loss": 0.87290168,
      "learning_rate": 7.697959086813912e-07,
      "loss": 0.89494026,
      "num_input_tokens_seen": 128667010,
      "step": 5984,
      "time_per_iteration": 2.835763454437256
    },
    {
      "auxiliary_loss_clip": 0.01169948,
      "auxiliary_loss_mlp": 0.01026545,
      "balance_loss_clip": 1.0421139,
      "balance_loss_mlp": 1.01887941,
      "epoch": 0.7196537004749595,
      "flos": 18770615809920.0,
      "grad_norm": 1.7196302940305246,
      "language_loss": 0.80380523,
      "learning_rate": 7.691818237530145e-07,
      "loss": 0.82577014,
      "num_input_tokens_seen": 128685870,
      "step": 5985,
      "time_per_iteration": 2.724641799926758
    },
    {
      "auxiliary_loss_clip": 0.01191071,
      "auxiliary_loss_mlp": 0.01027751,
      "balance_loss_clip": 1.04247952,
      "balance_loss_mlp": 1.02032435,
      "epoch": 0.7197739433655985,
      "flos": 24531009960960.0,
      "grad_norm": 1.7762412859283991,
      "language_loss": 0.77635419,
      "learning_rate": 7.685679255387774e-07,
      "loss": 0.79854244,
      "num_input_tokens_seen": 128704185,
      "step": 5986,
      "time_per_iteration": 2.755056142807007
    },
    {
      "auxiliary_loss_clip": 0.01173124,
      "auxiliary_loss_mlp": 0.01025923,
      "balance_loss_clip": 1.04559004,
      "balance_loss_mlp": 1.01817417,
      "epoch": 0.7198941862562376,
      "flos": 18040587793920.0,
      "grad_norm": 2.703584869940178,
      "language_loss": 0.76791573,
      "learning_rate": 7.679542141318065e-07,
      "loss": 0.7899062,
      "num_input_tokens_seen": 128721290,
      "step": 5987,
      "time_per_iteration": 2.741257667541504
    },
    {
      "auxiliary_loss_clip": 0.01161971,
      "auxiliary_loss_mlp": 0.01025533,
      "balance_loss_clip": 1.0421927,
      "balance_loss_mlp": 1.0178144,
      "epoch": 0.7200144291468767,
      "flos": 29022393542400.0,
      "grad_norm": 2.0449562909502914,
      "language_loss": 0.75970107,
      "learning_rate": 7.673406896252013e-07,
      "loss": 0.7815761,
      "num_input_tokens_seen": 128742665,
      "step": 5988,
      "time_per_iteration": 3.6536662578582764
    },
    {
      "auxiliary_loss_clip": 0.01174763,
      "auxiliary_loss_mlp": 0.01027686,
      "balance_loss_clip": 1.04108751,
      "balance_loss_mlp": 1.01882899,
      "epoch": 0.7201346720375158,
      "flos": 25374264624000.0,
      "grad_norm": 1.802508897935261,
      "language_loss": 0.78422868,
      "learning_rate": 7.667273521120347e-07,
      "loss": 0.80625319,
      "num_input_tokens_seen": 128762225,
      "step": 5989,
      "time_per_iteration": 2.7144596576690674
    },
    {
      "auxiliary_loss_clip": 0.01181719,
      "auxiliary_loss_mlp": 0.0102908,
      "balance_loss_clip": 1.04482687,
      "balance_loss_mlp": 1.02139711,
      "epoch": 0.7202549149281549,
      "flos": 14355614499840.0,
      "grad_norm": 2.1491400031449714,
      "language_loss": 0.79469699,
      "learning_rate": 7.661142016853468e-07,
      "loss": 0.81680501,
      "num_input_tokens_seen": 128779585,
      "step": 5990,
      "time_per_iteration": 2.6422312259674072
    },
    {
      "auxiliary_loss_clip": 0.0117729,
      "auxiliary_loss_mlp": 0.01026962,
      "balance_loss_clip": 1.04225576,
      "balance_loss_mlp": 1.01952636,
      "epoch": 0.7203751578187939,
      "flos": 23001682550400.0,
      "grad_norm": 2.626844354441396,
      "language_loss": 0.75183129,
      "learning_rate": 7.655012384381543e-07,
      "loss": 0.77387381,
      "num_input_tokens_seen": 128799070,
      "step": 5991,
      "time_per_iteration": 3.6095364093780518
    },
    {
      "auxiliary_loss_clip": 0.01172583,
      "auxiliary_loss_mlp": 0.01027552,
      "balance_loss_clip": 1.0487783,
      "balance_loss_mlp": 1.01991045,
      "epoch": 0.7204954007094331,
      "flos": 23692424065920.0,
      "grad_norm": 1.9007397562165749,
      "language_loss": 0.81860685,
      "learning_rate": 7.648884624634415e-07,
      "loss": 0.84060824,
      "num_input_tokens_seen": 128817620,
      "step": 5992,
      "time_per_iteration": 2.680189371109009
    },
    {
      "auxiliary_loss_clip": 0.0116899,
      "auxiliary_loss_mlp": 0.01024749,
      "balance_loss_clip": 1.04618406,
      "balance_loss_mlp": 1.01698232,
      "epoch": 0.7206156436000721,
      "flos": 16253026531200.0,
      "grad_norm": 1.9887204027802279,
      "language_loss": 0.8908428,
      "learning_rate": 7.642758738541683e-07,
      "loss": 0.91278017,
      "num_input_tokens_seen": 128834200,
      "step": 5993,
      "time_per_iteration": 2.626441717147827
    },
    {
      "auxiliary_loss_clip": 0.01075175,
      "auxiliary_loss_mlp": 0.01003731,
      "balance_loss_clip": 1.01451373,
      "balance_loss_mlp": 1.00290251,
      "epoch": 0.7207358864907112,
      "flos": 54377806504320.0,
      "grad_norm": 0.7631947159593238,
      "language_loss": 0.60734051,
      "learning_rate": 7.636634727032621e-07,
      "loss": 0.62812954,
      "num_input_tokens_seen": 128891305,
      "step": 5994,
      "time_per_iteration": 3.0789849758148193
    },
    {
      "auxiliary_loss_clip": 0.01180298,
      "auxiliary_loss_mlp": 0.01026861,
      "balance_loss_clip": 1.04111958,
      "balance_loss_mlp": 1.01848352,
      "epoch": 0.7208561293813504,
      "flos": 19135540033920.0,
      "grad_norm": 2.237882818368626,
      "language_loss": 0.79266214,
      "learning_rate": 7.630512591036231e-07,
      "loss": 0.81473368,
      "num_input_tokens_seen": 128910615,
      "step": 5995,
      "time_per_iteration": 3.631411552429199
    },
    {
      "auxiliary_loss_clip": 0.01173682,
      "auxiliary_loss_mlp": 0.01029047,
      "balance_loss_clip": 1.04737759,
      "balance_loss_mlp": 1.0209403,
      "epoch": 0.7209763722719894,
      "flos": 17748526308480.0,
      "grad_norm": 3.516620328463838,
      "language_loss": 0.64792836,
      "learning_rate": 7.624392331481255e-07,
      "loss": 0.66995561,
      "num_input_tokens_seen": 128928270,
      "step": 5996,
      "time_per_iteration": 2.594287157058716
    },
    {
      "auxiliary_loss_clip": 0.01073541,
      "auxiliary_loss_mlp": 0.01001823,
      "balance_loss_clip": 1.01350272,
      "balance_loss_mlp": 1.00097048,
      "epoch": 0.7210966151626285,
      "flos": 66819488716800.0,
      "grad_norm": 0.7480837524823812,
      "language_loss": 0.51806521,
      "learning_rate": 7.618273949296115e-07,
      "loss": 0.53881884,
      "num_input_tokens_seen": 128987780,
      "step": 5997,
      "time_per_iteration": 3.0650715827941895
    },
    {
      "auxiliary_loss_clip": 0.01165905,
      "auxiliary_loss_mlp": 0.01025983,
      "balance_loss_clip": 1.0406251,
      "balance_loss_mlp": 1.01736438,
      "epoch": 0.7212168580532676,
      "flos": 21141869080320.0,
      "grad_norm": 2.024754837725949,
      "language_loss": 0.68537778,
      "learning_rate": 7.612157445408987e-07,
      "loss": 0.70729667,
      "num_input_tokens_seen": 129005590,
      "step": 5998,
      "time_per_iteration": 2.6427805423736572
    },
    {
      "auxiliary_loss_clip": 0.01178252,
      "auxiliary_loss_mlp": 0.01029922,
      "balance_loss_clip": 1.04780734,
      "balance_loss_mlp": 1.02163696,
      "epoch": 0.7213371009439067,
      "flos": 22345738335360.0,
      "grad_norm": 13.214863018649561,
      "language_loss": 0.7423209,
      "learning_rate": 7.606042820747716e-07,
      "loss": 0.76440263,
      "num_input_tokens_seen": 129021995,
      "step": 5999,
      "time_per_iteration": 2.6715786457061768
    },
    {
      "auxiliary_loss_clip": 0.01180872,
      "auxiliary_loss_mlp": 0.01022724,
      "balance_loss_clip": 1.04930723,
      "balance_loss_mlp": 1.01508844,
      "epoch": 0.7214573438345457,
      "flos": 18515901490560.0,
      "grad_norm": 1.8031635888295152,
      "language_loss": 0.85481352,
      "learning_rate": 7.599930076239889e-07,
      "loss": 0.87684947,
      "num_input_tokens_seen": 129039280,
      "step": 6000,
      "time_per_iteration": 2.6769580841064453
    },
    {
      "auxiliary_loss_clip": 0.01181803,
      "auxiliary_loss_mlp": 0.01208042,
      "balance_loss_clip": 1.04382229,
      "balance_loss_mlp": 1.00067639,
      "epoch": 0.7215775867251849,
      "flos": 35736108606720.0,
      "grad_norm": 2.1515440387127223,
      "language_loss": 0.70495284,
      "learning_rate": 7.593819212812818e-07,
      "loss": 0.72885132,
      "num_input_tokens_seen": 129060860,
      "step": 6001,
      "time_per_iteration": 2.8117942810058594
    },
    {
      "auxiliary_loss_clip": 0.01170966,
      "auxiliary_loss_mlp": 0.01025012,
      "balance_loss_clip": 1.04675937,
      "balance_loss_mlp": 1.01728749,
      "epoch": 0.721697829615824,
      "flos": 20372410909440.0,
      "grad_norm": 2.3222089370030776,
      "language_loss": 0.7179147,
      "learning_rate": 7.587710231393508e-07,
      "loss": 0.73987448,
      "num_input_tokens_seen": 129079215,
      "step": 6002,
      "time_per_iteration": 2.6387386322021484
    },
    {
      "auxiliary_loss_clip": 0.01173835,
      "auxiliary_loss_mlp": 0.01023084,
      "balance_loss_clip": 1.03772759,
      "balance_loss_mlp": 1.01550817,
      "epoch": 0.721818072506463,
      "flos": 20229809915520.0,
      "grad_norm": 3.3324285252169474,
      "language_loss": 0.83885449,
      "learning_rate": 7.581603132908685e-07,
      "loss": 0.86082363,
      "num_input_tokens_seen": 129097185,
      "step": 6003,
      "time_per_iteration": 2.778022527694702
    },
    {
      "auxiliary_loss_clip": 0.0117246,
      "auxiliary_loss_mlp": 0.01024144,
      "balance_loss_clip": 1.042943,
      "balance_loss_mlp": 1.01619256,
      "epoch": 0.7219383153971022,
      "flos": 18186887888640.0,
      "grad_norm": 2.10913995626305,
      "language_loss": 0.78561401,
      "learning_rate": 7.575497918284795e-07,
      "loss": 0.80758011,
      "num_input_tokens_seen": 129114730,
      "step": 6004,
      "time_per_iteration": 2.6312365531921387
    },
    {
      "auxiliary_loss_clip": 0.01172458,
      "auxiliary_loss_mlp": 0.01032544,
      "balance_loss_clip": 1.04824758,
      "balance_loss_mlp": 1.02444959,
      "epoch": 0.7220585582877412,
      "flos": 17342124854400.0,
      "grad_norm": 2.107743707099797,
      "language_loss": 0.74583137,
      "learning_rate": 7.569394588447984e-07,
      "loss": 0.76788139,
      "num_input_tokens_seen": 129131745,
      "step": 6005,
      "time_per_iteration": 2.6400492191314697
    },
    {
      "auxiliary_loss_clip": 0.01164002,
      "auxiliary_loss_mlp": 0.01028236,
      "balance_loss_clip": 1.04471779,
      "balance_loss_mlp": 1.0208869,
      "epoch": 0.7221788011783803,
      "flos": 16976338704000.0,
      "grad_norm": 2.9091701277474495,
      "language_loss": 0.78415936,
      "learning_rate": 7.563293144324146e-07,
      "loss": 0.80608177,
      "num_input_tokens_seen": 129147295,
      "step": 6006,
      "time_per_iteration": 2.5909531116485596
    },
    {
      "auxiliary_loss_clip": 0.01167902,
      "auxiliary_loss_mlp": 0.01026736,
      "balance_loss_clip": 1.048805,
      "balance_loss_mlp": 1.0194881,
      "epoch": 0.7222990440690195,
      "flos": 26286359702400.0,
      "grad_norm": 1.8649367089900504,
      "language_loss": 0.80063939,
      "learning_rate": 7.557193586838834e-07,
      "loss": 0.82258582,
      "num_input_tokens_seen": 129162660,
      "step": 6007,
      "time_per_iteration": 2.6346116065979004
    },
    {
      "auxiliary_loss_clip": 0.01176989,
      "auxiliary_loss_mlp": 0.0102471,
      "balance_loss_clip": 1.04353178,
      "balance_loss_mlp": 1.0169971,
      "epoch": 0.7224192869596585,
      "flos": 17601687509760.0,
      "grad_norm": 2.26536964373722,
      "language_loss": 0.70766401,
      "learning_rate": 7.551095916917371e-07,
      "loss": 0.72968102,
      "num_input_tokens_seen": 129179990,
      "step": 6008,
      "time_per_iteration": 2.654680013656616
    },
    {
      "auxiliary_loss_clip": 0.0118854,
      "auxiliary_loss_mlp": 0.01028936,
      "balance_loss_clip": 1.04429126,
      "balance_loss_mlp": 1.02044845,
      "epoch": 0.7225395298502976,
      "flos": 12932331016320.0,
      "grad_norm": 3.3873043548773216,
      "language_loss": 0.66025335,
      "learning_rate": 7.545000135484758e-07,
      "loss": 0.68242812,
      "num_input_tokens_seen": 129197425,
      "step": 6009,
      "time_per_iteration": 3.6545321941375732
    },
    {
      "auxiliary_loss_clip": 0.01170805,
      "auxiliary_loss_mlp": 0.01208219,
      "balance_loss_clip": 1.04927194,
      "balance_loss_mlp": 1.00066328,
      "epoch": 0.7226597727409367,
      "flos": 29643899592960.0,
      "grad_norm": 2.0812517927953706,
      "language_loss": 0.62674409,
      "learning_rate": 7.538906243465714e-07,
      "loss": 0.65053433,
      "num_input_tokens_seen": 129217560,
      "step": 6010,
      "time_per_iteration": 2.6704535484313965
    },
    {
      "auxiliary_loss_clip": 0.0117205,
      "auxiliary_loss_mlp": 0.01025526,
      "balance_loss_clip": 1.0500294,
      "balance_loss_mlp": 1.01750958,
      "epoch": 0.7227800156315758,
      "flos": 13771635183360.0,
      "grad_norm": 1.903117999275114,
      "language_loss": 0.7856859,
      "learning_rate": 7.5328142417847e-07,
      "loss": 0.80766171,
      "num_input_tokens_seen": 129234325,
      "step": 6011,
      "time_per_iteration": 2.545747756958008
    },
    {
      "auxiliary_loss_clip": 0.01168296,
      "auxiliary_loss_mlp": 0.01029051,
      "balance_loss_clip": 1.04404747,
      "balance_loss_mlp": 1.02177584,
      "epoch": 0.7229002585222148,
      "flos": 20301882554880.0,
      "grad_norm": 1.5862967255537979,
      "language_loss": 0.69331396,
      "learning_rate": 7.526724131365838e-07,
      "loss": 0.71528745,
      "num_input_tokens_seen": 129255280,
      "step": 6012,
      "time_per_iteration": 2.7522599697113037
    },
    {
      "auxiliary_loss_clip": 0.01174336,
      "auxiliary_loss_mlp": 0.01028729,
      "balance_loss_clip": 1.04840803,
      "balance_loss_mlp": 1.02044415,
      "epoch": 0.723020501412854,
      "flos": 16581250033920.0,
      "grad_norm": 2.4519956294579432,
      "language_loss": 0.70205605,
      "learning_rate": 7.520635913133017e-07,
      "loss": 0.72408676,
      "num_input_tokens_seen": 129273910,
      "step": 6013,
      "time_per_iteration": 2.6624326705932617
    },
    {
      "auxiliary_loss_clip": 0.01177314,
      "auxiliary_loss_mlp": 0.01029392,
      "balance_loss_clip": 1.04727876,
      "balance_loss_mlp": 1.02034998,
      "epoch": 0.7231407443034931,
      "flos": 28548300908160.0,
      "grad_norm": 2.0556278190096666,
      "language_loss": 0.82723725,
      "learning_rate": 7.514549588009798e-07,
      "loss": 0.84930426,
      "num_input_tokens_seen": 129294785,
      "step": 6014,
      "time_per_iteration": 2.821371078491211
    },
    {
      "auxiliary_loss_clip": 0.01176821,
      "auxiliary_loss_mlp": 0.01027138,
      "balance_loss_clip": 1.04615974,
      "balance_loss_mlp": 1.01984191,
      "epoch": 0.7232609871941321,
      "flos": 30008536508160.0,
      "grad_norm": 5.7307645345584985,
      "language_loss": 0.70252264,
      "learning_rate": 7.508465156919492e-07,
      "loss": 0.72456217,
      "num_input_tokens_seen": 129318295,
      "step": 6015,
      "time_per_iteration": 3.6603710651397705
    },
    {
      "auxiliary_loss_clip": 0.01175549,
      "auxiliary_loss_mlp": 0.01025581,
      "balance_loss_clip": 1.04512715,
      "balance_loss_mlp": 1.01759362,
      "epoch": 0.7233812300847713,
      "flos": 16654005031680.0,
      "grad_norm": 3.160846269135812,
      "language_loss": 0.61804485,
      "learning_rate": 7.502382620785083e-07,
      "loss": 0.64005613,
      "num_input_tokens_seen": 129334845,
      "step": 6016,
      "time_per_iteration": 2.6361005306243896
    },
    {
      "auxiliary_loss_clip": 0.01079699,
      "auxiliary_loss_mlp": 0.01001801,
      "balance_loss_clip": 1.01555049,
      "balance_loss_mlp": 1.00096035,
      "epoch": 0.7235014729754103,
      "flos": 67258784050560.0,
      "grad_norm": 0.8169918233874316,
      "language_loss": 0.62519491,
      "learning_rate": 7.496301980529289e-07,
      "loss": 0.64600992,
      "num_input_tokens_seen": 129398055,
      "step": 6017,
      "time_per_iteration": 3.2801313400268555
    },
    {
      "auxiliary_loss_clip": 0.01170086,
      "auxiliary_loss_mlp": 0.01028814,
      "balance_loss_clip": 1.04837632,
      "balance_loss_mlp": 1.02108884,
      "epoch": 0.7236217158660494,
      "flos": 26943237671040.0,
      "grad_norm": 5.175397251064765,
      "language_loss": 0.74584436,
      "learning_rate": 7.490223237074547e-07,
      "loss": 0.76783335,
      "num_input_tokens_seen": 129417765,
      "step": 6018,
      "time_per_iteration": 3.540426015853882
    },
    {
      "auxiliary_loss_clip": 0.0117528,
      "auxiliary_loss_mlp": 0.01031702,
      "balance_loss_clip": 1.0416801,
      "balance_loss_mlp": 1.02358437,
      "epoch": 0.7237419587566886,
      "flos": 29423372042880.0,
      "grad_norm": 2.2244879725947255,
      "language_loss": 0.66123867,
      "learning_rate": 7.484146391342989e-07,
      "loss": 0.68330848,
      "num_input_tokens_seen": 129437560,
      "step": 6019,
      "time_per_iteration": 2.7453219890594482
    },
    {
      "auxiliary_loss_clip": 0.01168459,
      "auxiliary_loss_mlp": 0.010236,
      "balance_loss_clip": 1.04384184,
      "balance_loss_mlp": 1.01586592,
      "epoch": 0.7238622016473276,
      "flos": 17821496787840.0,
      "grad_norm": 2.8929910785933353,
      "language_loss": 0.56565392,
      "learning_rate": 7.478071444256484e-07,
      "loss": 0.58757454,
      "num_input_tokens_seen": 129455320,
      "step": 6020,
      "time_per_iteration": 2.6565308570861816
    },
    {
      "auxiliary_loss_clip": 0.01187617,
      "auxiliary_loss_mlp": 0.01023781,
      "balance_loss_clip": 1.04456186,
      "balance_loss_mlp": 1.01595545,
      "epoch": 0.7239824445379667,
      "flos": 25739117020800.0,
      "grad_norm": 2.4294742486767933,
      "language_loss": 0.79340428,
      "learning_rate": 7.471998396736579e-07,
      "loss": 0.81551826,
      "num_input_tokens_seen": 129475700,
      "step": 6021,
      "time_per_iteration": 3.593287706375122
    },
    {
      "auxiliary_loss_clip": 0.01180726,
      "auxiliary_loss_mlp": 0.01021911,
      "balance_loss_clip": 1.04590189,
      "balance_loss_mlp": 1.01422834,
      "epoch": 0.7241026874286057,
      "flos": 23148916398720.0,
      "grad_norm": 1.6713348585644836,
      "language_loss": 0.7589981,
      "learning_rate": 7.465927249704549e-07,
      "loss": 0.78102446,
      "num_input_tokens_seen": 129493585,
      "step": 6022,
      "time_per_iteration": 2.6877541542053223
    },
    {
      "auxiliary_loss_clip": 0.01168945,
      "auxiliary_loss_mlp": 0.01023314,
      "balance_loss_clip": 1.04507172,
      "balance_loss_mlp": 1.01549411,
      "epoch": 0.7242229303192449,
      "flos": 20266905686400.0,
      "grad_norm": 2.9018963217507387,
      "language_loss": 0.77422196,
      "learning_rate": 7.459858004081398e-07,
      "loss": 0.79614455,
      "num_input_tokens_seen": 129511555,
      "step": 6023,
      "time_per_iteration": 2.6237664222717285
    },
    {
      "auxiliary_loss_clip": 0.01083249,
      "auxiliary_loss_mlp": 0.01001093,
      "balance_loss_clip": 1.01462412,
      "balance_loss_mlp": 1.00026405,
      "epoch": 0.724343173209884,
      "flos": 62311659684480.0,
      "grad_norm": 0.6823406432648205,
      "language_loss": 0.57998109,
      "learning_rate": 7.453790660787815e-07,
      "loss": 0.60082448,
      "num_input_tokens_seen": 129579650,
      "step": 6024,
      "time_per_iteration": 3.400099277496338
    },
    {
      "auxiliary_loss_clip": 0.01177367,
      "auxiliary_loss_mlp": 0.01023879,
      "balance_loss_clip": 1.04600525,
      "balance_loss_mlp": 1.01590979,
      "epoch": 0.724463416100523,
      "flos": 35006403813120.0,
      "grad_norm": 4.315190248509519,
      "language_loss": 0.63718653,
      "learning_rate": 7.447725220744214e-07,
      "loss": 0.65919888,
      "num_input_tokens_seen": 129601895,
      "step": 6025,
      "time_per_iteration": 2.778233289718628
    },
    {
      "auxiliary_loss_clip": 0.01169214,
      "auxiliary_loss_mlp": 0.01028723,
      "balance_loss_clip": 1.04631901,
      "balance_loss_mlp": 1.02068877,
      "epoch": 0.7245836589911622,
      "flos": 21871968923520.0,
      "grad_norm": 2.0985856758683448,
      "language_loss": 0.77581108,
      "learning_rate": 7.441661684870717e-07,
      "loss": 0.79779041,
      "num_input_tokens_seen": 129622150,
      "step": 6026,
      "time_per_iteration": 2.642775297164917
    },
    {
      "auxiliary_loss_clip": 0.0116965,
      "auxiliary_loss_mlp": 0.01022776,
      "balance_loss_clip": 1.0489738,
      "balance_loss_mlp": 1.01510513,
      "epoch": 0.7247039018818012,
      "flos": 23006494972800.0,
      "grad_norm": 1.7121335736436227,
      "language_loss": 0.82058263,
      "learning_rate": 7.435600054087152e-07,
      "loss": 0.84250689,
      "num_input_tokens_seen": 129644315,
      "step": 6027,
      "time_per_iteration": 2.6705093383789062
    },
    {
      "auxiliary_loss_clip": 0.01172824,
      "auxiliary_loss_mlp": 0.01028587,
      "balance_loss_clip": 1.05056632,
      "balance_loss_mlp": 1.02074897,
      "epoch": 0.7248241447724403,
      "flos": 31722588587520.0,
      "grad_norm": 2.031544225542044,
      "language_loss": 0.74315584,
      "learning_rate": 7.42954032931308e-07,
      "loss": 0.76516998,
      "num_input_tokens_seen": 129665355,
      "step": 6028,
      "time_per_iteration": 2.6678268909454346
    },
    {
      "auxiliary_loss_clip": 0.01175098,
      "auxiliary_loss_mlp": 0.01026761,
      "balance_loss_clip": 1.04389977,
      "balance_loss_mlp": 1.01914072,
      "epoch": 0.7249443876630794,
      "flos": 34896984007680.0,
      "grad_norm": 1.921062084947247,
      "language_loss": 0.7451672,
      "learning_rate": 7.423482511467733e-07,
      "loss": 0.76718581,
      "num_input_tokens_seen": 129686125,
      "step": 6029,
      "time_per_iteration": 2.8333098888397217
    },
    {
      "auxiliary_loss_clip": 0.01168712,
      "auxiliary_loss_mlp": 0.01025997,
      "balance_loss_clip": 1.03748345,
      "balance_loss_mlp": 1.01843357,
      "epoch": 0.7250646305537185,
      "flos": 26359294268160.0,
      "grad_norm": 2.189577683659321,
      "language_loss": 0.64610606,
      "learning_rate": 7.417426601470099e-07,
      "loss": 0.66805315,
      "num_input_tokens_seen": 129706485,
      "step": 6030,
      "time_per_iteration": 2.833522081375122
    },
    {
      "auxiliary_loss_clip": 0.01175017,
      "auxiliary_loss_mlp": 0.01026814,
      "balance_loss_clip": 1.04801047,
      "balance_loss_mlp": 1.01851737,
      "epoch": 0.7251848734443576,
      "flos": 30081614728320.0,
      "grad_norm": 3.02291797529816,
      "language_loss": 0.78758872,
      "learning_rate": 7.411372600238841e-07,
      "loss": 0.80960697,
      "num_input_tokens_seen": 129727100,
      "step": 6031,
      "time_per_iteration": 2.823366641998291
    },
    {
      "auxiliary_loss_clip": 0.01169105,
      "auxiliary_loss_mlp": 0.0102876,
      "balance_loss_clip": 1.04741597,
      "balance_loss_mlp": 1.02070785,
      "epoch": 0.7253051163349967,
      "flos": 17785262943360.0,
      "grad_norm": 3.7990449475705272,
      "language_loss": 0.74413538,
      "learning_rate": 7.405320508692346e-07,
      "loss": 0.766114,
      "num_input_tokens_seen": 129745840,
      "step": 6032,
      "time_per_iteration": 2.604851007461548
    },
    {
      "auxiliary_loss_clip": 0.01165214,
      "auxiliary_loss_mlp": 0.01023592,
      "balance_loss_clip": 1.04741609,
      "balance_loss_mlp": 1.01594472,
      "epoch": 0.7254253592256358,
      "flos": 12641346938880.0,
      "grad_norm": 2.05527225436336,
      "language_loss": 0.75737178,
      "learning_rate": 7.399270327748727e-07,
      "loss": 0.7792598,
      "num_input_tokens_seen": 129763500,
      "step": 6033,
      "time_per_iteration": 2.5962862968444824
    },
    {
      "auxiliary_loss_clip": 0.01176067,
      "auxiliary_loss_mlp": 0.01206806,
      "balance_loss_clip": 1.04127407,
      "balance_loss_mlp": 1.00067532,
      "epoch": 0.7255456021162748,
      "flos": 27199208966400.0,
      "grad_norm": 1.7339122131815128,
      "language_loss": 0.74361193,
      "learning_rate": 7.39322205832577e-07,
      "loss": 0.76744068,
      "num_input_tokens_seen": 129784390,
      "step": 6034,
      "time_per_iteration": 2.7210114002227783
    },
    {
      "auxiliary_loss_clip": 0.01169205,
      "auxiliary_loss_mlp": 0.01021724,
      "balance_loss_clip": 1.04365301,
      "balance_loss_mlp": 1.01396334,
      "epoch": 0.725665845006914,
      "flos": 21288205088640.0,
      "grad_norm": 2.3268348676097688,
      "language_loss": 0.80852693,
      "learning_rate": 7.387175701341009e-07,
      "loss": 0.83043623,
      "num_input_tokens_seen": 129803060,
      "step": 6035,
      "time_per_iteration": 3.689413547515869
    },
    {
      "auxiliary_loss_clip": 0.0117063,
      "auxiliary_loss_mlp": 0.01020831,
      "balance_loss_clip": 1.0449326,
      "balance_loss_mlp": 1.01297474,
      "epoch": 0.7257860878975531,
      "flos": 16033684129920.0,
      "grad_norm": 2.713221028371312,
      "language_loss": 0.72445649,
      "learning_rate": 7.381131257711659e-07,
      "loss": 0.74637115,
      "num_input_tokens_seen": 129820165,
      "step": 6036,
      "time_per_iteration": 2.647446393966675
    },
    {
      "auxiliary_loss_clip": 0.01172826,
      "auxiliary_loss_mlp": 0.01028841,
      "balance_loss_clip": 1.0496397,
      "balance_loss_mlp": 1.02134323,
      "epoch": 0.7259063307881921,
      "flos": 12129943052160.0,
      "grad_norm": 1.7489560677756872,
      "language_loss": 0.84247619,
      "learning_rate": 7.375088728354677e-07,
      "loss": 0.86449283,
      "num_input_tokens_seen": 129835195,
      "step": 6037,
      "time_per_iteration": 2.587669610977173
    },
    {
      "auxiliary_loss_clip": 0.01179487,
      "auxiliary_loss_mlp": 0.010231,
      "balance_loss_clip": 1.04296613,
      "balance_loss_mlp": 1.01504123,
      "epoch": 0.7260265736788313,
      "flos": 30443845432320.0,
      "grad_norm": 1.594857323586159,
      "language_loss": 0.67565948,
      "learning_rate": 7.369048114186691e-07,
      "loss": 0.69768536,
      "num_input_tokens_seen": 129856240,
      "step": 6038,
      "time_per_iteration": 2.7642619609832764
    },
    {
      "auxiliary_loss_clip": 0.01185799,
      "auxiliary_loss_mlp": 0.01207535,
      "balance_loss_clip": 1.04571104,
      "balance_loss_mlp": 1.000669,
      "epoch": 0.7261468165694703,
      "flos": 21142264129920.0,
      "grad_norm": 1.8593674038939434,
      "language_loss": 0.83409727,
      "learning_rate": 7.363009416124055e-07,
      "loss": 0.85803056,
      "num_input_tokens_seen": 129875565,
      "step": 6039,
      "time_per_iteration": 2.6169800758361816
    },
    {
      "auxiliary_loss_clip": 0.0117889,
      "auxiliary_loss_mlp": 0.0102673,
      "balance_loss_clip": 1.04423523,
      "balance_loss_mlp": 1.01888597,
      "epoch": 0.7262670594601094,
      "flos": 22306308180480.0,
      "grad_norm": 2.287974976595948,
      "language_loss": 0.6313588,
      "learning_rate": 7.356972635082852e-07,
      "loss": 0.65341502,
      "num_input_tokens_seen": 129894420,
      "step": 6040,
      "time_per_iteration": 2.666898012161255
    },
    {
      "auxiliary_loss_clip": 0.01176272,
      "auxiliary_loss_mlp": 0.01025812,
      "balance_loss_clip": 1.0444057,
      "balance_loss_mlp": 1.01786721,
      "epoch": 0.7263873023507486,
      "flos": 25335049950720.0,
      "grad_norm": 3.462434986001913,
      "language_loss": 0.7527228,
      "learning_rate": 7.35093777197884e-07,
      "loss": 0.77474368,
      "num_input_tokens_seen": 129914490,
      "step": 6041,
      "time_per_iteration": 3.6051924228668213
    },
    {
      "auxiliary_loss_clip": 0.01170687,
      "auxiliary_loss_mlp": 0.01020632,
      "balance_loss_clip": 1.04550958,
      "balance_loss_mlp": 1.01316345,
      "epoch": 0.7265075452413876,
      "flos": 23878621192320.0,
      "grad_norm": 2.2250050820375384,
      "language_loss": 0.85789692,
      "learning_rate": 7.344904827727525e-07,
      "loss": 0.87981009,
      "num_input_tokens_seen": 129931670,
      "step": 6042,
      "time_per_iteration": 2.6127853393554688
    },
    {
      "auxiliary_loss_clip": 0.01179273,
      "auxiliary_loss_mlp": 0.01025396,
      "balance_loss_clip": 1.04303491,
      "balance_loss_mlp": 1.0176239,
      "epoch": 0.7266277881320267,
      "flos": 28724549967360.0,
      "grad_norm": 2.3301147978542263,
      "language_loss": 0.7351042,
      "learning_rate": 7.338873803244076e-07,
      "loss": 0.75715089,
      "num_input_tokens_seen": 129946905,
      "step": 6043,
      "time_per_iteration": 2.722466230392456
    },
    {
      "auxiliary_loss_clip": 0.01169884,
      "auxiliary_loss_mlp": 0.01024475,
      "balance_loss_clip": 1.04551196,
      "balance_loss_mlp": 1.01718259,
      "epoch": 0.7267480310226658,
      "flos": 24863507182080.0,
      "grad_norm": 2.381274142994816,
      "language_loss": 0.81127346,
      "learning_rate": 7.332844699443401e-07,
      "loss": 0.83321708,
      "num_input_tokens_seen": 129965505,
      "step": 6044,
      "time_per_iteration": 3.651366710662842
    },
    {
      "auxiliary_loss_clip": 0.01171408,
      "auxiliary_loss_mlp": 0.01024489,
      "balance_loss_clip": 1.04085791,
      "balance_loss_mlp": 1.01733041,
      "epoch": 0.7268682739133049,
      "flos": 27198490694400.0,
      "grad_norm": 1.7100782918077537,
      "language_loss": 0.75678897,
      "learning_rate": 7.326817517240121e-07,
      "loss": 0.77874792,
      "num_input_tokens_seen": 129987210,
      "step": 6045,
      "time_per_iteration": 2.7645299434661865
    },
    {
      "auxiliary_loss_clip": 0.01170898,
      "auxiliary_loss_mlp": 0.01207223,
      "balance_loss_clip": 1.04627752,
      "balance_loss_mlp": 1.00070405,
      "epoch": 0.7269885168039439,
      "flos": 33508138688640.0,
      "grad_norm": 1.7894787692784369,
      "language_loss": 0.83624512,
      "learning_rate": 7.320792257548545e-07,
      "loss": 0.86002636,
      "num_input_tokens_seen": 130008385,
      "step": 6046,
      "time_per_iteration": 2.7635130882263184
    },
    {
      "auxiliary_loss_clip": 0.011781,
      "auxiliary_loss_mlp": 0.01026532,
      "balance_loss_clip": 1.04549694,
      "balance_loss_mlp": 1.01859248,
      "epoch": 0.7271087596945831,
      "flos": 24313750548480.0,
      "grad_norm": 2.0132346671922603,
      "language_loss": 0.76769292,
      "learning_rate": 7.314768921282704e-07,
      "loss": 0.78973925,
      "num_input_tokens_seen": 130029040,
      "step": 6047,
      "time_per_iteration": 2.6720330715179443
    },
    {
      "auxiliary_loss_clip": 0.0117205,
      "auxiliary_loss_mlp": 0.0102345,
      "balance_loss_clip": 1.04545164,
      "balance_loss_mlp": 1.01599979,
      "epoch": 0.7272290025852222,
      "flos": 23805147922560.0,
      "grad_norm": 3.092049090143684,
      "language_loss": 0.73150897,
      "learning_rate": 7.30874750935633e-07,
      "loss": 0.75346392,
      "num_input_tokens_seen": 130048725,
      "step": 6048,
      "time_per_iteration": 3.598059892654419
    },
    {
      "auxiliary_loss_clip": 0.01176107,
      "auxiliary_loss_mlp": 0.01025849,
      "balance_loss_clip": 1.04543304,
      "balance_loss_mlp": 1.01806164,
      "epoch": 0.7273492454758612,
      "flos": 16720367408640.0,
      "grad_norm": 2.127106812358025,
      "language_loss": 0.79243141,
      "learning_rate": 7.30272802268286e-07,
      "loss": 0.81445092,
      "num_input_tokens_seen": 130065720,
      "step": 6049,
      "time_per_iteration": 2.658071279525757
    },
    {
      "auxiliary_loss_clip": 0.01154768,
      "auxiliary_loss_mlp": 0.01021828,
      "balance_loss_clip": 1.03597033,
      "balance_loss_mlp": 1.01452613,
      "epoch": 0.7274694883665004,
      "flos": 28031330413440.0,
      "grad_norm": 2.4203794895785453,
      "language_loss": 0.76525348,
      "learning_rate": 7.29671046217547e-07,
      "loss": 0.78701943,
      "num_input_tokens_seen": 130084830,
      "step": 6050,
      "time_per_iteration": 2.7880470752716064
    },
    {
      "auxiliary_loss_clip": 0.01175821,
      "auxiliary_loss_mlp": 0.0102689,
      "balance_loss_clip": 1.04387331,
      "balance_loss_mlp": 1.01965714,
      "epoch": 0.7275897312571394,
      "flos": 30372706546560.0,
      "grad_norm": 1.8517689936173551,
      "language_loss": 0.81862414,
      "learning_rate": 7.290694828746988e-07,
      "loss": 0.84065127,
      "num_input_tokens_seen": 130104495,
      "step": 6051,
      "time_per_iteration": 2.7801733016967773
    },
    {
      "auxiliary_loss_clip": 0.01179195,
      "auxiliary_loss_mlp": 0.01025334,
      "balance_loss_clip": 1.04181647,
      "balance_loss_mlp": 1.01726365,
      "epoch": 0.7277099741477785,
      "flos": 19204775498880.0,
      "grad_norm": 1.9732112388997967,
      "language_loss": 0.85980749,
      "learning_rate": 7.284681123310004e-07,
      "loss": 0.88185281,
      "num_input_tokens_seen": 130123210,
      "step": 6052,
      "time_per_iteration": 2.6420464515686035
    },
    {
      "auxiliary_loss_clip": 0.01170361,
      "auxiliary_loss_mlp": 0.01028903,
      "balance_loss_clip": 1.04668951,
      "balance_loss_mlp": 1.02049851,
      "epoch": 0.7278302170384175,
      "flos": 20667884186880.0,
      "grad_norm": 1.882757819742419,
      "language_loss": 0.79563457,
      "learning_rate": 7.27866934677678e-07,
      "loss": 0.81762725,
      "num_input_tokens_seen": 130142880,
      "step": 6053,
      "time_per_iteration": 2.6772356033325195
    },
    {
      "auxiliary_loss_clip": 0.01172376,
      "auxiliary_loss_mlp": 0.01022722,
      "balance_loss_clip": 1.04130518,
      "balance_loss_mlp": 1.01476753,
      "epoch": 0.7279504599290567,
      "flos": 19093200877440.0,
      "grad_norm": 1.7709456863994593,
      "language_loss": 0.78613555,
      "learning_rate": 7.272659500059297e-07,
      "loss": 0.80808651,
      "num_input_tokens_seen": 130160220,
      "step": 6054,
      "time_per_iteration": 2.640401840209961
    },
    {
      "auxiliary_loss_clip": 0.0116576,
      "auxiliary_loss_mlp": 0.01035846,
      "balance_loss_clip": 1.04627347,
      "balance_loss_mlp": 1.02825832,
      "epoch": 0.7280707028196958,
      "flos": 19062174504960.0,
      "grad_norm": 2.1576758862159964,
      "language_loss": 0.80504566,
      "learning_rate": 7.266651584069264e-07,
      "loss": 0.82706177,
      "num_input_tokens_seen": 130177885,
      "step": 6055,
      "time_per_iteration": 2.6671547889709473
    },
    {
      "auxiliary_loss_clip": 0.011751,
      "auxiliary_loss_mlp": 0.01020826,
      "balance_loss_clip": 1.04922199,
      "balance_loss_mlp": 1.01314342,
      "epoch": 0.7281909457103348,
      "flos": 37196308293120.0,
      "grad_norm": 1.693416151883688,
      "language_loss": 0.57223099,
      "learning_rate": 7.260645599718045e-07,
      "loss": 0.59419024,
      "num_input_tokens_seen": 130204240,
      "step": 6056,
      "time_per_iteration": 2.754126787185669
    },
    {
      "auxiliary_loss_clip": 0.01175666,
      "auxiliary_loss_mlp": 0.01028015,
      "balance_loss_clip": 1.04570651,
      "balance_loss_mlp": 1.01958084,
      "epoch": 0.728311188600974,
      "flos": 20667094087680.0,
      "grad_norm": 2.1402437204592535,
      "language_loss": 0.67696518,
      "learning_rate": 7.254641547916767e-07,
      "loss": 0.69900203,
      "num_input_tokens_seen": 130221735,
      "step": 6057,
      "time_per_iteration": 2.626509428024292
    },
    {
      "auxiliary_loss_clip": 0.01169412,
      "auxiliary_loss_mlp": 0.0102792,
      "balance_loss_clip": 1.04912782,
      "balance_loss_mlp": 1.02026367,
      "epoch": 0.728431431491613,
      "flos": 28840685616000.0,
      "grad_norm": 2.0843194765658324,
      "language_loss": 0.69362503,
      "learning_rate": 7.248639429576226e-07,
      "loss": 0.71559834,
      "num_input_tokens_seen": 130241190,
      "step": 6058,
      "time_per_iteration": 2.6458983421325684
    },
    {
      "auxiliary_loss_clip": 0.01171493,
      "auxiliary_loss_mlp": 0.01028527,
      "balance_loss_clip": 1.04552937,
      "balance_loss_mlp": 1.02058125,
      "epoch": 0.7285516743822521,
      "flos": 25991856092160.0,
      "grad_norm": 1.7618378000930508,
      "language_loss": 0.72283518,
      "learning_rate": 7.242639245606959e-07,
      "loss": 0.74483538,
      "num_input_tokens_seen": 130260980,
      "step": 6059,
      "time_per_iteration": 2.6955485343933105
    },
    {
      "auxiliary_loss_clip": 0.01179958,
      "auxiliary_loss_mlp": 0.01026182,
      "balance_loss_clip": 1.04434156,
      "balance_loss_mlp": 1.01839519,
      "epoch": 0.7286719172728913,
      "flos": 16399721675520.0,
      "grad_norm": 1.7932813378341474,
      "language_loss": 0.82473087,
      "learning_rate": 7.236640996919168e-07,
      "loss": 0.84679234,
      "num_input_tokens_seen": 130280025,
      "step": 6060,
      "time_per_iteration": 2.5910749435424805
    },
    {
      "auxiliary_loss_clip": 0.01173075,
      "auxiliary_loss_mlp": 0.01022831,
      "balance_loss_clip": 1.0460062,
      "balance_loss_mlp": 1.01540399,
      "epoch": 0.7287921601635303,
      "flos": 22018161277440.0,
      "grad_norm": 1.900294486648899,
      "language_loss": 0.70705914,
      "learning_rate": 7.230644684422782e-07,
      "loss": 0.72901821,
      "num_input_tokens_seen": 130300255,
      "step": 6061,
      "time_per_iteration": 2.6951181888580322
    },
    {
      "auxiliary_loss_clip": 0.01177426,
      "auxiliary_loss_mlp": 0.0102486,
      "balance_loss_clip": 1.04408574,
      "balance_loss_mlp": 1.01744497,
      "epoch": 0.7289124030541694,
      "flos": 24600927784320.0,
      "grad_norm": 2.2449159463506416,
      "language_loss": 0.81551564,
      "learning_rate": 7.224650309027451e-07,
      "loss": 0.83753848,
      "num_input_tokens_seen": 130320005,
      "step": 6062,
      "time_per_iteration": 3.642692804336548
    },
    {
      "auxiliary_loss_clip": 0.01176338,
      "auxiliary_loss_mlp": 0.0102463,
      "balance_loss_clip": 1.04977679,
      "balance_loss_mlp": 1.01709867,
      "epoch": 0.7290326459448085,
      "flos": 21393638484480.0,
      "grad_norm": 1.8085496797069813,
      "language_loss": 0.68354911,
      "learning_rate": 7.218657871642506e-07,
      "loss": 0.70555878,
      "num_input_tokens_seen": 130338810,
      "step": 6063,
      "time_per_iteration": 2.6189587116241455
    },
    {
      "auxiliary_loss_clip": 0.01171459,
      "auxiliary_loss_mlp": 0.01030584,
      "balance_loss_clip": 1.04862046,
      "balance_loss_mlp": 1.02271628,
      "epoch": 0.7291528888354476,
      "flos": 18587686821120.0,
      "grad_norm": 2.4184232777154486,
      "language_loss": 0.62257522,
      "learning_rate": 7.212667373177012e-07,
      "loss": 0.64459562,
      "num_input_tokens_seen": 130353805,
      "step": 6064,
      "time_per_iteration": 2.586331844329834
    },
    {
      "auxiliary_loss_clip": 0.011752,
      "auxiliary_loss_mlp": 0.01023546,
      "balance_loss_clip": 1.04294384,
      "balance_loss_mlp": 1.01583946,
      "epoch": 0.7292731317260867,
      "flos": 18951066760320.0,
      "grad_norm": 1.7883113103187156,
      "language_loss": 0.75312006,
      "learning_rate": 7.206678814539704e-07,
      "loss": 0.7751075,
      "num_input_tokens_seen": 130372105,
      "step": 6065,
      "time_per_iteration": 2.600961208343506
    },
    {
      "auxiliary_loss_clip": 0.0118725,
      "auxiliary_loss_mlp": 0.01023001,
      "balance_loss_clip": 1.04363358,
      "balance_loss_mlp": 1.01607227,
      "epoch": 0.7293933746167258,
      "flos": 21067569797760.0,
      "grad_norm": 1.50972904434719,
      "language_loss": 0.73072493,
      "learning_rate": 7.20069219663904e-07,
      "loss": 0.75282741,
      "num_input_tokens_seen": 130391990,
      "step": 6066,
      "time_per_iteration": 2.72440242767334
    },
    {
      "auxiliary_loss_clip": 0.01173888,
      "auxiliary_loss_mlp": 0.01022755,
      "balance_loss_clip": 1.04474509,
      "balance_loss_mlp": 1.01504266,
      "epoch": 0.7295136175073649,
      "flos": 22453326547200.0,
      "grad_norm": 2.5646158669632437,
      "language_loss": 0.79651904,
      "learning_rate": 7.1947075203832e-07,
      "loss": 0.8184855,
      "num_input_tokens_seen": 130411970,
      "step": 6067,
      "time_per_iteration": 2.6492133140563965
    },
    {
      "auxiliary_loss_clip": 0.01066859,
      "auxiliary_loss_mlp": 0.01003266,
      "balance_loss_clip": 1.01299965,
      "balance_loss_mlp": 1.00241935,
      "epoch": 0.7296338603980039,
      "flos": 56125506648960.0,
      "grad_norm": 0.9397794363096836,
      "language_loss": 0.6004132,
      "learning_rate": 7.188724786680049e-07,
      "loss": 0.62111437,
      "num_input_tokens_seen": 130472440,
      "step": 6068,
      "time_per_iteration": 4.106595516204834
    },
    {
      "auxiliary_loss_clip": 0.01172948,
      "auxiliary_loss_mlp": 0.010246,
      "balance_loss_clip": 1.04311669,
      "balance_loss_mlp": 1.01658916,
      "epoch": 0.7297541032886431,
      "flos": 25228287751680.0,
      "grad_norm": 2.011907760089191,
      "language_loss": 0.75903487,
      "learning_rate": 7.182743996437162e-07,
      "loss": 0.78101039,
      "num_input_tokens_seen": 130491975,
      "step": 6069,
      "time_per_iteration": 2.702606439590454
    },
    {
      "auxiliary_loss_clip": 0.01183413,
      "auxiliary_loss_mlp": 0.01023709,
      "balance_loss_clip": 1.04136407,
      "balance_loss_mlp": 1.01534641,
      "epoch": 0.7298743461792822,
      "flos": 26467600752000.0,
      "grad_norm": 2.2702994388587205,
      "language_loss": 0.68504643,
      "learning_rate": 7.176765150561819e-07,
      "loss": 0.70711762,
      "num_input_tokens_seen": 130510580,
      "step": 6070,
      "time_per_iteration": 2.779296636581421
    },
    {
      "auxiliary_loss_clip": 0.01170444,
      "auxiliary_loss_mlp": 0.01024375,
      "balance_loss_clip": 1.04647541,
      "balance_loss_mlp": 1.01656651,
      "epoch": 0.7299945890699212,
      "flos": 19569053278080.0,
      "grad_norm": 3.5898727971570197,
      "language_loss": 0.79884911,
      "learning_rate": 7.170788249961002e-07,
      "loss": 0.82079732,
      "num_input_tokens_seen": 130529090,
      "step": 6071,
      "time_per_iteration": 3.6017584800720215
    },
    {
      "auxiliary_loss_clip": 0.01165477,
      "auxiliary_loss_mlp": 0.01024504,
      "balance_loss_clip": 1.04620302,
      "balance_loss_mlp": 1.01707101,
      "epoch": 0.7301148319605604,
      "flos": 22928963466240.0,
      "grad_norm": 2.124958691107467,
      "language_loss": 0.88224667,
      "learning_rate": 7.164813295541418e-07,
      "loss": 0.90414655,
      "num_input_tokens_seen": 130548655,
      "step": 6072,
      "time_per_iteration": 2.5629706382751465
    },
    {
      "auxiliary_loss_clip": 0.01175935,
      "auxiliary_loss_mlp": 0.0102757,
      "balance_loss_clip": 1.04407406,
      "balance_loss_mlp": 1.02016139,
      "epoch": 0.7302350748511994,
      "flos": 25369703596800.0,
      "grad_norm": 2.402743886262588,
      "language_loss": 0.70065093,
      "learning_rate": 7.15884028820944e-07,
      "loss": 0.72268605,
      "num_input_tokens_seen": 130567710,
      "step": 6073,
      "time_per_iteration": 2.651902198791504
    },
    {
      "auxiliary_loss_clip": 0.01171675,
      "auxiliary_loss_mlp": 0.01023239,
      "balance_loss_clip": 1.04071689,
      "balance_loss_mlp": 1.01584172,
      "epoch": 0.7303553177418385,
      "flos": 27819170732160.0,
      "grad_norm": 2.4227645683048786,
      "language_loss": 0.60836822,
      "learning_rate": 7.152869228871185e-07,
      "loss": 0.63031733,
      "num_input_tokens_seen": 130590195,
      "step": 6074,
      "time_per_iteration": 2.7105233669281006
    },
    {
      "auxiliary_loss_clip": 0.01168951,
      "auxiliary_loss_mlp": 0.01029767,
      "balance_loss_clip": 1.04409242,
      "balance_loss_mlp": 1.02165473,
      "epoch": 0.7304755606324776,
      "flos": 24426510318720.0,
      "grad_norm": 2.170201648783621,
      "language_loss": 0.72552133,
      "learning_rate": 7.146900118432457e-07,
      "loss": 0.74750853,
      "num_input_tokens_seen": 130609940,
      "step": 6075,
      "time_per_iteration": 3.584022283554077
    },
    {
      "auxiliary_loss_clip": 0.01182271,
      "auxiliary_loss_mlp": 0.01029339,
      "balance_loss_clip": 1.03543496,
      "balance_loss_mlp": 1.02207911,
      "epoch": 0.7305958035231167,
      "flos": 23840483927040.0,
      "grad_norm": 1.7007165278754275,
      "language_loss": 0.85720789,
      "learning_rate": 7.140932957798753e-07,
      "loss": 0.87932402,
      "num_input_tokens_seen": 130628380,
      "step": 6076,
      "time_per_iteration": 2.66058611869812
    },
    {
      "auxiliary_loss_clip": 0.01176896,
      "auxiliary_loss_mlp": 0.01024526,
      "balance_loss_clip": 1.04321122,
      "balance_loss_mlp": 1.01687884,
      "epoch": 0.7307160464137558,
      "flos": 16726939597440.0,
      "grad_norm": 1.9726700672617432,
      "language_loss": 0.71404511,
      "learning_rate": 7.134967747875309e-07,
      "loss": 0.73605937,
      "num_input_tokens_seen": 130646590,
      "step": 6077,
      "time_per_iteration": 2.4986400604248047
    },
    {
      "auxiliary_loss_clip": 0.01166934,
      "auxiliary_loss_mlp": 0.01026199,
      "balance_loss_clip": 1.04413962,
      "balance_loss_mlp": 1.01856947,
      "epoch": 0.7308362893043949,
      "flos": 21798280172160.0,
      "grad_norm": 2.0013091800926253,
      "language_loss": 0.81815392,
      "learning_rate": 7.129004489567014e-07,
      "loss": 0.84008527,
      "num_input_tokens_seen": 130664070,
      "step": 6078,
      "time_per_iteration": 2.5649893283843994
    },
    {
      "auxiliary_loss_clip": 0.01180426,
      "auxiliary_loss_mlp": 0.010237,
      "balance_loss_clip": 1.04316533,
      "balance_loss_mlp": 1.01601696,
      "epoch": 0.730956532195034,
      "flos": 10707377840640.0,
      "grad_norm": 3.0020033742723475,
      "language_loss": 0.78051436,
      "learning_rate": 7.123043183778512e-07,
      "loss": 0.80255562,
      "num_input_tokens_seen": 130681400,
      "step": 6079,
      "time_per_iteration": 2.617706775665283
    },
    {
      "auxiliary_loss_clip": 0.01185052,
      "auxiliary_loss_mlp": 0.0103497,
      "balance_loss_clip": 1.04663706,
      "balance_loss_mlp": 1.02739096,
      "epoch": 0.731076775085673,
      "flos": 19791987039360.0,
      "grad_norm": 1.6329040449873569,
      "language_loss": 0.65623206,
      "learning_rate": 7.117083831414114e-07,
      "loss": 0.67843229,
      "num_input_tokens_seen": 130700675,
      "step": 6080,
      "time_per_iteration": 2.6315791606903076
    },
    {
      "auxiliary_loss_clip": 0.01166618,
      "auxiliary_loss_mlp": 0.01024708,
      "balance_loss_clip": 1.04800177,
      "balance_loss_mlp": 1.01711464,
      "epoch": 0.7311970179763122,
      "flos": 20447033414400.0,
      "grad_norm": 1.948949269932605,
      "language_loss": 0.69780695,
      "learning_rate": 7.11112643337787e-07,
      "loss": 0.71972018,
      "num_input_tokens_seen": 130719720,
      "step": 6081,
      "time_per_iteration": 2.5510735511779785
    },
    {
      "auxiliary_loss_clip": 0.01175203,
      "auxiliary_loss_mlp": 0.01031429,
      "balance_loss_clip": 1.04734182,
      "balance_loss_mlp": 1.02304292,
      "epoch": 0.7313172608669513,
      "flos": 18513818501760.0,
      "grad_norm": 3.025839194376698,
      "language_loss": 0.76617259,
      "learning_rate": 7.10517099057349e-07,
      "loss": 0.78823888,
      "num_input_tokens_seen": 130736670,
      "step": 6082,
      "time_per_iteration": 2.6240649223327637
    },
    {
      "auxiliary_loss_clip": 0.01171395,
      "auxiliary_loss_mlp": 0.01022813,
      "balance_loss_clip": 1.0427196,
      "balance_loss_mlp": 1.01443887,
      "epoch": 0.7314375037575903,
      "flos": 16180738410240.0,
      "grad_norm": 2.2644149142691314,
      "language_loss": 0.61624694,
      "learning_rate": 7.099217503904411e-07,
      "loss": 0.63818902,
      "num_input_tokens_seen": 130754525,
      "step": 6083,
      "time_per_iteration": 2.57378888130188
    },
    {
      "auxiliary_loss_clip": 0.01176933,
      "auxiliary_loss_mlp": 0.01023919,
      "balance_loss_clip": 1.04567027,
      "balance_loss_mlp": 1.01638508,
      "epoch": 0.7315577466482295,
      "flos": 17967940536960.0,
      "grad_norm": 1.971071497119586,
      "language_loss": 0.8977496,
      "learning_rate": 7.093265974273788e-07,
      "loss": 0.9197582,
      "num_input_tokens_seen": 130772420,
      "step": 6084,
      "time_per_iteration": 2.6594865322113037
    },
    {
      "auxiliary_loss_clip": 0.01171705,
      "auxiliary_loss_mlp": 0.01021742,
      "balance_loss_clip": 1.04498649,
      "balance_loss_mlp": 1.01439595,
      "epoch": 0.7316779895388685,
      "flos": 18405440190720.0,
      "grad_norm": 2.375995478795891,
      "language_loss": 0.72158003,
      "learning_rate": 7.087316402584447e-07,
      "loss": 0.74351454,
      "num_input_tokens_seen": 130791245,
      "step": 6085,
      "time_per_iteration": 2.584831953048706
    },
    {
      "auxiliary_loss_clip": 0.01167705,
      "auxiliary_loss_mlp": 0.01021061,
      "balance_loss_clip": 1.04705262,
      "balance_loss_mlp": 1.01326823,
      "epoch": 0.7317982324295076,
      "flos": 17928294900480.0,
      "grad_norm": 4.833338344470221,
      "language_loss": 0.86157048,
      "learning_rate": 7.081368789738953e-07,
      "loss": 0.88345814,
      "num_input_tokens_seen": 130808445,
      "step": 6086,
      "time_per_iteration": 2.595855951309204
    },
    {
      "auxiliary_loss_clip": 0.01167916,
      "auxiliary_loss_mlp": 0.01024041,
      "balance_loss_clip": 1.04054284,
      "balance_loss_mlp": 1.01596427,
      "epoch": 0.7319184753201466,
      "flos": 27229840289280.0,
      "grad_norm": 2.4118504381217276,
      "language_loss": 0.77664495,
      "learning_rate": 7.075423136639537e-07,
      "loss": 0.79856443,
      "num_input_tokens_seen": 130827700,
      "step": 6087,
      "time_per_iteration": 2.640087604522705
    },
    {
      "auxiliary_loss_clip": 0.01167303,
      "auxiliary_loss_mlp": 0.01022757,
      "balance_loss_clip": 1.03937721,
      "balance_loss_mlp": 1.01494622,
      "epoch": 0.7320387182107858,
      "flos": 37448544574080.0,
      "grad_norm": 1.7032880814841944,
      "language_loss": 0.74749959,
      "learning_rate": 7.069479444188149e-07,
      "loss": 0.76940018,
      "num_input_tokens_seen": 130848290,
      "step": 6088,
      "time_per_iteration": 3.7679195404052734
    },
    {
      "auxiliary_loss_clip": 0.01166964,
      "auxiliary_loss_mlp": 0.01024375,
      "balance_loss_clip": 1.04459989,
      "balance_loss_mlp": 1.01606059,
      "epoch": 0.7321589611014249,
      "flos": 17859023521920.0,
      "grad_norm": 2.0038948172246838,
      "language_loss": 0.8202799,
      "learning_rate": 7.063537713286453e-07,
      "loss": 0.84219331,
      "num_input_tokens_seen": 130865970,
      "step": 6089,
      "time_per_iteration": 2.678823947906494
    },
    {
      "auxiliary_loss_clip": 0.01179313,
      "auxiliary_loss_mlp": 0.01022772,
      "balance_loss_clip": 1.04439771,
      "balance_loss_mlp": 1.01490164,
      "epoch": 0.7322792039920639,
      "flos": 26100593539200.0,
      "grad_norm": 2.5301170412937193,
      "language_loss": 0.80995834,
      "learning_rate": 7.057597944835803e-07,
      "loss": 0.83197916,
      "num_input_tokens_seen": 130885245,
      "step": 6090,
      "time_per_iteration": 2.9008102416992188
    },
    {
      "auxiliary_loss_clip": 0.01183698,
      "auxiliary_loss_mlp": 0.01022817,
      "balance_loss_clip": 1.04355514,
      "balance_loss_mlp": 1.01541972,
      "epoch": 0.7323994468827031,
      "flos": 25369093065600.0,
      "grad_norm": 1.7152544402386882,
      "language_loss": 0.74643791,
      "learning_rate": 7.051660139737253e-07,
      "loss": 0.76850307,
      "num_input_tokens_seen": 130903465,
      "step": 6091,
      "time_per_iteration": 2.740128755569458
    },
    {
      "auxiliary_loss_clip": 0.01168631,
      "auxiliary_loss_mlp": 0.01207624,
      "balance_loss_clip": 1.04769456,
      "balance_loss_mlp": 1.00066066,
      "epoch": 0.7325196897733421,
      "flos": 26907075653760.0,
      "grad_norm": 2.4818775779690205,
      "language_loss": 0.76664698,
      "learning_rate": 7.045724298891565e-07,
      "loss": 0.79040951,
      "num_input_tokens_seen": 130922935,
      "step": 6092,
      "time_per_iteration": 2.632765054702759
    },
    {
      "auxiliary_loss_clip": 0.01171552,
      "auxiliary_loss_mlp": 0.01023417,
      "balance_loss_clip": 1.04792166,
      "balance_loss_mlp": 1.01606464,
      "epoch": 0.7326399326639812,
      "flos": 25775781828480.0,
      "grad_norm": 2.588693833219397,
      "language_loss": 0.69357997,
      "learning_rate": 7.039790423199192e-07,
      "loss": 0.71552968,
      "num_input_tokens_seen": 130942575,
      "step": 6093,
      "time_per_iteration": 2.7000885009765625
    },
    {
      "auxiliary_loss_clip": 0.01179797,
      "auxiliary_loss_mlp": 0.01023776,
      "balance_loss_clip": 1.04702747,
      "balance_loss_mlp": 1.01609254,
      "epoch": 0.7327601755546204,
      "flos": 21032269706880.0,
      "grad_norm": 2.0685576729876978,
      "language_loss": 0.7782768,
      "learning_rate": 7.033858513560322e-07,
      "loss": 0.80031252,
      "num_input_tokens_seen": 130958870,
      "step": 6094,
      "time_per_iteration": 2.6260764598846436
    },
    {
      "auxiliary_loss_clip": 0.01172635,
      "auxiliary_loss_mlp": 0.0102544,
      "balance_loss_clip": 1.04838634,
      "balance_loss_mlp": 1.01793551,
      "epoch": 0.7328804184452594,
      "flos": 16289224462080.0,
      "grad_norm": 3.49464108756735,
      "language_loss": 0.76473892,
      "learning_rate": 7.027928570874794e-07,
      "loss": 0.78671962,
      "num_input_tokens_seen": 130977060,
      "step": 6095,
      "time_per_iteration": 3.606250047683716
    },
    {
      "auxiliary_loss_clip": 0.01165776,
      "auxiliary_loss_mlp": 0.01024495,
      "balance_loss_clip": 1.04597795,
      "balance_loss_mlp": 1.01670432,
      "epoch": 0.7330006613358985,
      "flos": 17858233422720.0,
      "grad_norm": 2.2677010049444344,
      "language_loss": 0.85639298,
      "learning_rate": 7.022000596042194e-07,
      "loss": 0.87829566,
      "num_input_tokens_seen": 130994160,
      "step": 6096,
      "time_per_iteration": 2.613666296005249
    },
    {
      "auxiliary_loss_clip": 0.01176092,
      "auxiliary_loss_mlp": 0.01027176,
      "balance_loss_clip": 1.03957319,
      "balance_loss_mlp": 1.01997554,
      "epoch": 0.7331209042265376,
      "flos": 22492074343680.0,
      "grad_norm": 2.2383834330409544,
      "language_loss": 0.81681257,
      "learning_rate": 7.016074589961784e-07,
      "loss": 0.83884525,
      "num_input_tokens_seen": 131012725,
      "step": 6097,
      "time_per_iteration": 3.6135177612304688
    },
    {
      "auxiliary_loss_clip": 0.01168193,
      "auxiliary_loss_mlp": 0.01024442,
      "balance_loss_clip": 1.04457307,
      "balance_loss_mlp": 1.01643157,
      "epoch": 0.7332411471171767,
      "flos": 33072757937280.0,
      "grad_norm": 2.1554535537837998,
      "language_loss": 0.67179585,
      "learning_rate": 7.01015055353253e-07,
      "loss": 0.69372219,
      "num_input_tokens_seen": 131035150,
      "step": 6098,
      "time_per_iteration": 2.7379262447357178
    },
    {
      "auxiliary_loss_clip": 0.01168644,
      "auxiliary_loss_mlp": 0.01023179,
      "balance_loss_clip": 1.04071236,
      "balance_loss_mlp": 1.01497126,
      "epoch": 0.7333613900078157,
      "flos": 22743017735040.0,
      "grad_norm": 2.092597241842634,
      "language_loss": 0.78015321,
      "learning_rate": 7.004228487653123e-07,
      "loss": 0.80207145,
      "num_input_tokens_seen": 131055955,
      "step": 6099,
      "time_per_iteration": 2.676462173461914
    },
    {
      "auxiliary_loss_clip": 0.01171769,
      "auxiliary_loss_mlp": 0.01023937,
      "balance_loss_clip": 1.03924251,
      "balance_loss_mlp": 1.01639128,
      "epoch": 0.7334816328984549,
      "flos": 22346133384960.0,
      "grad_norm": 1.76414907789699,
      "language_loss": 0.78619492,
      "learning_rate": 6.998308393221906e-07,
      "loss": 0.80815196,
      "num_input_tokens_seen": 131074360,
      "step": 6100,
      "time_per_iteration": 2.718604803085327
    },
    {
      "auxiliary_loss_clip": 0.01178082,
      "auxiliary_loss_mlp": 0.01026458,
      "balance_loss_clip": 1.04393458,
      "balance_loss_mlp": 1.01905239,
      "epoch": 0.733601875789094,
      "flos": 20736149984640.0,
      "grad_norm": 1.9345050290367536,
      "language_loss": 0.70905542,
      "learning_rate": 6.992390271136977e-07,
      "loss": 0.7311008,
      "num_input_tokens_seen": 131090070,
      "step": 6101,
      "time_per_iteration": 2.7315850257873535
    },
    {
      "auxiliary_loss_clip": 0.0116359,
      "auxiliary_loss_mlp": 0.01021819,
      "balance_loss_clip": 1.04298365,
      "balance_loss_mlp": 1.014714,
      "epoch": 0.733722118679733,
      "flos": 22564362464640.0,
      "grad_norm": 2.1440807031467797,
      "language_loss": 0.85699397,
      "learning_rate": 6.986474122296094e-07,
      "loss": 0.87884808,
      "num_input_tokens_seen": 131109185,
      "step": 6102,
      "time_per_iteration": 3.4752206802368164
    },
    {
      "auxiliary_loss_clip": 0.01171893,
      "auxiliary_loss_mlp": 0.01025132,
      "balance_loss_clip": 1.04895329,
      "balance_loss_mlp": 1.01750898,
      "epoch": 0.7338423615703722,
      "flos": 20084192179200.0,
      "grad_norm": 1.8829993548328074,
      "language_loss": 0.72065765,
      "learning_rate": 6.980559947596751e-07,
      "loss": 0.74262792,
      "num_input_tokens_seen": 131127725,
      "step": 6103,
      "time_per_iteration": 2.6335842609405518
    },
    {
      "auxiliary_loss_clip": 0.01175915,
      "auxiliary_loss_mlp": 0.01022477,
      "balance_loss_clip": 1.03982329,
      "balance_loss_mlp": 1.01493716,
      "epoch": 0.7339626044610112,
      "flos": 21687675217920.0,
      "grad_norm": 2.0032377733116,
      "language_loss": 0.75689805,
      "learning_rate": 6.974647747936109e-07,
      "loss": 0.77888203,
      "num_input_tokens_seen": 131146110,
      "step": 6104,
      "time_per_iteration": 2.660097360610962
    },
    {
      "auxiliary_loss_clip": 0.01169097,
      "auxiliary_loss_mlp": 0.01207774,
      "balance_loss_clip": 1.04821742,
      "balance_loss_mlp": 1.00068092,
      "epoch": 0.7340828473516503,
      "flos": 15268248282240.0,
      "grad_norm": 2.2906367994797447,
      "language_loss": 0.82616985,
      "learning_rate": 6.968737524211039e-07,
      "loss": 0.84993857,
      "num_input_tokens_seen": 131162920,
      "step": 6105,
      "time_per_iteration": 2.6102664470672607
    },
    {
      "auxiliary_loss_clip": 0.01169296,
      "auxiliary_loss_mlp": 0.0102332,
      "balance_loss_clip": 1.04645801,
      "balance_loss_mlp": 1.01559508,
      "epoch": 0.7342030902422895,
      "flos": 22930112701440.0,
      "grad_norm": 2.0165540156504353,
      "language_loss": 0.80378711,
      "learning_rate": 6.962829277318132e-07,
      "loss": 0.82571328,
      "num_input_tokens_seen": 131182515,
      "step": 6106,
      "time_per_iteration": 2.586183786392212
    },
    {
      "auxiliary_loss_clip": 0.01172394,
      "auxiliary_loss_mlp": 0.01025313,
      "balance_loss_clip": 1.04880583,
      "balance_loss_mlp": 1.01803517,
      "epoch": 0.7343233331329285,
      "flos": 25847890381440.0,
      "grad_norm": 2.9030971924637443,
      "language_loss": 0.83530998,
      "learning_rate": 6.956923008153652e-07,
      "loss": 0.85728705,
      "num_input_tokens_seen": 131202280,
      "step": 6107,
      "time_per_iteration": 2.6769444942474365
    },
    {
      "auxiliary_loss_clip": 0.01169976,
      "auxiliary_loss_mlp": 0.01026476,
      "balance_loss_clip": 1.0441252,
      "balance_loss_mlp": 1.01945174,
      "epoch": 0.7344435760235676,
      "flos": 18478985287680.0,
      "grad_norm": 2.486326968455777,
      "language_loss": 0.8431794,
      "learning_rate": 6.951018717613593e-07,
      "loss": 0.8651439,
      "num_input_tokens_seen": 131221295,
      "step": 6108,
      "time_per_iteration": 2.5800600051879883
    },
    {
      "auxiliary_loss_clip": 0.01168809,
      "auxiliary_loss_mlp": 0.01025511,
      "balance_loss_clip": 1.04583406,
      "balance_loss_mlp": 1.01802146,
      "epoch": 0.7345638189142067,
      "flos": 17640040256640.0,
      "grad_norm": 1.6293362603073258,
      "language_loss": 0.78059196,
      "learning_rate": 6.945116406593614e-07,
      "loss": 0.80253518,
      "num_input_tokens_seen": 131240150,
      "step": 6109,
      "time_per_iteration": 2.6625802516937256
    },
    {
      "auxiliary_loss_clip": 0.01180725,
      "auxiliary_loss_mlp": 0.01025456,
      "balance_loss_clip": 1.04304242,
      "balance_loss_mlp": 1.01775467,
      "epoch": 0.7346840618048458,
      "flos": 20260225756800.0,
      "grad_norm": 2.205785031482178,
      "language_loss": 0.74288428,
      "learning_rate": 6.939216075989089e-07,
      "loss": 0.7649461,
      "num_input_tokens_seen": 131258080,
      "step": 6110,
      "time_per_iteration": 2.6590092182159424
    },
    {
      "auxiliary_loss_clip": 0.01170508,
      "auxiliary_loss_mlp": 0.01020609,
      "balance_loss_clip": 1.04315245,
      "balance_loss_mlp": 1.01316452,
      "epoch": 0.7348043046954849,
      "flos": 29023183641600.0,
      "grad_norm": 4.071555347023908,
      "language_loss": 0.66133565,
      "learning_rate": 6.933317726695109e-07,
      "loss": 0.68324691,
      "num_input_tokens_seen": 131279310,
      "step": 6111,
      "time_per_iteration": 2.7053678035736084
    },
    {
      "auxiliary_loss_clip": 0.01171145,
      "auxiliary_loss_mlp": 0.01022903,
      "balance_loss_clip": 1.04376364,
      "balance_loss_mlp": 1.01533866,
      "epoch": 0.734924547586124,
      "flos": 17931203902080.0,
      "grad_norm": 4.009826078565068,
      "language_loss": 0.80068552,
      "learning_rate": 6.92742135960644e-07,
      "loss": 0.82262594,
      "num_input_tokens_seen": 131297010,
      "step": 6112,
      "time_per_iteration": 2.653128147125244
    },
    {
      "auxiliary_loss_clip": 0.0107432,
      "auxiliary_loss_mlp": 0.01001739,
      "balance_loss_clip": 1.01330638,
      "balance_loss_mlp": 1.0008564,
      "epoch": 0.7350447904767631,
      "flos": 63588319850880.0,
      "grad_norm": 0.8230896548921521,
      "language_loss": 0.55623519,
      "learning_rate": 6.921526975617556e-07,
      "loss": 0.57699579,
      "num_input_tokens_seen": 131356470,
      "step": 6113,
      "time_per_iteration": 3.193995475769043
    },
    {
      "auxiliary_loss_clip": 0.01174764,
      "auxiliary_loss_mlp": 0.01025165,
      "balance_loss_clip": 1.04280818,
      "balance_loss_mlp": 1.01714802,
      "epoch": 0.7351650333674021,
      "flos": 21580015178880.0,
      "grad_norm": 1.8955984090625646,
      "language_loss": 0.7561546,
      "learning_rate": 6.915634575622631e-07,
      "loss": 0.7781539,
      "num_input_tokens_seen": 131374985,
      "step": 6114,
      "time_per_iteration": 2.671565532684326
    },
    {
      "auxiliary_loss_clip": 0.01166034,
      "auxiliary_loss_mlp": 0.01019355,
      "balance_loss_clip": 1.04726064,
      "balance_loss_mlp": 1.0119822,
      "epoch": 0.7352852762580413,
      "flos": 18186349184640.0,
      "grad_norm": 2.607667297615874,
      "language_loss": 0.70477599,
      "learning_rate": 6.909744160515532e-07,
      "loss": 0.72662991,
      "num_input_tokens_seen": 131393125,
      "step": 6115,
      "time_per_iteration": 3.4952971935272217
    },
    {
      "auxiliary_loss_clip": 0.0117204,
      "auxiliary_loss_mlp": 0.0102532,
      "balance_loss_clip": 1.04391503,
      "balance_loss_mlp": 1.01785433,
      "epoch": 0.7354055191486804,
      "flos": 38910073063680.0,
      "grad_norm": 1.8374699236986383,
      "language_loss": 0.69356072,
      "learning_rate": 6.903855731189849e-07,
      "loss": 0.71553433,
      "num_input_tokens_seen": 131415760,
      "step": 6116,
      "time_per_iteration": 2.7993087768554688
    },
    {
      "auxiliary_loss_clip": 0.01180086,
      "auxiliary_loss_mlp": 0.0102859,
      "balance_loss_clip": 1.04588675,
      "balance_loss_mlp": 1.02026343,
      "epoch": 0.7355257620393194,
      "flos": 16289978647680.0,
      "grad_norm": 11.80972999855257,
      "language_loss": 0.82191503,
      "learning_rate": 6.897969288538825e-07,
      "loss": 0.84400189,
      "num_input_tokens_seen": 131433705,
      "step": 6117,
      "time_per_iteration": 2.601910352706909
    },
    {
      "auxiliary_loss_clip": 0.01171735,
      "auxiliary_loss_mlp": 0.01026128,
      "balance_loss_clip": 1.04458046,
      "balance_loss_mlp": 1.01883805,
      "epoch": 0.7356460049299585,
      "flos": 18114240631680.0,
      "grad_norm": 1.9084403924917828,
      "language_loss": 0.81513709,
      "learning_rate": 6.892084833455452e-07,
      "loss": 0.83711571,
      "num_input_tokens_seen": 131453275,
      "step": 6118,
      "time_per_iteration": 2.6465413570404053
    },
    {
      "auxiliary_loss_clip": 0.01168802,
      "auxiliary_loss_mlp": 0.01022357,
      "balance_loss_clip": 1.04760349,
      "balance_loss_mlp": 1.01547909,
      "epoch": 0.7357662478205976,
      "flos": 21325193118720.0,
      "grad_norm": 1.4615937179287353,
      "language_loss": 0.83595967,
      "learning_rate": 6.886202366832384e-07,
      "loss": 0.85787123,
      "num_input_tokens_seen": 131474960,
      "step": 6119,
      "time_per_iteration": 2.699786424636841
    },
    {
      "auxiliary_loss_clip": 0.01174507,
      "auxiliary_loss_mlp": 0.01021419,
      "balance_loss_clip": 1.04193962,
      "balance_loss_mlp": 1.01358104,
      "epoch": 0.7358864907112367,
      "flos": 14246841139200.0,
      "grad_norm": 2.157757368247733,
      "language_loss": 0.73361814,
      "learning_rate": 6.880321889561987e-07,
      "loss": 0.75557745,
      "num_input_tokens_seen": 131492935,
      "step": 6120,
      "time_per_iteration": 2.617363214492798
    },
    {
      "auxiliary_loss_clip": 0.01170737,
      "auxiliary_loss_mlp": 0.01028098,
      "balance_loss_clip": 1.04304552,
      "balance_loss_mlp": 1.01967597,
      "epoch": 0.7360067336018757,
      "flos": 22309684058880.0,
      "grad_norm": 2.1366431104411734,
      "language_loss": 0.65356755,
      "learning_rate": 6.874443402536338e-07,
      "loss": 0.67555594,
      "num_input_tokens_seen": 131512025,
      "step": 6121,
      "time_per_iteration": 2.77795672416687
    },
    {
      "auxiliary_loss_clip": 0.01173464,
      "auxiliary_loss_mlp": 0.01020837,
      "balance_loss_clip": 1.04435015,
      "balance_loss_mlp": 1.0132432,
      "epoch": 0.7361269764925149,
      "flos": 25554607833600.0,
      "grad_norm": 1.8467202335218675,
      "language_loss": 0.80339962,
      "learning_rate": 6.868566906647177e-07,
      "loss": 0.82534266,
      "num_input_tokens_seen": 131532975,
      "step": 6122,
      "time_per_iteration": 3.73568058013916
    },
    {
      "auxiliary_loss_clip": 0.01170225,
      "auxiliary_loss_mlp": 0.01029851,
      "balance_loss_clip": 1.04503918,
      "balance_loss_mlp": 1.02179837,
      "epoch": 0.736247219383154,
      "flos": 20376505059840.0,
      "grad_norm": 5.220297743325479,
      "language_loss": 0.83727461,
      "learning_rate": 6.862692402785984e-07,
      "loss": 0.8592754,
      "num_input_tokens_seen": 131553225,
      "step": 6123,
      "time_per_iteration": 2.771449089050293
    },
    {
      "auxiliary_loss_clip": 0.0107985,
      "auxiliary_loss_mlp": 0.00998615,
      "balance_loss_clip": 1.01406431,
      "balance_loss_mlp": 0.99778032,
      "epoch": 0.736367462273793,
      "flos": 70339525735680.0,
      "grad_norm": 0.6813960098725127,
      "language_loss": 0.49544957,
      "learning_rate": 6.856819891843899e-07,
      "loss": 0.51623422,
      "num_input_tokens_seen": 131617930,
      "step": 6124,
      "time_per_iteration": 4.154621124267578
    },
    {
      "auxiliary_loss_clip": 0.01180758,
      "auxiliary_loss_mlp": 0.01028522,
      "balance_loss_clip": 1.0403775,
      "balance_loss_mlp": 1.02074909,
      "epoch": 0.7364877051644322,
      "flos": 22412711243520.0,
      "grad_norm": 5.228764574751437,
      "language_loss": 0.72272086,
      "learning_rate": 6.8509493747118e-07,
      "loss": 0.74481362,
      "num_input_tokens_seen": 131636740,
      "step": 6125,
      "time_per_iteration": 2.6987462043762207
    },
    {
      "auxiliary_loss_clip": 0.01167698,
      "auxiliary_loss_mlp": 0.01024787,
      "balance_loss_clip": 1.04772437,
      "balance_loss_mlp": 1.01723528,
      "epoch": 0.7366079480550712,
      "flos": 12130266274560.0,
      "grad_norm": 1.9680311717077656,
      "language_loss": 0.88555861,
      "learning_rate": 6.845080852280221e-07,
      "loss": 0.90748346,
      "num_input_tokens_seen": 131653810,
      "step": 6126,
      "time_per_iteration": 2.597024917602539
    },
    {
      "auxiliary_loss_clip": 0.01175959,
      "auxiliary_loss_mlp": 0.0102174,
      "balance_loss_clip": 1.04249549,
      "balance_loss_mlp": 1.01487994,
      "epoch": 0.7367281909457103,
      "flos": 15049336844160.0,
      "grad_norm": 1.6948484992825472,
      "language_loss": 0.74480212,
      "learning_rate": 6.839214325439409e-07,
      "loss": 0.76677912,
      "num_input_tokens_seen": 131671505,
      "step": 6127,
      "time_per_iteration": 2.6082751750946045
    },
    {
      "auxiliary_loss_clip": 0.01164158,
      "auxiliary_loss_mlp": 0.01022001,
      "balance_loss_clip": 1.04328489,
      "balance_loss_mlp": 1.01482725,
      "epoch": 0.7368484338363495,
      "flos": 23510752053120.0,
      "grad_norm": 1.6536899235458284,
      "language_loss": 0.71843827,
      "learning_rate": 6.833349795079327e-07,
      "loss": 0.74029982,
      "num_input_tokens_seen": 131690615,
      "step": 6128,
      "time_per_iteration": 2.6920909881591797
    },
    {
      "auxiliary_loss_clip": 0.01176562,
      "auxiliary_loss_mlp": 0.01027976,
      "balance_loss_clip": 1.04545975,
      "balance_loss_mlp": 1.02067709,
      "epoch": 0.7369686767269885,
      "flos": 27417833095680.0,
      "grad_norm": 2.495302831042931,
      "language_loss": 0.6868459,
      "learning_rate": 6.827487262089613e-07,
      "loss": 0.70889127,
      "num_input_tokens_seen": 131711120,
      "step": 6129,
      "time_per_iteration": 3.5567915439605713
    },
    {
      "auxiliary_loss_clip": 0.01077333,
      "auxiliary_loss_mlp": 0.01001803,
      "balance_loss_clip": 1.0136261,
      "balance_loss_mlp": 1.00101614,
      "epoch": 0.7370889196176276,
      "flos": 70293343824000.0,
      "grad_norm": 1.017122866256442,
      "language_loss": 0.56879365,
      "learning_rate": 6.821626727359606e-07,
      "loss": 0.58958507,
      "num_input_tokens_seen": 131776680,
      "step": 6130,
      "time_per_iteration": 3.2234716415405273
    },
    {
      "auxiliary_loss_clip": 0.01174634,
      "auxiliary_loss_mlp": 0.01026723,
      "balance_loss_clip": 1.04820442,
      "balance_loss_mlp": 1.018682,
      "epoch": 0.7372091625082667,
      "flos": 18040839189120.0,
      "grad_norm": 2.33338354493766,
      "language_loss": 0.7750864,
      "learning_rate": 6.815768191778348e-07,
      "loss": 0.79709995,
      "num_input_tokens_seen": 131794760,
      "step": 6131,
      "time_per_iteration": 2.6079165935516357
    },
    {
      "auxiliary_loss_clip": 0.01166891,
      "auxiliary_loss_mlp": 0.01026925,
      "balance_loss_clip": 1.04476905,
      "balance_loss_mlp": 1.01902771,
      "epoch": 0.7373294053989058,
      "flos": 33726331854720.0,
      "grad_norm": 1.8501607990625202,
      "language_loss": 0.7303704,
      "learning_rate": 6.809911656234569e-07,
      "loss": 0.75230861,
      "num_input_tokens_seen": 131816735,
      "step": 6132,
      "time_per_iteration": 2.6753034591674805
    },
    {
      "auxiliary_loss_clip": 0.01177006,
      "auxiliary_loss_mlp": 0.01022017,
      "balance_loss_clip": 1.04149199,
      "balance_loss_mlp": 1.01467085,
      "epoch": 0.7374496482895448,
      "flos": 21506326427520.0,
      "grad_norm": 3.4740006421269207,
      "language_loss": 0.78620064,
      "learning_rate": 6.804057121616707e-07,
      "loss": 0.80819082,
      "num_input_tokens_seen": 131834940,
      "step": 6133,
      "time_per_iteration": 2.700634717941284
    },
    {
      "auxiliary_loss_clip": 0.01170637,
      "auxiliary_loss_mlp": 0.01025274,
      "balance_loss_clip": 1.04588628,
      "balance_loss_mlp": 1.01711392,
      "epoch": 0.737569891180184,
      "flos": 24936908624640.0,
      "grad_norm": 1.8308439263522636,
      "language_loss": 0.71857369,
      "learning_rate": 6.798204588812888e-07,
      "loss": 0.74053282,
      "num_input_tokens_seen": 131854355,
      "step": 6134,
      "time_per_iteration": 2.6550774574279785
    },
    {
      "auxiliary_loss_clip": 0.01170289,
      "auxiliary_loss_mlp": 0.01207631,
      "balance_loss_clip": 1.03769863,
      "balance_loss_mlp": 1.00064015,
      "epoch": 0.7376901340708231,
      "flos": 20664544222080.0,
      "grad_norm": 2.2862832972730285,
      "language_loss": 0.75365466,
      "learning_rate": 6.792354058710937e-07,
      "loss": 0.77743387,
      "num_input_tokens_seen": 131871825,
      "step": 6135,
      "time_per_iteration": 2.724475622177124
    },
    {
      "auxiliary_loss_clip": 0.01159851,
      "auxiliary_loss_mlp": 0.01020783,
      "balance_loss_clip": 1.0447042,
      "balance_loss_mlp": 1.01314807,
      "epoch": 0.7378103769614621,
      "flos": 23805794367360.0,
      "grad_norm": 1.8359873010163146,
      "language_loss": 0.64888036,
      "learning_rate": 6.786505532198374e-07,
      "loss": 0.67068672,
      "num_input_tokens_seen": 131890770,
      "step": 6136,
      "time_per_iteration": 2.6084980964660645
    },
    {
      "auxiliary_loss_clip": 0.01168279,
      "auxiliary_loss_mlp": 0.0102352,
      "balance_loss_clip": 1.04747272,
      "balance_loss_mlp": 1.01554465,
      "epoch": 0.7379306198521013,
      "flos": 22237216369920.0,
      "grad_norm": 1.8781228058790669,
      "language_loss": 0.85288668,
      "learning_rate": 6.780659010162411e-07,
      "loss": 0.87480462,
      "num_input_tokens_seen": 131909720,
      "step": 6137,
      "time_per_iteration": 2.634641408920288
    },
    {
      "auxiliary_loss_clip": 0.01180156,
      "auxiliary_loss_mlp": 0.01022836,
      "balance_loss_clip": 1.04434562,
      "balance_loss_mlp": 1.01559138,
      "epoch": 0.7380508627427403,
      "flos": 14903108576640.0,
      "grad_norm": 2.314852792730772,
      "language_loss": 0.83108091,
      "learning_rate": 6.774814493489975e-07,
      "loss": 0.85311079,
      "num_input_tokens_seen": 131927395,
      "step": 6138,
      "time_per_iteration": 2.6506781578063965
    },
    {
      "auxiliary_loss_clip": 0.01166206,
      "auxiliary_loss_mlp": 0.01025725,
      "balance_loss_clip": 1.0452466,
      "balance_loss_mlp": 1.01858759,
      "epoch": 0.7381711056333794,
      "flos": 21685843624320.0,
      "grad_norm": 1.7819985288752482,
      "language_loss": 0.65977025,
      "learning_rate": 6.768971983067655e-07,
      "loss": 0.68168956,
      "num_input_tokens_seen": 131947725,
      "step": 6139,
      "time_per_iteration": 2.72049880027771
    },
    {
      "auxiliary_loss_clip": 0.01066519,
      "auxiliary_loss_mlp": 0.01000817,
      "balance_loss_clip": 1.01271987,
      "balance_loss_mlp": 0.99991721,
      "epoch": 0.7382913485240186,
      "flos": 52404263596800.0,
      "grad_norm": 1.0027263683791974,
      "language_loss": 0.6772024,
      "learning_rate": 6.763131479781772e-07,
      "loss": 0.69787574,
      "num_input_tokens_seen": 131997485,
      "step": 6140,
      "time_per_iteration": 2.968113422393799
    },
    {
      "auxiliary_loss_clip": 0.01161123,
      "auxiliary_loss_mlp": 0.01021834,
      "balance_loss_clip": 1.04336107,
      "balance_loss_mlp": 1.01458049,
      "epoch": 0.7384115914146576,
      "flos": 21798818876160.0,
      "grad_norm": 3.0942468619821586,
      "language_loss": 0.7630446,
      "learning_rate": 6.757292984518316e-07,
      "loss": 0.7848742,
      "num_input_tokens_seen": 132016885,
      "step": 6141,
      "time_per_iteration": 2.604030132293701
    },
    {
      "auxiliary_loss_clip": 0.01073954,
      "auxiliary_loss_mlp": 0.01001422,
      "balance_loss_clip": 1.01317203,
      "balance_loss_mlp": 1.00055742,
      "epoch": 0.7385318343052967,
      "flos": 61494331662720.0,
      "grad_norm": 0.7395741485321272,
      "language_loss": 0.5638026,
      "learning_rate": 6.751456498162981e-07,
      "loss": 0.58455634,
      "num_input_tokens_seen": 132075920,
      "step": 6142,
      "time_per_iteration": 4.033464431762695
    },
    {
      "auxiliary_loss_clip": 0.01167274,
      "auxiliary_loss_mlp": 0.01021056,
      "balance_loss_clip": 1.04213369,
      "balance_loss_mlp": 1.01451731,
      "epoch": 0.7386520771959358,
      "flos": 17013757697280.0,
      "grad_norm": 2.0188827952082327,
      "language_loss": 0.85827053,
      "learning_rate": 6.745622021601174e-07,
      "loss": 0.88015389,
      "num_input_tokens_seen": 132092945,
      "step": 6143,
      "time_per_iteration": 2.5580644607543945
    },
    {
      "auxiliary_loss_clip": 0.0117752,
      "auxiliary_loss_mlp": 0.01020683,
      "balance_loss_clip": 1.04249811,
      "balance_loss_mlp": 1.01336098,
      "epoch": 0.7387723200865749,
      "flos": 18770759464320.0,
      "grad_norm": 2.0566425986502286,
      "language_loss": 0.69985718,
      "learning_rate": 6.739789555717954e-07,
      "loss": 0.72183919,
      "num_input_tokens_seen": 132109920,
      "step": 6144,
      "time_per_iteration": 2.6714210510253906
    },
    {
      "auxiliary_loss_clip": 0.01166026,
      "auxiliary_loss_mlp": 0.01024848,
      "balance_loss_clip": 1.04712498,
      "balance_loss_mlp": 1.01740336,
      "epoch": 0.738892562977214,
      "flos": 22525542840960.0,
      "grad_norm": 2.102680577773653,
      "language_loss": 0.77540457,
      "learning_rate": 6.733959101398124e-07,
      "loss": 0.79731333,
      "num_input_tokens_seen": 132128050,
      "step": 6145,
      "time_per_iteration": 2.6264684200286865
    },
    {
      "auxiliary_loss_clip": 0.01169628,
      "auxiliary_loss_mlp": 0.01024062,
      "balance_loss_clip": 1.04265761,
      "balance_loss_mlp": 1.01616478,
      "epoch": 0.7390128058678531,
      "flos": 21501478091520.0,
      "grad_norm": 1.85905998759783,
      "language_loss": 0.81742561,
      "learning_rate": 6.728130659526143e-07,
      "loss": 0.8393625,
      "num_input_tokens_seen": 132145860,
      "step": 6146,
      "time_per_iteration": 2.6300315856933594
    },
    {
      "auxiliary_loss_clip": 0.01173579,
      "auxiliary_loss_mlp": 0.01028279,
      "balance_loss_clip": 1.04560995,
      "balance_loss_mlp": 1.0207901,
      "epoch": 0.7391330487584922,
      "flos": 25776176878080.0,
      "grad_norm": 2.4853043020881604,
      "language_loss": 0.7132206,
      "learning_rate": 6.7223042309862e-07,
      "loss": 0.73523921,
      "num_input_tokens_seen": 132166060,
      "step": 6147,
      "time_per_iteration": 2.7299280166625977
    },
    {
      "auxiliary_loss_clip": 0.01165564,
      "auxiliary_loss_mlp": 0.01026357,
      "balance_loss_clip": 1.04340184,
      "balance_loss_mlp": 1.01953793,
      "epoch": 0.7392532916491312,
      "flos": 28366736636160.0,
      "grad_norm": 1.9490869418972652,
      "language_loss": 0.73870325,
      "learning_rate": 6.716479816662144e-07,
      "loss": 0.76062238,
      "num_input_tokens_seen": 132187790,
      "step": 6148,
      "time_per_iteration": 3.567202568054199
    },
    {
      "auxiliary_loss_clip": 0.01175794,
      "auxiliary_loss_mlp": 0.01023976,
      "balance_loss_clip": 1.04394412,
      "balance_loss_mlp": 1.01665926,
      "epoch": 0.7393735345397703,
      "flos": 23585877348480.0,
      "grad_norm": 24.041356242592883,
      "language_loss": 0.73353124,
      "learning_rate": 6.710657417437531e-07,
      "loss": 0.75552893,
      "num_input_tokens_seen": 132207495,
      "step": 6149,
      "time_per_iteration": 2.699763536453247
    },
    {
      "auxiliary_loss_clip": 0.01169045,
      "auxiliary_loss_mlp": 0.01023824,
      "balance_loss_clip": 1.04303753,
      "balance_loss_mlp": 1.01676106,
      "epoch": 0.7394937774304094,
      "flos": 19974772373760.0,
      "grad_norm": 2.111368693097346,
      "language_loss": 0.80002326,
      "learning_rate": 6.704837034195628e-07,
      "loss": 0.82195199,
      "num_input_tokens_seen": 132225960,
      "step": 6150,
      "time_per_iteration": 3.6044445037841797
    },
    {
      "auxiliary_loss_clip": 0.01162724,
      "auxiliary_loss_mlp": 0.01029227,
      "balance_loss_clip": 1.0445739,
      "balance_loss_mlp": 1.02168119,
      "epoch": 0.7396140203210485,
      "flos": 23478037741440.0,
      "grad_norm": 2.044576966775737,
      "language_loss": 0.85102546,
      "learning_rate": 6.699018667819376e-07,
      "loss": 0.87294495,
      "num_input_tokens_seen": 132245360,
      "step": 6151,
      "time_per_iteration": 2.651968240737915
    },
    {
      "auxiliary_loss_clip": 0.01165947,
      "auxiliary_loss_mlp": 0.01027228,
      "balance_loss_clip": 1.0431571,
      "balance_loss_mlp": 1.01940751,
      "epoch": 0.7397342632116876,
      "flos": 25555433846400.0,
      "grad_norm": 2.663375281546125,
      "language_loss": 0.72767437,
      "learning_rate": 6.693202319191415e-07,
      "loss": 0.74960607,
      "num_input_tokens_seen": 132267095,
      "step": 6152,
      "time_per_iteration": 2.6411333084106445
    },
    {
      "auxiliary_loss_clip": 0.01166084,
      "auxiliary_loss_mlp": 0.0102623,
      "balance_loss_clip": 1.04955673,
      "balance_loss_mlp": 1.01861238,
      "epoch": 0.7398545061023267,
      "flos": 24755021130240.0,
      "grad_norm": 5.092503663442073,
      "language_loss": 0.74696052,
      "learning_rate": 6.687387989194084e-07,
      "loss": 0.76888371,
      "num_input_tokens_seen": 132286610,
      "step": 6153,
      "time_per_iteration": 2.5680480003356934
    },
    {
      "auxiliary_loss_clip": 0.01166594,
      "auxiliary_loss_mlp": 0.01023823,
      "balance_loss_clip": 1.04530728,
      "balance_loss_mlp": 1.01646149,
      "epoch": 0.7399747489929658,
      "flos": 16508602776960.0,
      "grad_norm": 1.8978392654598828,
      "language_loss": 0.79386914,
      "learning_rate": 6.681575678709404e-07,
      "loss": 0.81577325,
      "num_input_tokens_seen": 132305300,
      "step": 6154,
      "time_per_iteration": 2.662022352218628
    },
    {
      "auxiliary_loss_clip": 0.01169454,
      "auxiliary_loss_mlp": 0.01024156,
      "balance_loss_clip": 1.04642534,
      "balance_loss_mlp": 1.01686954,
      "epoch": 0.7400949918836048,
      "flos": 24097065753600.0,
      "grad_norm": 2.4640978821524366,
      "language_loss": 0.70867062,
      "learning_rate": 6.67576538861911e-07,
      "loss": 0.73060679,
      "num_input_tokens_seen": 132323875,
      "step": 6155,
      "time_per_iteration": 2.605081558227539
    },
    {
      "auxiliary_loss_clip": 0.01166767,
      "auxiliary_loss_mlp": 0.01023185,
      "balance_loss_clip": 1.0428468,
      "balance_loss_mlp": 1.01617599,
      "epoch": 0.740215234774244,
      "flos": 21802517976960.0,
      "grad_norm": 1.5954047429135811,
      "language_loss": 0.81909895,
      "learning_rate": 6.669957119804612e-07,
      "loss": 0.84099847,
      "num_input_tokens_seen": 132345510,
      "step": 6156,
      "time_per_iteration": 3.7093281745910645
    },
    {
      "auxiliary_loss_clip": 0.01178832,
      "auxiliary_loss_mlp": 0.01023339,
      "balance_loss_clip": 1.045048,
      "balance_loss_mlp": 1.01637697,
      "epoch": 0.7403354776648831,
      "flos": 18733196816640.0,
      "grad_norm": 4.196088615368916,
      "language_loss": 0.72427583,
      "learning_rate": 6.66415087314702e-07,
      "loss": 0.7462976,
      "num_input_tokens_seen": 132360465,
      "step": 6157,
      "time_per_iteration": 2.6091060638427734
    },
    {
      "auxiliary_loss_clip": 0.01172521,
      "auxiliary_loss_mlp": 0.01018941,
      "balance_loss_clip": 1.04319811,
      "balance_loss_mlp": 1.01177657,
      "epoch": 0.7404557205555221,
      "flos": 16909581277440.0,
      "grad_norm": 2.3305240339283464,
      "language_loss": 0.73173404,
      "learning_rate": 6.65834664952714e-07,
      "loss": 0.75364864,
      "num_input_tokens_seen": 132377915,
      "step": 6158,
      "time_per_iteration": 2.5675947666168213
    },
    {
      "auxiliary_loss_clip": 0.01173713,
      "auxiliary_loss_mlp": 0.01021244,
      "balance_loss_clip": 1.04135704,
      "balance_loss_mlp": 1.01430035,
      "epoch": 0.7405759634461613,
      "flos": 21214408596480.0,
      "grad_norm": 1.975238018428664,
      "language_loss": 0.76057392,
      "learning_rate": 6.652544449825457e-07,
      "loss": 0.78252357,
      "num_input_tokens_seen": 132398170,
      "step": 6159,
      "time_per_iteration": 2.6725330352783203
    },
    {
      "auxiliary_loss_clip": 0.0117891,
      "auxiliary_loss_mlp": 0.01027414,
      "balance_loss_clip": 1.04321384,
      "balance_loss_mlp": 1.01997495,
      "epoch": 0.7406962063368003,
      "flos": 20480106862080.0,
      "grad_norm": 1.837543077118546,
      "language_loss": 0.7705825,
      "learning_rate": 6.646744274922182e-07,
      "loss": 0.79264569,
      "num_input_tokens_seen": 132416615,
      "step": 6160,
      "time_per_iteration": 2.605009078979492
    },
    {
      "auxiliary_loss_clip": 0.01172607,
      "auxiliary_loss_mlp": 0.01021558,
      "balance_loss_clip": 1.04298401,
      "balance_loss_mlp": 1.01397061,
      "epoch": 0.7408164492274394,
      "flos": 19791915212160.0,
      "grad_norm": 4.748000667152247,
      "language_loss": 0.75588185,
      "learning_rate": 6.640946125697171e-07,
      "loss": 0.77782345,
      "num_input_tokens_seen": 132434145,
      "step": 6161,
      "time_per_iteration": 2.6896309852600098
    },
    {
      "auxiliary_loss_clip": 0.01168672,
      "auxiliary_loss_mlp": 0.01020989,
      "balance_loss_clip": 1.04354048,
      "balance_loss_mlp": 1.01325846,
      "epoch": 0.7409366921180786,
      "flos": 29204855654400.0,
      "grad_norm": 2.4171665593821916,
      "language_loss": 0.75918591,
      "learning_rate": 6.635150003030017e-07,
      "loss": 0.78108251,
      "num_input_tokens_seen": 132452670,
      "step": 6162,
      "time_per_iteration": 2.6799216270446777
    },
    {
      "auxiliary_loss_clip": 0.01176411,
      "auxiliary_loss_mlp": 0.01022681,
      "balance_loss_clip": 1.03851223,
      "balance_loss_mlp": 1.01542139,
      "epoch": 0.7410569350087176,
      "flos": 22930004960640.0,
      "grad_norm": 2.6650587001838058,
      "language_loss": 0.85876167,
      "learning_rate": 6.629355907799981e-07,
      "loss": 0.88075268,
      "num_input_tokens_seen": 132472475,
      "step": 6163,
      "time_per_iteration": 2.7069647312164307
    },
    {
      "auxiliary_loss_clip": 0.0116865,
      "auxiliary_loss_mlp": 0.01023773,
      "balance_loss_clip": 1.04314375,
      "balance_loss_mlp": 1.01636767,
      "epoch": 0.7411771778993567,
      "flos": 30440397726720.0,
      "grad_norm": 1.7460732538747041,
      "language_loss": 0.69454336,
      "learning_rate": 6.623563840886015e-07,
      "loss": 0.71646756,
      "num_input_tokens_seen": 132493400,
      "step": 6164,
      "time_per_iteration": 2.674400806427002
    },
    {
      "auxiliary_loss_clip": 0.01163379,
      "auxiliary_loss_mlp": 0.01020984,
      "balance_loss_clip": 1.04277563,
      "balance_loss_mlp": 1.01368856,
      "epoch": 0.7412974207899958,
      "flos": 20522050968960.0,
      "grad_norm": 1.9408021914048819,
      "language_loss": 0.69591141,
      "learning_rate": 6.617773803166795e-07,
      "loss": 0.71775508,
      "num_input_tokens_seen": 132511725,
      "step": 6165,
      "time_per_iteration": 2.6427910327911377
    },
    {
      "auxiliary_loss_clip": 0.01178409,
      "auxiliary_loss_mlp": 0.01207912,
      "balance_loss_clip": 1.04588509,
      "balance_loss_mlp": 1.00073421,
      "epoch": 0.7414176636806349,
      "flos": 22090700793600.0,
      "grad_norm": 2.2307273706634003,
      "language_loss": 0.81807745,
      "learning_rate": 6.611985795520634e-07,
      "loss": 0.84194064,
      "num_input_tokens_seen": 132530270,
      "step": 6166,
      "time_per_iteration": 2.628579616546631
    },
    {
      "auxiliary_loss_clip": 0.0118453,
      "auxiliary_loss_mlp": 0.01024856,
      "balance_loss_clip": 1.04526854,
      "balance_loss_mlp": 1.01712489,
      "epoch": 0.7415379065712739,
      "flos": 25155245445120.0,
      "grad_norm": 2.2303121200900304,
      "language_loss": 0.77743983,
      "learning_rate": 6.606199818825588e-07,
      "loss": 0.79953372,
      "num_input_tokens_seen": 132550725,
      "step": 6167,
      "time_per_iteration": 2.715894937515259
    },
    {
      "auxiliary_loss_clip": 0.01175193,
      "auxiliary_loss_mlp": 0.0102189,
      "balance_loss_clip": 1.0421977,
      "balance_loss_mlp": 1.01489246,
      "epoch": 0.7416581494619131,
      "flos": 16871731320960.0,
      "grad_norm": 2.127912716381052,
      "language_loss": 0.81581855,
      "learning_rate": 6.600415873959377e-07,
      "loss": 0.83778942,
      "num_input_tokens_seen": 132568600,
      "step": 6168,
      "time_per_iteration": 2.6331076622009277
    },
    {
      "auxiliary_loss_clip": 0.01176854,
      "auxiliary_loss_mlp": 0.01207228,
      "balance_loss_clip": 1.03784204,
      "balance_loss_mlp": 1.00081325,
      "epoch": 0.7417783923525522,
      "flos": 28438881102720.0,
      "grad_norm": 2.4363301228118037,
      "language_loss": 0.64990765,
      "learning_rate": 6.594633961799437e-07,
      "loss": 0.67374843,
      "num_input_tokens_seen": 132587640,
      "step": 6169,
      "time_per_iteration": 3.751650333404541
    },
    {
      "auxiliary_loss_clip": 0.01182462,
      "auxiliary_loss_mlp": 0.01022212,
      "balance_loss_clip": 1.04354692,
      "balance_loss_mlp": 1.01509547,
      "epoch": 0.7418986352431912,
      "flos": 20084299920000.0,
      "grad_norm": 1.6896863061892768,
      "language_loss": 0.81626332,
      "learning_rate": 6.588854083222857e-07,
      "loss": 0.83831006,
      "num_input_tokens_seen": 132607075,
      "step": 6170,
      "time_per_iteration": 2.7109103202819824
    },
    {
      "auxiliary_loss_clip": 0.01175369,
      "auxiliary_loss_mlp": 0.01027659,
      "balance_loss_clip": 1.04549694,
      "balance_loss_mlp": 1.01928496,
      "epoch": 0.7420188781338304,
      "flos": 18259571059200.0,
      "grad_norm": 2.938333363124948,
      "language_loss": 0.80526936,
      "learning_rate": 6.583076239106444e-07,
      "loss": 0.82729965,
      "num_input_tokens_seen": 132625580,
      "step": 6171,
      "time_per_iteration": 2.599522352218628
    },
    {
      "auxiliary_loss_clip": 0.01174993,
      "auxiliary_loss_mlp": 0.01021787,
      "balance_loss_clip": 1.04367781,
      "balance_loss_mlp": 1.0141933,
      "epoch": 0.7421391210244694,
      "flos": 13771994319360.0,
      "grad_norm": 3.105985496725403,
      "language_loss": 0.75717497,
      "learning_rate": 6.577300430326707e-07,
      "loss": 0.77914274,
      "num_input_tokens_seen": 132640525,
      "step": 6172,
      "time_per_iteration": 2.657707691192627
    },
    {
      "auxiliary_loss_clip": 0.01170503,
      "auxiliary_loss_mlp": 0.01020128,
      "balance_loss_clip": 1.04277682,
      "balance_loss_mlp": 1.01310325,
      "epoch": 0.7422593639151085,
      "flos": 15961683317760.0,
      "grad_norm": 2.2589656256614705,
      "language_loss": 0.71875972,
      "learning_rate": 6.571526657759821e-07,
      "loss": 0.74066603,
      "num_input_tokens_seen": 132656265,
      "step": 6173,
      "time_per_iteration": 2.6515188217163086
    },
    {
      "auxiliary_loss_clip": 0.01163057,
      "auxiliary_loss_mlp": 0.01021132,
      "balance_loss_clip": 1.04324532,
      "balance_loss_mlp": 1.01378918,
      "epoch": 0.7423796068057477,
      "flos": 30114400867200.0,
      "grad_norm": 1.7173961281026962,
      "language_loss": 0.70545673,
      "learning_rate": 6.565754922281663e-07,
      "loss": 0.72729862,
      "num_input_tokens_seen": 132678510,
      "step": 6174,
      "time_per_iteration": 2.7072696685791016
    },
    {
      "auxiliary_loss_clip": 0.01169653,
      "auxiliary_loss_mlp": 0.01026542,
      "balance_loss_clip": 1.04264486,
      "balance_loss_mlp": 1.01911235,
      "epoch": 0.7424998496963867,
      "flos": 20521907314560.0,
      "grad_norm": 1.8444992306606212,
      "language_loss": 0.78328723,
      "learning_rate": 6.559985224767801e-07,
      "loss": 0.80524921,
      "num_input_tokens_seen": 132696385,
      "step": 6175,
      "time_per_iteration": 3.5721962451934814
    },
    {
      "auxiliary_loss_clip": 0.01179854,
      "auxiliary_loss_mlp": 0.01028714,
      "balance_loss_clip": 1.04271305,
      "balance_loss_mlp": 1.02129889,
      "epoch": 0.7426200925870258,
      "flos": 21871573873920.0,
      "grad_norm": 3.202020389266677,
      "language_loss": 0.75199926,
      "learning_rate": 6.55421756609349e-07,
      "loss": 0.77408487,
      "num_input_tokens_seen": 132714640,
      "step": 6176,
      "time_per_iteration": 2.648716926574707
    },
    {
      "auxiliary_loss_clip": 0.01165119,
      "auxiliary_loss_mlp": 0.01028297,
      "balance_loss_clip": 1.04664373,
      "balance_loss_mlp": 1.02030408,
      "epoch": 0.7427403354776649,
      "flos": 26432049265920.0,
      "grad_norm": 2.3842238367583968,
      "language_loss": 0.7890175,
      "learning_rate": 6.54845194713369e-07,
      "loss": 0.81095159,
      "num_input_tokens_seen": 132735590,
      "step": 6177,
      "time_per_iteration": 3.6484413146972656
    },
    {
      "auxiliary_loss_clip": 0.01164339,
      "auxiliary_loss_mlp": 0.01029215,
      "balance_loss_clip": 1.04464865,
      "balance_loss_mlp": 1.02231884,
      "epoch": 0.742860578368304,
      "flos": 19898390102400.0,
      "grad_norm": 3.484197999820618,
      "language_loss": 0.7983672,
      "learning_rate": 6.542688368763034e-07,
      "loss": 0.82030272,
      "num_input_tokens_seen": 132753995,
      "step": 6178,
      "time_per_iteration": 2.648409366607666
    },
    {
      "auxiliary_loss_clip": 0.01169488,
      "auxiliary_loss_mlp": 0.0102104,
      "balance_loss_clip": 1.04760766,
      "balance_loss_mlp": 1.01393223,
      "epoch": 0.742980821258943,
      "flos": 24827201510400.0,
      "grad_norm": 1.74906272000895,
      "language_loss": 0.77043766,
      "learning_rate": 6.536926831855854e-07,
      "loss": 0.7923429,
      "num_input_tokens_seen": 132773160,
      "step": 6179,
      "time_per_iteration": 2.773009777069092
    },
    {
      "auxiliary_loss_clip": 0.011696,
      "auxiliary_loss_mlp": 0.01022707,
      "balance_loss_clip": 1.04451895,
      "balance_loss_mlp": 1.01528001,
      "epoch": 0.7431010641495821,
      "flos": 25228646887680.0,
      "grad_norm": 2.170892614415503,
      "language_loss": 0.72927308,
      "learning_rate": 6.531167337286165e-07,
      "loss": 0.75119615,
      "num_input_tokens_seen": 132793180,
      "step": 6180,
      "time_per_iteration": 2.64804744720459
    },
    {
      "auxiliary_loss_clip": 0.0117102,
      "auxiliary_loss_mlp": 0.01020727,
      "balance_loss_clip": 1.04612553,
      "balance_loss_mlp": 1.01360989,
      "epoch": 0.7432213070402213,
      "flos": 21762369550080.0,
      "grad_norm": 1.4646930918072154,
      "language_loss": 0.8001883,
      "learning_rate": 6.52540988592768e-07,
      "loss": 0.82210577,
      "num_input_tokens_seen": 132814200,
      "step": 6181,
      "time_per_iteration": 3.580437183380127
    },
    {
      "auxiliary_loss_clip": 0.01170625,
      "auxiliary_loss_mlp": 0.01021662,
      "balance_loss_clip": 1.04251528,
      "balance_loss_mlp": 1.01444364,
      "epoch": 0.7433415499308603,
      "flos": 14793832425600.0,
      "grad_norm": 3.2781544908491367,
      "language_loss": 0.83305883,
      "learning_rate": 6.519654478653814e-07,
      "loss": 0.85498172,
      "num_input_tokens_seen": 132832565,
      "step": 6182,
      "time_per_iteration": 2.650104284286499
    },
    {
      "auxiliary_loss_clip": 0.01080927,
      "auxiliary_loss_mlp": 0.01001885,
      "balance_loss_clip": 1.0120219,
      "balance_loss_mlp": 1.00102699,
      "epoch": 0.7434617928214994,
      "flos": 67155577297920.0,
      "grad_norm": 0.7509834480723848,
      "language_loss": 0.56060028,
      "learning_rate": 6.51390111633763e-07,
      "loss": 0.58142841,
      "num_input_tokens_seen": 132897840,
      "step": 6183,
      "time_per_iteration": 3.2433762550354004
    },
    {
      "auxiliary_loss_clip": 0.01177661,
      "auxiliary_loss_mlp": 0.01020394,
      "balance_loss_clip": 1.03805423,
      "balance_loss_mlp": 1.01322389,
      "epoch": 0.7435820357121385,
      "flos": 27377576928000.0,
      "grad_norm": 2.1772589643384963,
      "language_loss": 0.76412034,
      "learning_rate": 6.508149799851932e-07,
      "loss": 0.78610086,
      "num_input_tokens_seen": 132919505,
      "step": 6184,
      "time_per_iteration": 2.872917652130127
    },
    {
      "auxiliary_loss_clip": 0.0116634,
      "auxiliary_loss_mlp": 0.01020299,
      "balance_loss_clip": 1.04271591,
      "balance_loss_mlp": 1.01314926,
      "epoch": 0.7437022786027776,
      "flos": 23987645948160.0,
      "grad_norm": 8.12006341224109,
      "language_loss": 0.61425257,
      "learning_rate": 6.502400530069183e-07,
      "loss": 0.63611889,
      "num_input_tokens_seen": 132939390,
      "step": 6185,
      "time_per_iteration": 2.635432720184326
    },
    {
      "auxiliary_loss_clip": 0.01176202,
      "auxiliary_loss_mlp": 0.01030339,
      "balance_loss_clip": 1.04400826,
      "balance_loss_mlp": 1.02241755,
      "epoch": 0.7438225214934167,
      "flos": 21866761451520.0,
      "grad_norm": 3.169381864804302,
      "language_loss": 0.68567181,
      "learning_rate": 6.496653307861535e-07,
      "loss": 0.70773721,
      "num_input_tokens_seen": 132960060,
      "step": 6186,
      "time_per_iteration": 2.651542901992798
    },
    {
      "auxiliary_loss_clip": 0.01174822,
      "auxiliary_loss_mlp": 0.01026811,
      "balance_loss_clip": 1.04670441,
      "balance_loss_mlp": 1.01939917,
      "epoch": 0.7439427643840558,
      "flos": 20230097224320.0,
      "grad_norm": 2.2567748209289915,
      "language_loss": 0.65580535,
      "learning_rate": 6.490908134100857e-07,
      "loss": 0.6778217,
      "num_input_tokens_seen": 132978525,
      "step": 6187,
      "time_per_iteration": 2.684253692626953
    },
    {
      "auxiliary_loss_clip": 0.0117465,
      "auxiliary_loss_mlp": 0.01025449,
      "balance_loss_clip": 1.04646564,
      "balance_loss_mlp": 1.01768887,
      "epoch": 0.7440630072746949,
      "flos": 20849915335680.0,
      "grad_norm": 2.7340799455185536,
      "language_loss": 0.6972208,
      "learning_rate": 6.48516500965866e-07,
      "loss": 0.71922183,
      "num_input_tokens_seen": 132998460,
      "step": 6188,
      "time_per_iteration": 2.637392044067383
    },
    {
      "auxiliary_loss_clip": 0.01170642,
      "auxiliary_loss_mlp": 0.01021924,
      "balance_loss_clip": 1.0427624,
      "balance_loss_mlp": 1.01449168,
      "epoch": 0.7441832501653339,
      "flos": 26503762769280.0,
      "grad_norm": 2.175662032435744,
      "language_loss": 0.81966329,
      "learning_rate": 6.479423935406192e-07,
      "loss": 0.84158897,
      "num_input_tokens_seen": 133018445,
      "step": 6189,
      "time_per_iteration": 2.6246883869171143
    },
    {
      "auxiliary_loss_clip": 0.01073787,
      "auxiliary_loss_mlp": 0.01001567,
      "balance_loss_clip": 1.01367784,
      "balance_loss_mlp": 1.00066078,
      "epoch": 0.7443034930559731,
      "flos": 68602848088320.0,
      "grad_norm": 0.8028950974303979,
      "language_loss": 0.620112,
      "learning_rate": 6.473684912214357e-07,
      "loss": 0.64086556,
      "num_input_tokens_seen": 133082005,
      "step": 6190,
      "time_per_iteration": 3.4322431087493896
    },
    {
      "auxiliary_loss_clip": 0.01169871,
      "auxiliary_loss_mlp": 0.01021282,
      "balance_loss_clip": 1.04743719,
      "balance_loss_mlp": 1.01378131,
      "epoch": 0.7444237359466122,
      "flos": 18654982951680.0,
      "grad_norm": 2.3721875535380206,
      "language_loss": 0.69604748,
      "learning_rate": 6.467947940953778e-07,
      "loss": 0.71795905,
      "num_input_tokens_seen": 133100530,
      "step": 6191,
      "time_per_iteration": 2.6698977947235107
    },
    {
      "auxiliary_loss_clip": 0.01171792,
      "auxiliary_loss_mlp": 0.01024725,
      "balance_loss_clip": 1.04375136,
      "balance_loss_mlp": 1.01754522,
      "epoch": 0.7445439788372512,
      "flos": 22817604326400.0,
      "grad_norm": 1.9059936999101126,
      "language_loss": 0.7233693,
      "learning_rate": 6.462213022494732e-07,
      "loss": 0.74533439,
      "num_input_tokens_seen": 133119775,
      "step": 6192,
      "time_per_iteration": 2.6635239124298096
    },
    {
      "auxiliary_loss_clip": 0.01074203,
      "auxiliary_loss_mlp": 0.01002236,
      "balance_loss_clip": 1.01286602,
      "balance_loss_mlp": 1.00138938,
      "epoch": 0.7446642217278904,
      "flos": 67045690615680.0,
      "grad_norm": 0.7759109325556389,
      "language_loss": 0.60949469,
      "learning_rate": 6.456480157707201e-07,
      "loss": 0.63025904,
      "num_input_tokens_seen": 133184550,
      "step": 6193,
      "time_per_iteration": 3.1958985328674316
    },
    {
      "auxiliary_loss_clip": 0.01167525,
      "auxiliary_loss_mlp": 0.01025319,
      "balance_loss_clip": 1.04095054,
      "balance_loss_mlp": 1.01735592,
      "epoch": 0.7447844646185294,
      "flos": 17417465631360.0,
      "grad_norm": 2.0646455532951924,
      "language_loss": 0.84819126,
      "learning_rate": 6.450749347460866e-07,
      "loss": 0.87011969,
      "num_input_tokens_seen": 133201525,
      "step": 6194,
      "time_per_iteration": 2.625082492828369
    },
    {
      "auxiliary_loss_clip": 0.0116773,
      "auxiliary_loss_mlp": 0.01027386,
      "balance_loss_clip": 1.04692674,
      "balance_loss_mlp": 1.02014089,
      "epoch": 0.7449047075091685,
      "flos": 26615876094720.0,
      "grad_norm": 2.2146283702932603,
      "language_loss": 0.78885841,
      "learning_rate": 6.445020592625083e-07,
      "loss": 0.81080949,
      "num_input_tokens_seen": 133222175,
      "step": 6195,
      "time_per_iteration": 3.6635632514953613
    },
    {
      "auxiliary_loss_clip": 0.01165999,
      "auxiliary_loss_mlp": 0.0102303,
      "balance_loss_clip": 1.04608393,
      "balance_loss_mlp": 1.01519752,
      "epoch": 0.7450249503998077,
      "flos": 14170458867840.0,
      "grad_norm": 2.4689925892072133,
      "language_loss": 0.80329597,
      "learning_rate": 6.4392938940689e-07,
      "loss": 0.82518619,
      "num_input_tokens_seen": 133237590,
      "step": 6196,
      "time_per_iteration": 2.608854055404663
    },
    {
      "auxiliary_loss_clip": 0.0117533,
      "auxiliary_loss_mlp": 0.01207401,
      "balance_loss_clip": 1.04254687,
      "balance_loss_mlp": 1.00058746,
      "epoch": 0.7451451932904467,
      "flos": 19606687752960.0,
      "grad_norm": 2.3774533401330054,
      "language_loss": 0.71310902,
      "learning_rate": 6.433569252661049e-07,
      "loss": 0.73693633,
      "num_input_tokens_seen": 133255590,
      "step": 6197,
      "time_per_iteration": 2.6610164642333984
    },
    {
      "auxiliary_loss_clip": 0.01170447,
      "auxiliary_loss_mlp": 0.01021914,
      "balance_loss_clip": 1.04293633,
      "balance_loss_mlp": 1.01505983,
      "epoch": 0.7452654361810858,
      "flos": 12495405980160.0,
      "grad_norm": 1.9000485320032605,
      "language_loss": 0.71833587,
      "learning_rate": 6.427846669269952e-07,
      "loss": 0.74025953,
      "num_input_tokens_seen": 133273210,
      "step": 6198,
      "time_per_iteration": 2.711005926132202
    },
    {
      "auxiliary_loss_clip": 0.01169576,
      "auxiliary_loss_mlp": 0.0102455,
      "balance_loss_clip": 1.05043828,
      "balance_loss_mlp": 1.01763856,
      "epoch": 0.7453856790717249,
      "flos": 22127329687680.0,
      "grad_norm": 2.273306573081826,
      "language_loss": 0.82882297,
      "learning_rate": 6.422126144763729e-07,
      "loss": 0.85076416,
      "num_input_tokens_seen": 133292600,
      "step": 6199,
      "time_per_iteration": 2.5826034545898438
    },
    {
      "auxiliary_loss_clip": 0.01173657,
      "auxiliary_loss_mlp": 0.01207783,
      "balance_loss_clip": 1.03948784,
      "balance_loss_mlp": 1.00071478,
      "epoch": 0.745505921962364,
      "flos": 20010682995840.0,
      "grad_norm": 2.3228848275266056,
      "language_loss": 0.7723456,
      "learning_rate": 6.416407680010174e-07,
      "loss": 0.79615998,
      "num_input_tokens_seen": 133306960,
      "step": 6200,
      "time_per_iteration": 2.7210781574249268
    },
    {
      "auxiliary_loss_clip": 0.0119045,
      "auxiliary_loss_mlp": 0.01024803,
      "balance_loss_clip": 1.04449284,
      "balance_loss_mlp": 1.01667845,
      "epoch": 0.745626164853003,
      "flos": 24677884673280.0,
      "grad_norm": 1.8736599092291815,
      "language_loss": 0.81026781,
      "learning_rate": 6.410691275876774e-07,
      "loss": 0.83242041,
      "num_input_tokens_seen": 133326380,
      "step": 6201,
      "time_per_iteration": 2.727959156036377
    },
    {
      "auxiliary_loss_clip": 0.01179466,
      "auxiliary_loss_mlp": 0.01024863,
      "balance_loss_clip": 1.04652274,
      "balance_loss_mlp": 1.01734114,
      "epoch": 0.7457464077436422,
      "flos": 14538830797440.0,
      "grad_norm": 7.1463069334429274,
      "language_loss": 0.76569581,
      "learning_rate": 6.404976933230704e-07,
      "loss": 0.7877391,
      "num_input_tokens_seen": 133342900,
      "step": 6202,
      "time_per_iteration": 3.5300889015197754
    },
    {
      "auxiliary_loss_clip": 0.01176925,
      "auxiliary_loss_mlp": 0.01027222,
      "balance_loss_clip": 1.04507518,
      "balance_loss_mlp": 1.01934195,
      "epoch": 0.7458666506342813,
      "flos": 34021194600960.0,
      "grad_norm": 2.205657142075462,
      "language_loss": 0.726439,
      "learning_rate": 6.399264652938813e-07,
      "loss": 0.74848044,
      "num_input_tokens_seen": 133363805,
      "step": 6203,
      "time_per_iteration": 3.6837031841278076
    },
    {
      "auxiliary_loss_clip": 0.01171266,
      "auxiliary_loss_mlp": 0.01021753,
      "balance_loss_clip": 1.04426408,
      "balance_loss_mlp": 1.01441026,
      "epoch": 0.7459868935249203,
      "flos": 24279025075200.0,
      "grad_norm": 1.9126764534029137,
      "language_loss": 0.74470842,
      "learning_rate": 6.393554435867679e-07,
      "loss": 0.76663864,
      "num_input_tokens_seen": 133384655,
      "step": 6204,
      "time_per_iteration": 2.6617510318756104
    },
    {
      "auxiliary_loss_clip": 0.01171388,
      "auxiliary_loss_mlp": 0.01023009,
      "balance_loss_clip": 1.04176438,
      "balance_loss_mlp": 1.01522434,
      "epoch": 0.7461071364155595,
      "flos": 21908777385600.0,
      "grad_norm": 4.666425872659282,
      "language_loss": 0.84049976,
      "learning_rate": 6.387846282883502e-07,
      "loss": 0.86244369,
      "num_input_tokens_seen": 133401185,
      "step": 6205,
      "time_per_iteration": 2.73494815826416
    },
    {
      "auxiliary_loss_clip": 0.0116508,
      "auxiliary_loss_mlp": 0.01027061,
      "balance_loss_clip": 1.04640043,
      "balance_loss_mlp": 1.01976252,
      "epoch": 0.7462273793061985,
      "flos": 22889712879360.0,
      "grad_norm": 3.7646707505011214,
      "language_loss": 0.77041173,
      "learning_rate": 6.38214019485223e-07,
      "loss": 0.79233313,
      "num_input_tokens_seen": 133420010,
      "step": 6206,
      "time_per_iteration": 2.611055612564087
    },
    {
      "auxiliary_loss_clip": 0.01175743,
      "auxiliary_loss_mlp": 0.01023921,
      "balance_loss_clip": 1.03826296,
      "balance_loss_mlp": 1.01659,
      "epoch": 0.7463476221968376,
      "flos": 19968451580160.0,
      "grad_norm": 2.2890188710924733,
      "language_loss": 0.71834385,
      "learning_rate": 6.376436172639461e-07,
      "loss": 0.74034047,
      "num_input_tokens_seen": 133437855,
      "step": 6207,
      "time_per_iteration": 2.810961961746216
    },
    {
      "auxiliary_loss_clip": 0.01185172,
      "auxiliary_loss_mlp": 0.01024956,
      "balance_loss_clip": 1.03910136,
      "balance_loss_mlp": 1.01668835,
      "epoch": 0.7464678650874768,
      "flos": 16836610798080.0,
      "grad_norm": 2.4820567699184206,
      "language_loss": 0.65042681,
      "learning_rate": 6.370734217110487e-07,
      "loss": 0.67252803,
      "num_input_tokens_seen": 133456600,
      "step": 6208,
      "time_per_iteration": 3.750943899154663
    },
    {
      "auxiliary_loss_clip": 0.01177144,
      "auxiliary_loss_mlp": 0.01027774,
      "balance_loss_clip": 1.04765177,
      "balance_loss_mlp": 1.01975131,
      "epoch": 0.7465881079781158,
      "flos": 48100869843840.0,
      "grad_norm": 1.5257513561918854,
      "language_loss": 0.64419436,
      "learning_rate": 6.36503432913031e-07,
      "loss": 0.66624355,
      "num_input_tokens_seen": 133479745,
      "step": 6209,
      "time_per_iteration": 2.871565818786621
    },
    {
      "auxiliary_loss_clip": 0.01167064,
      "auxiliary_loss_mlp": 0.0102455,
      "balance_loss_clip": 1.04570162,
      "balance_loss_mlp": 1.01692915,
      "epoch": 0.7467083508687549,
      "flos": 19677359761920.0,
      "grad_norm": 2.7564851309915204,
      "language_loss": 0.69009024,
      "learning_rate": 6.359336509563569e-07,
      "loss": 0.71200633,
      "num_input_tokens_seen": 133495765,
      "step": 6210,
      "time_per_iteration": 2.6957931518554688
    },
    {
      "auxiliary_loss_clip": 0.01162331,
      "auxiliary_loss_mlp": 0.01027873,
      "balance_loss_clip": 1.04181015,
      "balance_loss_mlp": 1.02001739,
      "epoch": 0.7468285937593939,
      "flos": 17895436934400.0,
      "grad_norm": 1.9274874540690061,
      "language_loss": 0.80667901,
      "learning_rate": 6.353640759274641e-07,
      "loss": 0.82858109,
      "num_input_tokens_seen": 133514655,
      "step": 6211,
      "time_per_iteration": 2.7086503505706787
    },
    {
      "auxiliary_loss_clip": 0.0116602,
      "auxiliary_loss_mlp": 0.01024314,
      "balance_loss_clip": 1.04289055,
      "balance_loss_mlp": 1.01665461,
      "epoch": 0.7469488366500331,
      "flos": 23141446369920.0,
      "grad_norm": 2.5319238354348625,
      "language_loss": 0.75219691,
      "learning_rate": 6.347947079127556e-07,
      "loss": 0.77410018,
      "num_input_tokens_seen": 133532555,
      "step": 6212,
      "time_per_iteration": 2.676828145980835
    },
    {
      "auxiliary_loss_clip": 0.01168738,
      "auxiliary_loss_mlp": 0.01023206,
      "balance_loss_clip": 1.04397154,
      "balance_loss_mlp": 1.01526666,
      "epoch": 0.7470690795406721,
      "flos": 16690849407360.0,
      "grad_norm": 2.1103610591200015,
      "language_loss": 0.7713027,
      "learning_rate": 6.342255469986053e-07,
      "loss": 0.79322219,
      "num_input_tokens_seen": 133551300,
      "step": 6213,
      "time_per_iteration": 2.596496105194092
    },
    {
      "auxiliary_loss_clip": 0.01163957,
      "auxiliary_loss_mlp": 0.01022324,
      "balance_loss_clip": 1.04593086,
      "balance_loss_mlp": 1.01498687,
      "epoch": 0.7471893224313112,
      "flos": 25192700352000.0,
      "grad_norm": 1.8205524702106384,
      "language_loss": 0.76733255,
      "learning_rate": 6.336565932713533e-07,
      "loss": 0.78919542,
      "num_input_tokens_seen": 133570725,
      "step": 6214,
      "time_per_iteration": 2.796154260635376
    },
    {
      "auxiliary_loss_clip": 0.01169225,
      "auxiliary_loss_mlp": 0.01027048,
      "balance_loss_clip": 1.04671431,
      "balance_loss_mlp": 1.01934373,
      "epoch": 0.7473095653219504,
      "flos": 22526225199360.0,
      "grad_norm": 2.1006118412343406,
      "language_loss": 0.78055084,
      "learning_rate": 6.330878468173088e-07,
      "loss": 0.8025136,
      "num_input_tokens_seen": 133590790,
      "step": 6215,
      "time_per_iteration": 2.695960521697998
    },
    {
      "auxiliary_loss_clip": 0.01161638,
      "auxiliary_loss_mlp": 0.01023814,
      "balance_loss_clip": 1.04374504,
      "balance_loss_mlp": 1.01630437,
      "epoch": 0.7474298082125894,
      "flos": 18113989236480.0,
      "grad_norm": 18.159371418789355,
      "language_loss": 0.72960079,
      "learning_rate": 6.32519307722752e-07,
      "loss": 0.75145537,
      "num_input_tokens_seen": 133608685,
      "step": 6216,
      "time_per_iteration": 2.6282753944396973
    },
    {
      "auxiliary_loss_clip": 0.01082562,
      "auxiliary_loss_mlp": 0.0100052,
      "balance_loss_clip": 1.01547861,
      "balance_loss_mlp": 0.99966735,
      "epoch": 0.7475500511032285,
      "flos": 62086535193600.0,
      "grad_norm": 0.8389954789232758,
      "language_loss": 0.54967511,
      "learning_rate": 6.31950976073929e-07,
      "loss": 0.57050592,
      "num_input_tokens_seen": 133662775,
      "step": 6217,
      "time_per_iteration": 3.1627631187438965
    },
    {
      "auxiliary_loss_clip": 0.01171926,
      "auxiliary_loss_mlp": 0.0102544,
      "balance_loss_clip": 1.04177403,
      "balance_loss_mlp": 1.01818621,
      "epoch": 0.7476702939938676,
      "flos": 17785586165760.0,
      "grad_norm": 2.262664365520446,
      "language_loss": 0.80884552,
      "learning_rate": 6.31382851957055e-07,
      "loss": 0.83081913,
      "num_input_tokens_seen": 133679595,
      "step": 6218,
      "time_per_iteration": 2.6530821323394775
    },
    {
      "auxiliary_loss_clip": 0.01169484,
      "auxiliary_loss_mlp": 0.01207667,
      "balance_loss_clip": 1.0425607,
      "balance_loss_mlp": 1.00072944,
      "epoch": 0.7477905368845067,
      "flos": 27927944092800.0,
      "grad_norm": 2.39673568268299,
      "language_loss": 0.71270049,
      "learning_rate": 6.308149354583143e-07,
      "loss": 0.73647201,
      "num_input_tokens_seen": 133699000,
      "step": 6219,
      "time_per_iteration": 2.7145092487335205
    },
    {
      "auxiliary_loss_clip": 0.01174506,
      "auxiliary_loss_mlp": 0.01024395,
      "balance_loss_clip": 1.04708815,
      "balance_loss_mlp": 1.01611638,
      "epoch": 0.7479107797751458,
      "flos": 26870374932480.0,
      "grad_norm": 2.319699912273592,
      "language_loss": 0.81995666,
      "learning_rate": 6.302472266638586e-07,
      "loss": 0.84194565,
      "num_input_tokens_seen": 133719540,
      "step": 6220,
      "time_per_iteration": 2.6541435718536377
    },
    {
      "auxiliary_loss_clip": 0.01175314,
      "auxiliary_loss_mlp": 0.01024852,
      "balance_loss_clip": 1.04976988,
      "balance_loss_mlp": 1.01683521,
      "epoch": 0.7480310226657849,
      "flos": 33943375785600.0,
      "grad_norm": 2.246093630838979,
      "language_loss": 0.70263076,
      "learning_rate": 6.296797256598101e-07,
      "loss": 0.72463238,
      "num_input_tokens_seen": 133741020,
      "step": 6221,
      "time_per_iteration": 2.7460455894470215
    },
    {
      "auxiliary_loss_clip": 0.01165855,
      "auxiliary_loss_mlp": 0.01021773,
      "balance_loss_clip": 1.04206729,
      "balance_loss_mlp": 1.0143255,
      "epoch": 0.748151265556424,
      "flos": 24826555065600.0,
      "grad_norm": 2.7745963858421145,
      "language_loss": 0.81549752,
      "learning_rate": 6.291124325322576e-07,
      "loss": 0.83737379,
      "num_input_tokens_seen": 133761145,
      "step": 6222,
      "time_per_iteration": 3.7246854305267334
    },
    {
      "auxiliary_loss_clip": 0.01175753,
      "auxiliary_loss_mlp": 0.01021557,
      "balance_loss_clip": 1.04373431,
      "balance_loss_mlp": 1.01405239,
      "epoch": 0.748271508447063,
      "flos": 38399351535360.0,
      "grad_norm": 1.6649650797974695,
      "language_loss": 0.62179041,
      "learning_rate": 6.285453473672595e-07,
      "loss": 0.64376354,
      "num_input_tokens_seen": 133783715,
      "step": 6223,
      "time_per_iteration": 2.840841054916382
    },
    {
      "auxiliary_loss_clip": 0.01163888,
      "auxiliary_loss_mlp": 0.01023741,
      "balance_loss_clip": 1.04501832,
      "balance_loss_mlp": 1.0162909,
      "epoch": 0.7483917513377022,
      "flos": 21541842000000.0,
      "grad_norm": 2.1881475340244574,
      "language_loss": 0.75600612,
      "learning_rate": 6.279784702508415e-07,
      "loss": 0.77788246,
      "num_input_tokens_seen": 133804465,
      "step": 6224,
      "time_per_iteration": 2.6732871532440186
    },
    {
      "auxiliary_loss_clip": 0.01088612,
      "auxiliary_loss_mlp": 0.01002755,
      "balance_loss_clip": 1.01200294,
      "balance_loss_mlp": 1.00180721,
      "epoch": 0.7485119942283412,
      "flos": 62314532772480.0,
      "grad_norm": 0.7976771342941439,
      "language_loss": 0.58506584,
      "learning_rate": 6.274118012689979e-07,
      "loss": 0.6059795,
      "num_input_tokens_seen": 133866365,
      "step": 6225,
      "time_per_iteration": 3.307328939437866
    },
    {
      "auxiliary_loss_clip": 0.01163279,
      "auxiliary_loss_mlp": 0.01020937,
      "balance_loss_clip": 1.0425235,
      "balance_loss_mlp": 1.01350391,
      "epoch": 0.7486322371189803,
      "flos": 29937613104000.0,
      "grad_norm": 1.6201926796138735,
      "language_loss": 0.68576646,
      "learning_rate": 6.268453405076943e-07,
      "loss": 0.70760858,
      "num_input_tokens_seen": 133888760,
      "step": 6226,
      "time_per_iteration": 2.7539923191070557
    },
    {
      "auxiliary_loss_clip": 0.01171975,
      "auxiliary_loss_mlp": 0.01021111,
      "balance_loss_clip": 1.04333329,
      "balance_loss_mlp": 1.01403022,
      "epoch": 0.7487524800096195,
      "flos": 18949414734720.0,
      "grad_norm": 2.1539254440036157,
      "language_loss": 0.82786667,
      "learning_rate": 6.262790880528592e-07,
      "loss": 0.84979749,
      "num_input_tokens_seen": 133906380,
      "step": 6227,
      "time_per_iteration": 2.6870381832122803
    },
    {
      "auxiliary_loss_clip": 0.01185712,
      "auxiliary_loss_mlp": 0.01027786,
      "balance_loss_clip": 1.0419116,
      "balance_loss_mlp": 1.01972723,
      "epoch": 0.7488727229002585,
      "flos": 18697393935360.0,
      "grad_norm": 2.4055038116115908,
      "language_loss": 0.79866457,
      "learning_rate": 6.257130439903951e-07,
      "loss": 0.82079959,
      "num_input_tokens_seen": 133922875,
      "step": 6228,
      "time_per_iteration": 3.6099812984466553
    },
    {
      "auxiliary_loss_clip": 0.01170071,
      "auxiliary_loss_mlp": 0.01024583,
      "balance_loss_clip": 1.04941308,
      "balance_loss_mlp": 1.01698923,
      "epoch": 0.7489929657908976,
      "flos": 23623368168960.0,
      "grad_norm": 1.863471889990684,
      "language_loss": 0.81610715,
      "learning_rate": 6.251472084061695e-07,
      "loss": 0.8380537,
      "num_input_tokens_seen": 133941795,
      "step": 6229,
      "time_per_iteration": 2.6594347953796387
    },
    {
      "auxiliary_loss_clip": 0.01170043,
      "auxiliary_loss_mlp": 0.01023259,
      "balance_loss_clip": 1.04815173,
      "balance_loss_mlp": 1.016294,
      "epoch": 0.7491132086815367,
      "flos": 20551533056640.0,
      "grad_norm": 2.177730027192914,
      "language_loss": 0.89019859,
      "learning_rate": 6.245815813860191e-07,
      "loss": 0.91213167,
      "num_input_tokens_seen": 133957305,
      "step": 6230,
      "time_per_iteration": 3.4827921390533447
    },
    {
      "auxiliary_loss_clip": 0.01169362,
      "auxiliary_loss_mlp": 0.01022352,
      "balance_loss_clip": 1.04639697,
      "balance_loss_mlp": 1.01465118,
      "epoch": 0.7492334515721758,
      "flos": 23003011353600.0,
      "grad_norm": 2.2344937870178616,
      "language_loss": 0.70601797,
      "learning_rate": 6.240161630157495e-07,
      "loss": 0.72793514,
      "num_input_tokens_seen": 133976660,
      "step": 6231,
      "time_per_iteration": 2.6775729656219482
    },
    {
      "auxiliary_loss_clip": 0.01170277,
      "auxiliary_loss_mlp": 0.0102219,
      "balance_loss_clip": 1.04805243,
      "balance_loss_mlp": 1.01494837,
      "epoch": 0.7493536944628149,
      "flos": 16398823835520.0,
      "grad_norm": 2.1792535609360164,
      "language_loss": 0.70464545,
      "learning_rate": 6.23450953381133e-07,
      "loss": 0.72657013,
      "num_input_tokens_seen": 133994750,
      "step": 6232,
      "time_per_iteration": 2.529935359954834
    },
    {
      "auxiliary_loss_clip": 0.01164726,
      "auxiliary_loss_mlp": 0.0102277,
      "balance_loss_clip": 1.04287386,
      "balance_loss_mlp": 1.01537895,
      "epoch": 0.749473937353454,
      "flos": 15338561155200.0,
      "grad_norm": 3.3711084647995557,
      "language_loss": 0.67932302,
      "learning_rate": 6.228859525679131e-07,
      "loss": 0.70119798,
      "num_input_tokens_seen": 134009165,
      "step": 6233,
      "time_per_iteration": 2.699355363845825
    },
    {
      "auxiliary_loss_clip": 0.01169584,
      "auxiliary_loss_mlp": 0.01025171,
      "balance_loss_clip": 1.04605818,
      "balance_loss_mlp": 1.01744628,
      "epoch": 0.7495941802440931,
      "flos": 18951138587520.0,
      "grad_norm": 3.919643967475904,
      "language_loss": 0.80032825,
      "learning_rate": 6.223211606617986e-07,
      "loss": 0.82227576,
      "num_input_tokens_seen": 134027585,
      "step": 6234,
      "time_per_iteration": 2.6612250804901123
    },
    {
      "auxiliary_loss_clip": 0.01167562,
      "auxiliary_loss_mlp": 0.01024671,
      "balance_loss_clip": 1.04909182,
      "balance_loss_mlp": 1.0182699,
      "epoch": 0.7497144231347321,
      "flos": 22492469393280.0,
      "grad_norm": 1.7584771770528513,
      "language_loss": 0.84289181,
      "learning_rate": 6.217565777484701e-07,
      "loss": 0.86481416,
      "num_input_tokens_seen": 134046680,
      "step": 6235,
      "time_per_iteration": 3.5933027267456055
    },
    {
      "auxiliary_loss_clip": 0.01167525,
      "auxiliary_loss_mlp": 0.012076,
      "balance_loss_clip": 1.04362702,
      "balance_loss_mlp": 1.0006237,
      "epoch": 0.7498346660253713,
      "flos": 24243509502720.0,
      "grad_norm": 1.7921588788285319,
      "language_loss": 0.8007912,
      "learning_rate": 6.211922039135722e-07,
      "loss": 0.82454246,
      "num_input_tokens_seen": 134066825,
      "step": 6236,
      "time_per_iteration": 2.8303005695343018
    },
    {
      "auxiliary_loss_clip": 0.01167512,
      "auxiliary_loss_mlp": 0.01024544,
      "balance_loss_clip": 1.04767203,
      "balance_loss_mlp": 1.01692045,
      "epoch": 0.7499549089160104,
      "flos": 24387080163840.0,
      "grad_norm": 2.727710091168939,
      "language_loss": 0.80898082,
      "learning_rate": 6.206280392427201e-07,
      "loss": 0.83090138,
      "num_input_tokens_seen": 134086410,
      "step": 6237,
      "time_per_iteration": 2.59197735786438
    },
    {
      "auxiliary_loss_clip": 0.0116323,
      "auxiliary_loss_mlp": 0.01021852,
      "balance_loss_clip": 1.04303169,
      "balance_loss_mlp": 1.01456213,
      "epoch": 0.7500751518066494,
      "flos": 34057320704640.0,
      "grad_norm": 2.229606735279985,
      "language_loss": 0.7393834,
      "learning_rate": 6.200640838214983e-07,
      "loss": 0.76123422,
      "num_input_tokens_seen": 134109185,
      "step": 6238,
      "time_per_iteration": 2.7736873626708984
    },
    {
      "auxiliary_loss_clip": 0.01166513,
      "auxiliary_loss_mlp": 0.01027401,
      "balance_loss_clip": 1.04711771,
      "balance_loss_mlp": 1.01982486,
      "epoch": 0.7501953946972886,
      "flos": 18843586289280.0,
      "grad_norm": 2.142400029721762,
      "language_loss": 0.67072469,
      "learning_rate": 6.195003377354578e-07,
      "loss": 0.69266379,
      "num_input_tokens_seen": 134128455,
      "step": 6239,
      "time_per_iteration": 2.729727029800415
    },
    {
      "auxiliary_loss_clip": 0.01166699,
      "auxiliary_loss_mlp": 0.01025344,
      "balance_loss_clip": 1.04393888,
      "balance_loss_mlp": 1.01726437,
      "epoch": 0.7503156375879276,
      "flos": 20257675891200.0,
      "grad_norm": 2.5551176886723366,
      "language_loss": 0.73970145,
      "learning_rate": 6.189368010701183e-07,
      "loss": 0.76162189,
      "num_input_tokens_seen": 134145515,
      "step": 6240,
      "time_per_iteration": 2.689788818359375
    },
    {
      "auxiliary_loss_clip": 0.01174268,
      "auxiliary_loss_mlp": 0.01023303,
      "balance_loss_clip": 1.04490983,
      "balance_loss_mlp": 1.01556587,
      "epoch": 0.7504358804785667,
      "flos": 13480040574720.0,
      "grad_norm": 2.042197649016927,
      "language_loss": 0.76471305,
      "learning_rate": 6.183734739109683e-07,
      "loss": 0.78668875,
      "num_input_tokens_seen": 134163335,
      "step": 6241,
      "time_per_iteration": 2.603666305541992
    },
    {
      "auxiliary_loss_clip": 0.01176904,
      "auxiliary_loss_mlp": 0.01023928,
      "balance_loss_clip": 1.04710031,
      "balance_loss_mlp": 1.01610148,
      "epoch": 0.7505561233692057,
      "flos": 29461042431360.0,
      "grad_norm": 2.3898982262695854,
      "language_loss": 0.69115937,
      "learning_rate": 6.178103563434629e-07,
      "loss": 0.71316767,
      "num_input_tokens_seen": 134182335,
      "step": 6242,
      "time_per_iteration": 2.6950936317443848
    },
    {
      "auxiliary_loss_clip": 0.01165909,
      "auxiliary_loss_mlp": 0.0102457,
      "balance_loss_clip": 1.04582715,
      "balance_loss_mlp": 1.01723886,
      "epoch": 0.7506763662598449,
      "flos": 20302457172480.0,
      "grad_norm": 1.8834609341971316,
      "language_loss": 0.83806121,
      "learning_rate": 6.172474484530283e-07,
      "loss": 0.85996604,
      "num_input_tokens_seen": 134201070,
      "step": 6243,
      "time_per_iteration": 2.6009106636047363
    },
    {
      "auxiliary_loss_clip": 0.01162961,
      "auxiliary_loss_mlp": 0.01023748,
      "balance_loss_clip": 1.04054976,
      "balance_loss_mlp": 1.01603556,
      "epoch": 0.750796609150484,
      "flos": 37230961939200.0,
      "grad_norm": 2.75792490176286,
      "language_loss": 0.75796235,
      "learning_rate": 6.166847503250563e-07,
      "loss": 0.77982944,
      "num_input_tokens_seen": 134223310,
      "step": 6244,
      "time_per_iteration": 2.808687210083008
    },
    {
      "auxiliary_loss_clip": 0.01173003,
      "auxiliary_loss_mlp": 0.01024745,
      "balance_loss_clip": 1.04339933,
      "balance_loss_mlp": 1.01733625,
      "epoch": 0.750916852041123,
      "flos": 19609417186560.0,
      "grad_norm": 3.2862490360572947,
      "language_loss": 0.79046905,
      "learning_rate": 6.161222620449078e-07,
      "loss": 0.81244653,
      "num_input_tokens_seen": 134242085,
      "step": 6245,
      "time_per_iteration": 2.7042198181152344
    },
    {
      "auxiliary_loss_clip": 0.01176875,
      "auxiliary_loss_mlp": 0.0102682,
      "balance_loss_clip": 1.0435524,
      "balance_loss_mlp": 1.01929808,
      "epoch": 0.7510370949317622,
      "flos": 25112690807040.0,
      "grad_norm": 2.063749883644172,
      "language_loss": 0.80097485,
      "learning_rate": 6.155599836979117e-07,
      "loss": 0.82301176,
      "num_input_tokens_seen": 134260770,
      "step": 6246,
      "time_per_iteration": 2.6969869136810303
    },
    {
      "auxiliary_loss_clip": 0.01177532,
      "auxiliary_loss_mlp": 0.01025143,
      "balance_loss_clip": 1.03990006,
      "balance_loss_mlp": 1.01711416,
      "epoch": 0.7511573378224012,
      "flos": 19062282245760.0,
      "grad_norm": 2.130473781118687,
      "language_loss": 0.82140446,
      "learning_rate": 6.149979153693649e-07,
      "loss": 0.84343117,
      "num_input_tokens_seen": 134278025,
      "step": 6247,
      "time_per_iteration": 2.746976375579834
    },
    {
      "auxiliary_loss_clip": 0.01167962,
      "auxiliary_loss_mlp": 0.01020506,
      "balance_loss_clip": 1.04519749,
      "balance_loss_mlp": 1.01316226,
      "epoch": 0.7512775807130403,
      "flos": 19937676602880.0,
      "grad_norm": 1.94060066690122,
      "language_loss": 0.76930696,
      "learning_rate": 6.144360571445343e-07,
      "loss": 0.79119164,
      "num_input_tokens_seen": 134297170,
      "step": 6248,
      "time_per_iteration": 3.700371503829956
    },
    {
      "auxiliary_loss_clip": 0.01165579,
      "auxiliary_loss_mlp": 0.01021575,
      "balance_loss_clip": 1.04629803,
      "balance_loss_mlp": 1.01412475,
      "epoch": 0.7513978236036795,
      "flos": 20739920912640.0,
      "grad_norm": 1.7310628845700466,
      "language_loss": 0.80463648,
      "learning_rate": 6.138744091086509e-07,
      "loss": 0.82650799,
      "num_input_tokens_seen": 134316755,
      "step": 6249,
      "time_per_iteration": 2.666694402694702
    },
    {
      "auxiliary_loss_clip": 0.01178617,
      "auxiliary_loss_mlp": 0.01024194,
      "balance_loss_clip": 1.04464638,
      "balance_loss_mlp": 1.01685667,
      "epoch": 0.7515180664943185,
      "flos": 27563163523200.0,
      "grad_norm": 2.2963777756395887,
      "language_loss": 0.72837949,
      "learning_rate": 6.133129713469183e-07,
      "loss": 0.75040758,
      "num_input_tokens_seen": 134335960,
      "step": 6250,
      "time_per_iteration": 2.744213819503784
    },
    {
      "auxiliary_loss_clip": 0.01184488,
      "auxiliary_loss_mlp": 0.01024468,
      "balance_loss_clip": 1.04233479,
      "balance_loss_mlp": 1.01630211,
      "epoch": 0.7516383093849576,
      "flos": 33803181002880.0,
      "grad_norm": 3.5244975467744504,
      "language_loss": 0.64391041,
      "learning_rate": 6.127517439445053e-07,
      "loss": 0.66599989,
      "num_input_tokens_seen": 134356805,
      "step": 6251,
      "time_per_iteration": 2.995534896850586
    },
    {
      "auxiliary_loss_clip": 0.01171001,
      "auxiliary_loss_mlp": 0.0102408,
      "balance_loss_clip": 1.04214239,
      "balance_loss_mlp": 1.01682353,
      "epoch": 0.7517585522755967,
      "flos": 29746172592000.0,
      "grad_norm": 1.9655740348692599,
      "language_loss": 0.81715852,
      "learning_rate": 6.121907269865498e-07,
      "loss": 0.8391093,
      "num_input_tokens_seen": 134376295,
      "step": 6252,
      "time_per_iteration": 2.8554770946502686
    },
    {
      "auxiliary_loss_clip": 0.01082723,
      "auxiliary_loss_mlp": 0.01001592,
      "balance_loss_clip": 1.01401258,
      "balance_loss_mlp": 1.00072813,
      "epoch": 0.7518787951662358,
      "flos": 69807974319360.0,
      "grad_norm": 0.922285367060105,
      "language_loss": 0.67276132,
      "learning_rate": 6.116299205581577e-07,
      "loss": 0.69360447,
      "num_input_tokens_seen": 134431125,
      "step": 6253,
      "time_per_iteration": 3.1976370811462402
    },
    {
      "auxiliary_loss_clip": 0.01171421,
      "auxiliary_loss_mlp": 0.01024862,
      "balance_loss_clip": 1.04846454,
      "balance_loss_mlp": 1.01681209,
      "epoch": 0.7519990380568748,
      "flos": 34203225749760.0,
      "grad_norm": 2.358993290295363,
      "language_loss": 0.68461233,
      "learning_rate": 6.110693247444018e-07,
      "loss": 0.70657516,
      "num_input_tokens_seen": 134452960,
      "step": 6254,
      "time_per_iteration": 2.764317512512207
    },
    {
      "auxiliary_loss_clip": 0.0116198,
      "auxiliary_loss_mlp": 0.0102245,
      "balance_loss_clip": 1.04095459,
      "balance_loss_mlp": 1.01548266,
      "epoch": 0.752119280947514,
      "flos": 21725704742400.0,
      "grad_norm": 1.7745170494639517,
      "language_loss": 0.82482052,
      "learning_rate": 6.105089396303258e-07,
      "loss": 0.84666485,
      "num_input_tokens_seen": 134471350,
      "step": 6255,
      "time_per_iteration": 3.747568368911743
    },
    {
      "auxiliary_loss_clip": 0.01174317,
      "auxiliary_loss_mlp": 0.0102718,
      "balance_loss_clip": 1.04468298,
      "balance_loss_mlp": 1.01911592,
      "epoch": 0.7522395238381531,
      "flos": 32742774668160.0,
      "grad_norm": 1.8988155146854842,
      "language_loss": 0.75619012,
      "learning_rate": 6.099487653009383e-07,
      "loss": 0.7782051,
      "num_input_tokens_seen": 134490695,
      "step": 6256,
      "time_per_iteration": 2.8004395961761475
    },
    {
      "auxiliary_loss_clip": 0.01166651,
      "auxiliary_loss_mlp": 0.01021212,
      "balance_loss_clip": 1.04364717,
      "balance_loss_mlp": 1.01456857,
      "epoch": 0.7523597667287921,
      "flos": 23476026579840.0,
      "grad_norm": 2.282375048521687,
      "language_loss": 0.83339441,
      "learning_rate": 6.093888018412192e-07,
      "loss": 0.85527313,
      "num_input_tokens_seen": 134506885,
      "step": 6257,
      "time_per_iteration": 3.5980939865112305
    },
    {
      "auxiliary_loss_clip": 0.01074456,
      "auxiliary_loss_mlp": 0.01002206,
      "balance_loss_clip": 1.01293087,
      "balance_loss_mlp": 1.0012877,
      "epoch": 0.7524800096194313,
      "flos": 67346730501120.0,
      "grad_norm": 0.7075966513076396,
      "language_loss": 0.54636437,
      "learning_rate": 6.088290493361125e-07,
      "loss": 0.56713092,
      "num_input_tokens_seen": 134571770,
      "step": 6258,
      "time_per_iteration": 3.3831918239593506
    },
    {
      "auxiliary_loss_clip": 0.01168363,
      "auxiliary_loss_mlp": 0.01022623,
      "balance_loss_clip": 1.03996801,
      "balance_loss_mlp": 1.01523232,
      "epoch": 0.7526002525100703,
      "flos": 13006055681280.0,
      "grad_norm": 1.9473473813694206,
      "language_loss": 0.71743929,
      "learning_rate": 6.082695078705322e-07,
      "loss": 0.73934919,
      "num_input_tokens_seen": 134589250,
      "step": 6259,
      "time_per_iteration": 2.756321668624878
    },
    {
      "auxiliary_loss_clip": 0.01163735,
      "auxiliary_loss_mlp": 0.01026655,
      "balance_loss_clip": 1.04609978,
      "balance_loss_mlp": 1.0188055,
      "epoch": 0.7527204954007094,
      "flos": 21397229844480.0,
      "grad_norm": 2.1686999747276623,
      "language_loss": 0.68974394,
      "learning_rate": 6.077101775293618e-07,
      "loss": 0.71164787,
      "num_input_tokens_seen": 134608075,
      "step": 6260,
      "time_per_iteration": 2.6912238597869873
    },
    {
      "auxiliary_loss_clip": 0.01171914,
      "auxiliary_loss_mlp": 0.01024702,
      "balance_loss_clip": 1.04574633,
      "balance_loss_mlp": 1.0160594,
      "epoch": 0.7528407382913486,
      "flos": 18947188091520.0,
      "grad_norm": 2.378682796161618,
      "language_loss": 0.82445413,
      "learning_rate": 6.071510583974504e-07,
      "loss": 0.84642035,
      "num_input_tokens_seen": 134623260,
      "step": 6261,
      "time_per_iteration": 3.6251423358917236
    },
    {
      "auxiliary_loss_clip": 0.01168737,
      "auxiliary_loss_mlp": 0.01029342,
      "balance_loss_clip": 1.04750752,
      "balance_loss_mlp": 1.02170682,
      "epoch": 0.7529609811819876,
      "flos": 15231798956160.0,
      "grad_norm": 1.8993357216408069,
      "language_loss": 0.72264349,
      "learning_rate": 6.065921505596161e-07,
      "loss": 0.74462426,
      "num_input_tokens_seen": 134641540,
      "step": 6262,
      "time_per_iteration": 2.6707708835601807
    },
    {
      "auxiliary_loss_clip": 0.01175141,
      "auxiliary_loss_mlp": 0.01023762,
      "balance_loss_clip": 1.04433131,
      "balance_loss_mlp": 1.01622856,
      "epoch": 0.7530812240726267,
      "flos": 19354487385600.0,
      "grad_norm": 1.6324180882784005,
      "language_loss": 0.76989102,
      "learning_rate": 6.060334541006445e-07,
      "loss": 0.79188001,
      "num_input_tokens_seen": 134660035,
      "step": 6263,
      "time_per_iteration": 2.666229486465454
    },
    {
      "auxiliary_loss_clip": 0.01176299,
      "auxiliary_loss_mlp": 0.01028052,
      "balance_loss_clip": 1.04129601,
      "balance_loss_mlp": 1.02045226,
      "epoch": 0.7532014669632658,
      "flos": 27748247328000.0,
      "grad_norm": 1.8973829585301685,
      "language_loss": 0.68859321,
      "learning_rate": 6.05474969105289e-07,
      "loss": 0.71063668,
      "num_input_tokens_seen": 134683025,
      "step": 6264,
      "time_per_iteration": 2.7379560470581055
    },
    {
      "auxiliary_loss_clip": 0.01172419,
      "auxiliary_loss_mlp": 0.01021921,
      "balance_loss_clip": 1.04721856,
      "balance_loss_mlp": 1.01403522,
      "epoch": 0.7533217098539049,
      "flos": 14137421333760.0,
      "grad_norm": 2.210700206416431,
      "language_loss": 0.73803616,
      "learning_rate": 6.049166956582725e-07,
      "loss": 0.75997961,
      "num_input_tokens_seen": 134701290,
      "step": 6265,
      "time_per_iteration": 2.6557257175445557
    },
    {
      "auxiliary_loss_clip": 0.01163847,
      "auxiliary_loss_mlp": 0.01019913,
      "balance_loss_clip": 1.04337561,
      "balance_loss_mlp": 1.01263547,
      "epoch": 0.753441952744544,
      "flos": 26429068437120.0,
      "grad_norm": 2.2193140600759236,
      "language_loss": 0.8781209,
      "learning_rate": 6.043586338442841e-07,
      "loss": 0.89995849,
      "num_input_tokens_seen": 134720345,
      "step": 6266,
      "time_per_iteration": 2.658714771270752
    },
    {
      "auxiliary_loss_clip": 0.01163394,
      "auxiliary_loss_mlp": 0.01023674,
      "balance_loss_clip": 1.04706192,
      "balance_loss_mlp": 1.01716769,
      "epoch": 0.7535621956351831,
      "flos": 23878621192320.0,
      "grad_norm": 1.4767434013295082,
      "language_loss": 0.73286116,
      "learning_rate": 6.038007837479815e-07,
      "loss": 0.75473177,
      "num_input_tokens_seen": 134741450,
      "step": 6267,
      "time_per_iteration": 2.708380699157715
    },
    {
      "auxiliary_loss_clip": 0.01166619,
      "auxiliary_loss_mlp": 0.01020709,
      "balance_loss_clip": 1.04542732,
      "balance_loss_mlp": 1.01300764,
      "epoch": 0.7536824385258222,
      "flos": 21795873960960.0,
      "grad_norm": 1.9845996033428897,
      "language_loss": 0.64215595,
      "learning_rate": 6.032431454539897e-07,
      "loss": 0.66402924,
      "num_input_tokens_seen": 134760295,
      "step": 6268,
      "time_per_iteration": 2.6771557331085205
    },
    {
      "auxiliary_loss_clip": 0.01177193,
      "auxiliary_loss_mlp": 0.01026177,
      "balance_loss_clip": 1.04293513,
      "balance_loss_mlp": 1.01926589,
      "epoch": 0.7538026814164612,
      "flos": 28911644933760.0,
      "grad_norm": 2.6505494951900848,
      "language_loss": 0.81900656,
      "learning_rate": 6.026857190469014e-07,
      "loss": 0.84104025,
      "num_input_tokens_seen": 134782050,
      "step": 6269,
      "time_per_iteration": 2.7319695949554443
    },
    {
      "auxiliary_loss_clip": 0.0117395,
      "auxiliary_loss_mlp": 0.01023333,
      "balance_loss_clip": 1.04379141,
      "balance_loss_mlp": 1.01582015,
      "epoch": 0.7539229243071004,
      "flos": 21104701482240.0,
      "grad_norm": 1.9329035870315023,
      "language_loss": 0.73964494,
      "learning_rate": 6.0212850461128e-07,
      "loss": 0.76161778,
      "num_input_tokens_seen": 134801170,
      "step": 6270,
      "time_per_iteration": 2.6662180423736572
    },
    {
      "auxiliary_loss_clip": 0.01174506,
      "auxiliary_loss_mlp": 0.01025297,
      "balance_loss_clip": 1.04242635,
      "balance_loss_mlp": 1.01738167,
      "epoch": 0.7540431671977395,
      "flos": 15158469340800.0,
      "grad_norm": 3.1334242063687054,
      "language_loss": 0.74645662,
      "learning_rate": 6.015715022316516e-07,
      "loss": 0.76845467,
      "num_input_tokens_seen": 134819150,
      "step": 6271,
      "time_per_iteration": 2.622650623321533
    },
    {
      "auxiliary_loss_clip": 0.01177016,
      "auxiliary_loss_mlp": 0.0102076,
      "balance_loss_clip": 1.03925192,
      "balance_loss_mlp": 1.01288366,
      "epoch": 0.7541634100883785,
      "flos": 18770579896320.0,
      "grad_norm": 2.73721680482021,
      "language_loss": 0.78530538,
      "learning_rate": 6.010147119925154e-07,
      "loss": 0.80728316,
      "num_input_tokens_seen": 134836905,
      "step": 6272,
      "time_per_iteration": 2.6439239978790283
    },
    {
      "auxiliary_loss_clip": 0.01166252,
      "auxiliary_loss_mlp": 0.01025103,
      "balance_loss_clip": 1.04162359,
      "balance_loss_mlp": 1.01741433,
      "epoch": 0.7542836529790176,
      "flos": 20594770053120.0,
      "grad_norm": 2.335369099999269,
      "language_loss": 0.66395998,
      "learning_rate": 6.004581339783348e-07,
      "loss": 0.68587351,
      "num_input_tokens_seen": 134855225,
      "step": 6273,
      "time_per_iteration": 2.631531238555908
    },
    {
      "auxiliary_loss_clip": 0.01175828,
      "auxiliary_loss_mlp": 0.01033906,
      "balance_loss_clip": 1.04810393,
      "balance_loss_mlp": 1.02544236,
      "epoch": 0.7544038958696567,
      "flos": 19095104298240.0,
      "grad_norm": 2.585179974573381,
      "language_loss": 0.68818641,
      "learning_rate": 5.999017682735425e-07,
      "loss": 0.7102837,
      "num_input_tokens_seen": 134871615,
      "step": 6274,
      "time_per_iteration": 2.5766677856445312
    },
    {
      "auxiliary_loss_clip": 0.0118548,
      "auxiliary_loss_mlp": 0.0102803,
      "balance_loss_clip": 1.0399332,
      "balance_loss_mlp": 1.02023649,
      "epoch": 0.7545241387602958,
      "flos": 31723306859520.0,
      "grad_norm": 3.919001346828993,
      "language_loss": 0.6625005,
      "learning_rate": 5.993456149625387e-07,
      "loss": 0.68463564,
      "num_input_tokens_seen": 134892765,
      "step": 6275,
      "time_per_iteration": 3.7685580253601074
    },
    {
      "auxiliary_loss_clip": 0.01166161,
      "auxiliary_loss_mlp": 0.01021186,
      "balance_loss_clip": 1.0420351,
      "balance_loss_mlp": 1.01450086,
      "epoch": 0.7546443816509348,
      "flos": 20296495514880.0,
      "grad_norm": 1.72007753333387,
      "language_loss": 0.82379508,
      "learning_rate": 5.987896741296909e-07,
      "loss": 0.84566849,
      "num_input_tokens_seen": 134910505,
      "step": 6276,
      "time_per_iteration": 2.737164258956909
    },
    {
      "auxiliary_loss_clip": 0.0117212,
      "auxiliary_loss_mlp": 0.01026376,
      "balance_loss_clip": 1.04621315,
      "balance_loss_mlp": 1.01896977,
      "epoch": 0.754764624541574,
      "flos": 23696159080320.0,
      "grad_norm": 2.1964482414485462,
      "language_loss": 0.78590274,
      "learning_rate": 5.982339458593361e-07,
      "loss": 0.80788773,
      "num_input_tokens_seen": 134930445,
      "step": 6277,
      "time_per_iteration": 2.6741011142730713
    },
    {
      "auxiliary_loss_clip": 0.01164681,
      "auxiliary_loss_mlp": 0.01207794,
      "balance_loss_clip": 1.04539013,
      "balance_loss_mlp": 1.00065386,
      "epoch": 0.7548848674322131,
      "flos": 25337204766720.0,
      "grad_norm": 1.9586122434053523,
      "language_loss": 0.83942991,
      "learning_rate": 5.976784302357767e-07,
      "loss": 0.86315465,
      "num_input_tokens_seen": 134951010,
      "step": 6278,
      "time_per_iteration": 2.675119638442993
    },
    {
      "auxiliary_loss_clip": 0.01171861,
      "auxiliary_loss_mlp": 0.01026132,
      "balance_loss_clip": 1.04606628,
      "balance_loss_mlp": 1.01875293,
      "epoch": 0.7550051103228521,
      "flos": 19573147428480.0,
      "grad_norm": 2.081079871251538,
      "language_loss": 0.73501319,
      "learning_rate": 5.971231273432855e-07,
      "loss": 0.75699311,
      "num_input_tokens_seen": 134970495,
      "step": 6279,
      "time_per_iteration": 2.6167070865631104
    },
    {
      "auxiliary_loss_clip": 0.01073974,
      "auxiliary_loss_mlp": 0.01002546,
      "balance_loss_clip": 1.01315761,
      "balance_loss_mlp": 1.00166416,
      "epoch": 0.7551253532134913,
      "flos": 64150068648960.0,
      "grad_norm": 0.8133630009240177,
      "language_loss": 0.54575109,
      "learning_rate": 5.965680372661e-07,
      "loss": 0.56651628,
      "num_input_tokens_seen": 135028060,
      "step": 6280,
      "time_per_iteration": 3.076066255569458
    },
    {
      "auxiliary_loss_clip": 0.01173476,
      "auxiliary_loss_mlp": 0.01021038,
      "balance_loss_clip": 1.04601824,
      "balance_loss_mlp": 1.01440394,
      "epoch": 0.7552455961041303,
      "flos": 26067986968320.0,
      "grad_norm": 1.7356503633757758,
      "language_loss": 0.56094062,
      "learning_rate": 5.960131600884266e-07,
      "loss": 0.58288574,
      "num_input_tokens_seen": 135047330,
      "step": 6281,
      "time_per_iteration": 2.7471208572387695
    },
    {
      "auxiliary_loss_clip": 0.01176389,
      "auxiliary_loss_mlp": 0.01020757,
      "balance_loss_clip": 1.04282165,
      "balance_loss_mlp": 1.01400113,
      "epoch": 0.7553658389947694,
      "flos": 24498223822080.0,
      "grad_norm": 2.13073433966484,
      "language_loss": 0.76239508,
      "learning_rate": 5.954584958944413e-07,
      "loss": 0.78436655,
      "num_input_tokens_seen": 135065995,
      "step": 6282,
      "time_per_iteration": 3.6580421924591064
    },
    {
      "auxiliary_loss_clip": 0.01177126,
      "auxiliary_loss_mlp": 0.01207952,
      "balance_loss_clip": 1.04156649,
      "balance_loss_mlp": 1.00072837,
      "epoch": 0.7554860818854086,
      "flos": 21799465320960.0,
      "grad_norm": 4.3283958343404025,
      "language_loss": 0.81918067,
      "learning_rate": 5.949040447682854e-07,
      "loss": 0.84303147,
      "num_input_tokens_seen": 135085820,
      "step": 6283,
      "time_per_iteration": 2.7077174186706543
    },
    {
      "auxiliary_loss_clip": 0.01177865,
      "auxiliary_loss_mlp": 0.01023923,
      "balance_loss_clip": 1.0450536,
      "balance_loss_mlp": 1.01645148,
      "epoch": 0.7556063247760476,
      "flos": 16362123114240.0,
      "grad_norm": 2.1980419425978948,
      "language_loss": 0.68339628,
      "learning_rate": 5.943498067940686e-07,
      "loss": 0.70541418,
      "num_input_tokens_seen": 135102845,
      "step": 6284,
      "time_per_iteration": 3.5872294902801514
    },
    {
      "auxiliary_loss_clip": 0.01165427,
      "auxiliary_loss_mlp": 0.01027063,
      "balance_loss_clip": 1.04793143,
      "balance_loss_mlp": 1.01958215,
      "epoch": 0.7557265676666867,
      "flos": 27235155502080.0,
      "grad_norm": 2.0475408418446843,
      "language_loss": 0.81657708,
      "learning_rate": 5.937957820558686e-07,
      "loss": 0.83850193,
      "num_input_tokens_seen": 135122190,
      "step": 6285,
      "time_per_iteration": 2.7253129482269287
    },
    {
      "auxiliary_loss_clip": 0.01081628,
      "auxiliary_loss_mlp": 0.0100114,
      "balance_loss_clip": 1.01247323,
      "balance_loss_mlp": 1.00023389,
      "epoch": 0.7558468105573258,
      "flos": 62189131415040.0,
      "grad_norm": 0.8516534136954582,
      "language_loss": 0.6522153,
      "learning_rate": 5.932419706377296e-07,
      "loss": 0.67304295,
      "num_input_tokens_seen": 135180495,
      "step": 6286,
      "time_per_iteration": 3.1487114429473877
    },
    {
      "auxiliary_loss_clip": 0.01172083,
      "auxiliary_loss_mlp": 0.01021626,
      "balance_loss_clip": 1.04556811,
      "balance_loss_mlp": 1.01413321,
      "epoch": 0.7559670534479649,
      "flos": 33249078823680.0,
      "grad_norm": 2.3510022241449855,
      "language_loss": 0.74155223,
      "learning_rate": 5.92688372623666e-07,
      "loss": 0.76348925,
      "num_input_tokens_seen": 135199200,
      "step": 6287,
      "time_per_iteration": 3.6720056533813477
    },
    {
      "auxiliary_loss_clip": 0.01169363,
      "auxiliary_loss_mlp": 0.0102126,
      "balance_loss_clip": 1.04334116,
      "balance_loss_mlp": 1.01373243,
      "epoch": 0.7560872963386039,
      "flos": 14064379027200.0,
      "grad_norm": 1.9691688214823944,
      "language_loss": 0.73743534,
      "learning_rate": 5.921349880976574e-07,
      "loss": 0.7593416,
      "num_input_tokens_seen": 135217035,
      "step": 6288,
      "time_per_iteration": 2.6567115783691406
    },
    {
      "auxiliary_loss_clip": 0.01173259,
      "auxiliary_loss_mlp": 0.01207918,
      "balance_loss_clip": 1.04188609,
      "balance_loss_mlp": 1.00079918,
      "epoch": 0.7562075392292431,
      "flos": 20412307941120.0,
      "grad_norm": 1.923807930131741,
      "language_loss": 0.81595629,
      "learning_rate": 5.915818171436515e-07,
      "loss": 0.83976805,
      "num_input_tokens_seen": 135236370,
      "step": 6289,
      "time_per_iteration": 2.6311373710632324
    },
    {
      "auxiliary_loss_clip": 0.01171159,
      "auxiliary_loss_mlp": 0.01025738,
      "balance_loss_clip": 1.04100835,
      "balance_loss_mlp": 1.01832294,
      "epoch": 0.7563277821198822,
      "flos": 20376792368640.0,
      "grad_norm": 1.7017975506372063,
      "language_loss": 0.74651515,
      "learning_rate": 5.910288598455642e-07,
      "loss": 0.76848412,
      "num_input_tokens_seen": 135255720,
      "step": 6290,
      "time_per_iteration": 2.7142133712768555
    },
    {
      "auxiliary_loss_clip": 0.01176137,
      "auxiliary_loss_mlp": 0.01027299,
      "balance_loss_clip": 1.04687142,
      "balance_loss_mlp": 1.01952362,
      "epoch": 0.7564480250105212,
      "flos": 18588261438720.0,
      "grad_norm": 2.374194882269843,
      "language_loss": 0.74481058,
      "learning_rate": 5.90476116287278e-07,
      "loss": 0.76684487,
      "num_input_tokens_seen": 135273320,
      "step": 6291,
      "time_per_iteration": 2.609341859817505
    },
    {
      "auxiliary_loss_clip": 0.01174088,
      "auxiliary_loss_mlp": 0.01026694,
      "balance_loss_clip": 1.04805088,
      "balance_loss_mlp": 1.01934457,
      "epoch": 0.7565682679011604,
      "flos": 21215521918080.0,
      "grad_norm": 1.7918573747592088,
      "language_loss": 0.68179142,
      "learning_rate": 5.899235865526456e-07,
      "loss": 0.70379925,
      "num_input_tokens_seen": 135292615,
      "step": 6292,
      "time_per_iteration": 2.692394495010376
    },
    {
      "auxiliary_loss_clip": 0.01166406,
      "auxiliary_loss_mlp": 0.01024897,
      "balance_loss_clip": 1.04134417,
      "balance_loss_mlp": 1.01811731,
      "epoch": 0.7566885107917994,
      "flos": 20449008662400.0,
      "grad_norm": 1.8795627747299455,
      "language_loss": 0.82327205,
      "learning_rate": 5.893712707254825e-07,
      "loss": 0.84518504,
      "num_input_tokens_seen": 135310075,
      "step": 6293,
      "time_per_iteration": 2.6886074542999268
    },
    {
      "auxiliary_loss_clip": 0.01171586,
      "auxiliary_loss_mlp": 0.01022179,
      "balance_loss_clip": 1.03837204,
      "balance_loss_mlp": 1.01411414,
      "epoch": 0.7568087536824385,
      "flos": 19025832919680.0,
      "grad_norm": 3.240236153315225,
      "language_loss": 0.66553843,
      "learning_rate": 5.888191688895769e-07,
      "loss": 0.6874761,
      "num_input_tokens_seen": 135327335,
      "step": 6294,
      "time_per_iteration": 2.817352056503296
    },
    {
      "auxiliary_loss_clip": 0.01166437,
      "auxiliary_loss_mlp": 0.01026961,
      "balance_loss_clip": 1.0446794,
      "balance_loss_mlp": 1.01889706,
      "epoch": 0.7569289965730777,
      "flos": 15225442248960.0,
      "grad_norm": 2.098099329547896,
      "language_loss": 0.61802137,
      "learning_rate": 5.882672811286813e-07,
      "loss": 0.63995528,
      "num_input_tokens_seen": 135343615,
      "step": 6295,
      "time_per_iteration": 2.6440954208374023
    },
    {
      "auxiliary_loss_clip": 0.01169564,
      "auxiliary_loss_mlp": 0.01026564,
      "balance_loss_clip": 1.04737353,
      "balance_loss_mlp": 1.01889277,
      "epoch": 0.7570492394637167,
      "flos": 20769367086720.0,
      "grad_norm": 2.193236239736664,
      "language_loss": 0.69404304,
      "learning_rate": 5.877156075265166e-07,
      "loss": 0.71600437,
      "num_input_tokens_seen": 135359880,
      "step": 6296,
      "time_per_iteration": 2.6422247886657715
    },
    {
      "auxiliary_loss_clip": 0.01170412,
      "auxiliary_loss_mlp": 0.01021857,
      "balance_loss_clip": 1.04207754,
      "balance_loss_mlp": 1.01401925,
      "epoch": 0.7571694823543558,
      "flos": 15664091137920.0,
      "grad_norm": 3.983718207939817,
      "language_loss": 0.6979562,
      "learning_rate": 5.871641481667715e-07,
      "loss": 0.71987891,
      "num_input_tokens_seen": 135374325,
      "step": 6297,
      "time_per_iteration": 2.6328744888305664
    },
    {
      "auxiliary_loss_clip": 0.01180379,
      "auxiliary_loss_mlp": 0.01025897,
      "balance_loss_clip": 1.04221892,
      "balance_loss_mlp": 1.01817811,
      "epoch": 0.7572897252449949,
      "flos": 25409241492480.0,
      "grad_norm": 2.1166287787886744,
      "language_loss": 0.84526426,
      "learning_rate": 5.866129031331011e-07,
      "loss": 0.86732703,
      "num_input_tokens_seen": 135393980,
      "step": 6298,
      "time_per_iteration": 2.775425672531128
    },
    {
      "auxiliary_loss_clip": 0.01172303,
      "auxiliary_loss_mlp": 0.01023172,
      "balance_loss_clip": 1.04167271,
      "balance_loss_mlp": 1.01577783,
      "epoch": 0.757409968135634,
      "flos": 24279348297600.0,
      "grad_norm": 2.1732585983601207,
      "language_loss": 0.83516663,
      "learning_rate": 5.8606187250913e-07,
      "loss": 0.85712135,
      "num_input_tokens_seen": 135412030,
      "step": 6299,
      "time_per_iteration": 2.7518348693847656
    },
    {
      "auxiliary_loss_clip": 0.01168256,
      "auxiliary_loss_mlp": 0.01207751,
      "balance_loss_clip": 1.04704881,
      "balance_loss_mlp": 1.00071752,
      "epoch": 0.757530211026273,
      "flos": 24133766474880.0,
      "grad_norm": 1.94814549859387,
      "language_loss": 0.83950377,
      "learning_rate": 5.855110563784482e-07,
      "loss": 0.86326385,
      "num_input_tokens_seen": 135430565,
      "step": 6300,
      "time_per_iteration": 2.638334274291992
    },
    {
      "auxiliary_loss_clip": 0.0116333,
      "auxiliary_loss_mlp": 0.01207725,
      "balance_loss_clip": 1.04330409,
      "balance_loss_mlp": 1.00063527,
      "epoch": 0.7576504539169122,
      "flos": 23951807153280.0,
      "grad_norm": 3.3794944542014145,
      "language_loss": 0.64237362,
      "learning_rate": 5.849604548246156e-07,
      "loss": 0.66608417,
      "num_input_tokens_seen": 135451675,
      "step": 6301,
      "time_per_iteration": 2.6585283279418945
    },
    {
      "auxiliary_loss_clip": 0.0117815,
      "auxiliary_loss_mlp": 0.01207353,
      "balance_loss_clip": 1.04652429,
      "balance_loss_mlp": 1.00072157,
      "epoch": 0.7577706968075513,
      "flos": 21251360712960.0,
      "grad_norm": 2.3631145193906646,
      "language_loss": 0.80131423,
      "learning_rate": 5.844100679311565e-07,
      "loss": 0.82516927,
      "num_input_tokens_seen": 135470635,
      "step": 6302,
      "time_per_iteration": 3.6427175998687744
    },
    {
      "auxiliary_loss_clip": 0.01175898,
      "auxiliary_loss_mlp": 0.01022259,
      "balance_loss_clip": 1.04740226,
      "balance_loss_mlp": 1.01424801,
      "epoch": 0.7578909396981903,
      "flos": 18296595002880.0,
      "grad_norm": 2.2441439989378007,
      "language_loss": 0.76423812,
      "learning_rate": 5.838598957815637e-07,
      "loss": 0.78621972,
      "num_input_tokens_seen": 135487865,
      "step": 6303,
      "time_per_iteration": 2.744770050048828
    },
    {
      "auxiliary_loss_clip": 0.01164586,
      "auxiliary_loss_mlp": 0.01022077,
      "balance_loss_clip": 1.04212713,
      "balance_loss_mlp": 1.01490331,
      "epoch": 0.7580111825888295,
      "flos": 25373869574400.0,
      "grad_norm": 1.8021184825859373,
      "language_loss": 0.85577905,
      "learning_rate": 5.833099384592996e-07,
      "loss": 0.87764573,
      "num_input_tokens_seen": 135508440,
      "step": 6304,
      "time_per_iteration": 2.7538821697235107
    },
    {
      "auxiliary_loss_clip": 0.01168844,
      "auxiliary_loss_mlp": 0.01026563,
      "balance_loss_clip": 1.04411888,
      "balance_loss_mlp": 1.01882052,
      "epoch": 0.7581314254794685,
      "flos": 23768662682880.0,
      "grad_norm": 2.1697511764133206,
      "language_loss": 0.71444964,
      "learning_rate": 5.827601960477913e-07,
      "loss": 0.7364037,
      "num_input_tokens_seen": 135526365,
      "step": 6305,
      "time_per_iteration": 2.6927731037139893
    },
    {
      "auxiliary_loss_clip": 0.01165179,
      "auxiliary_loss_mlp": 0.01028709,
      "balance_loss_clip": 1.04328561,
      "balance_loss_mlp": 1.02215207,
      "epoch": 0.7582516683701076,
      "flos": 22054610603520.0,
      "grad_norm": 1.8165939764363057,
      "language_loss": 0.7063325,
      "learning_rate": 5.822106686304344e-07,
      "loss": 0.72827137,
      "num_input_tokens_seen": 135545655,
      "step": 6306,
      "time_per_iteration": 2.7074501514434814
    },
    {
      "auxiliary_loss_clip": 0.01181,
      "auxiliary_loss_mlp": 0.01023321,
      "balance_loss_clip": 1.04254055,
      "balance_loss_mlp": 1.01569784,
      "epoch": 0.7583719112607467,
      "flos": 31649725848960.0,
      "grad_norm": 2.757678750829436,
      "language_loss": 0.5800038,
      "learning_rate": 5.816613562905919e-07,
      "loss": 0.60204697,
      "num_input_tokens_seen": 135566840,
      "step": 6307,
      "time_per_iteration": 2.7301230430603027
    },
    {
      "auxiliary_loss_clip": 0.01169865,
      "auxiliary_loss_mlp": 0.01024216,
      "balance_loss_clip": 1.04549778,
      "balance_loss_mlp": 1.0170157,
      "epoch": 0.7584921541513858,
      "flos": 33068376478080.0,
      "grad_norm": 1.592871689379479,
      "language_loss": 0.6995827,
      "learning_rate": 5.811122591115933e-07,
      "loss": 0.72152346,
      "num_input_tokens_seen": 135587825,
      "step": 6308,
      "time_per_iteration": 3.7221426963806152
    },
    {
      "auxiliary_loss_clip": 0.0117403,
      "auxiliary_loss_mlp": 0.01024158,
      "balance_loss_clip": 1.04602957,
      "balance_loss_mlp": 1.01702988,
      "epoch": 0.7586123970420249,
      "flos": 23326350606720.0,
      "grad_norm": 3.173822211956781,
      "language_loss": 0.71017855,
      "learning_rate": 5.805633771767376e-07,
      "loss": 0.73216045,
      "num_input_tokens_seen": 135605220,
      "step": 6309,
      "time_per_iteration": 2.7204060554504395
    },
    {
      "auxiliary_loss_clip": 0.01170179,
      "auxiliary_loss_mlp": 0.01023851,
      "balance_loss_clip": 1.04525495,
      "balance_loss_mlp": 1.01578081,
      "epoch": 0.7587326399326639,
      "flos": 18334229477760.0,
      "grad_norm": 1.866581586087557,
      "language_loss": 0.77576548,
      "learning_rate": 5.800147105692888e-07,
      "loss": 0.79770577,
      "num_input_tokens_seen": 135624795,
      "step": 6310,
      "time_per_iteration": 3.5717575550079346
    },
    {
      "auxiliary_loss_clip": 0.01169141,
      "auxiliary_loss_mlp": 0.01021803,
      "balance_loss_clip": 1.04299545,
      "balance_loss_mlp": 1.01467121,
      "epoch": 0.7588528828233031,
      "flos": 17275080119040.0,
      "grad_norm": 2.0820495905416023,
      "language_loss": 0.79094332,
      "learning_rate": 5.794662593724795e-07,
      "loss": 0.81285274,
      "num_input_tokens_seen": 135643800,
      "step": 6311,
      "time_per_iteration": 2.687204360961914
    },
    {
      "auxiliary_loss_clip": 0.01169644,
      "auxiliary_loss_mlp": 0.0102959,
      "balance_loss_clip": 1.04921031,
      "balance_loss_mlp": 1.02184153,
      "epoch": 0.7589731257139422,
      "flos": 17713621267200.0,
      "grad_norm": 1.9004769791779974,
      "language_loss": 0.75287789,
      "learning_rate": 5.789180236695091e-07,
      "loss": 0.77487028,
      "num_input_tokens_seen": 135660655,
      "step": 6312,
      "time_per_iteration": 2.572277784347534
    },
    {
      "auxiliary_loss_clip": 0.01164033,
      "auxiliary_loss_mlp": 0.01024125,
      "balance_loss_clip": 1.04507518,
      "balance_loss_mlp": 1.01746452,
      "epoch": 0.7590933686045812,
      "flos": 15961072786560.0,
      "grad_norm": 1.902185172538873,
      "language_loss": 0.85087192,
      "learning_rate": 5.78370003543544e-07,
      "loss": 0.8727535,
      "num_input_tokens_seen": 135679410,
      "step": 6313,
      "time_per_iteration": 2.6266520023345947
    },
    {
      "auxiliary_loss_clip": 0.01169698,
      "auxiliary_loss_mlp": 0.01207843,
      "balance_loss_clip": 1.04626453,
      "balance_loss_mlp": 1.00076187,
      "epoch": 0.7592136114952204,
      "flos": 21068072588160.0,
      "grad_norm": 4.9774704559448395,
      "language_loss": 0.84161496,
      "learning_rate": 5.778221990777203e-07,
      "loss": 0.86539036,
      "num_input_tokens_seen": 135697150,
      "step": 6314,
      "time_per_iteration": 3.682196617126465
    },
    {
      "auxiliary_loss_clip": 0.0117526,
      "auxiliary_loss_mlp": 0.01024951,
      "balance_loss_clip": 1.04653382,
      "balance_loss_mlp": 1.01732135,
      "epoch": 0.7593338543858594,
      "flos": 25297666871040.0,
      "grad_norm": 2.7751559839849897,
      "language_loss": 0.82714027,
      "learning_rate": 5.772746103551372e-07,
      "loss": 0.84914237,
      "num_input_tokens_seen": 135712545,
      "step": 6315,
      "time_per_iteration": 2.6532464027404785
    },
    {
      "auxiliary_loss_clip": 0.01166367,
      "auxiliary_loss_mlp": 0.0102033,
      "balance_loss_clip": 1.04355645,
      "balance_loss_mlp": 1.01279545,
      "epoch": 0.7594540972764985,
      "flos": 31832367528960.0,
      "grad_norm": 1.8204814394016953,
      "language_loss": 0.72125602,
      "learning_rate": 5.767272374588648e-07,
      "loss": 0.74312305,
      "num_input_tokens_seen": 135733950,
      "step": 6316,
      "time_per_iteration": 2.8267982006073
    },
    {
      "auxiliary_loss_clip": 0.01169124,
      "auxiliary_loss_mlp": 0.01021927,
      "balance_loss_clip": 1.04788446,
      "balance_loss_mlp": 1.01434517,
      "epoch": 0.7595743401671377,
      "flos": 37597250880000.0,
      "grad_norm": 4.237836547281151,
      "language_loss": 0.78241849,
      "learning_rate": 5.76180080471939e-07,
      "loss": 0.80432904,
      "num_input_tokens_seen": 135757120,
      "step": 6317,
      "time_per_iteration": 2.753903865814209
    },
    {
      "auxiliary_loss_clip": 0.0117322,
      "auxiliary_loss_mlp": 0.01024016,
      "balance_loss_clip": 1.04857659,
      "balance_loss_mlp": 1.01621938,
      "epoch": 0.7596945830577767,
      "flos": 18287724343680.0,
      "grad_norm": 3.6106567754324805,
      "language_loss": 0.72553861,
      "learning_rate": 5.756331394773631e-07,
      "loss": 0.74751091,
      "num_input_tokens_seen": 135773335,
      "step": 6318,
      "time_per_iteration": 2.536177396774292
    },
    {
      "auxiliary_loss_clip": 0.01178071,
      "auxiliary_loss_mlp": 0.01208185,
      "balance_loss_clip": 1.03835344,
      "balance_loss_mlp": 1.00065136,
      "epoch": 0.7598148259484158,
      "flos": 22233122219520.0,
      "grad_norm": 2.159819851915767,
      "language_loss": 0.76176351,
      "learning_rate": 5.750864145581071e-07,
      "loss": 0.78562605,
      "num_input_tokens_seen": 135792555,
      "step": 6319,
      "time_per_iteration": 2.7563717365264893
    },
    {
      "auxiliary_loss_clip": 0.01166478,
      "auxiliary_loss_mlp": 0.01024621,
      "balance_loss_clip": 1.04792714,
      "balance_loss_mlp": 1.01748657,
      "epoch": 0.7599350688390549,
      "flos": 27161718145920.0,
      "grad_norm": 2.3848017350416124,
      "language_loss": 0.86310804,
      "learning_rate": 5.745399057971085e-07,
      "loss": 0.885019,
      "num_input_tokens_seen": 135813690,
      "step": 6320,
      "time_per_iteration": 2.63360595703125
    },
    {
      "auxiliary_loss_clip": 0.0117378,
      "auxiliary_loss_mlp": 0.01025848,
      "balance_loss_clip": 1.04615641,
      "balance_loss_mlp": 1.01851678,
      "epoch": 0.760055311729694,
      "flos": 15560704817280.0,
      "grad_norm": 3.4975149869972104,
      "language_loss": 0.7535814,
      "learning_rate": 5.739936132772738e-07,
      "loss": 0.77557766,
      "num_input_tokens_seen": 135832255,
      "step": 6321,
      "time_per_iteration": 2.6764650344848633
    },
    {
      "auxiliary_loss_clip": 0.01164931,
      "auxiliary_loss_mlp": 0.01024139,
      "balance_loss_clip": 1.04535806,
      "balance_loss_mlp": 1.01630092,
      "epoch": 0.760175554620333,
      "flos": 25155496840320.0,
      "grad_norm": 5.287313713662496,
      "language_loss": 0.74213374,
      "learning_rate": 5.734475370814733e-07,
      "loss": 0.76402438,
      "num_input_tokens_seen": 135851935,
      "step": 6322,
      "time_per_iteration": 2.7275121212005615
    },
    {
      "auxiliary_loss_clip": 0.01168769,
      "auxiliary_loss_mlp": 0.01022593,
      "balance_loss_clip": 1.04230583,
      "balance_loss_mlp": 1.01523185,
      "epoch": 0.7602957975109722,
      "flos": 24353791234560.0,
      "grad_norm": 2.1516832066451133,
      "language_loss": 0.78534168,
      "learning_rate": 5.729016772925483e-07,
      "loss": 0.80725527,
      "num_input_tokens_seen": 135873510,
      "step": 6323,
      "time_per_iteration": 2.7095837593078613
    },
    {
      "auxiliary_loss_clip": 0.01174047,
      "auxiliary_loss_mlp": 0.01023325,
      "balance_loss_clip": 1.04328942,
      "balance_loss_mlp": 1.01537967,
      "epoch": 0.7604160404016113,
      "flos": 25192664438400.0,
      "grad_norm": 5.927468218179777,
      "language_loss": 0.70845854,
      "learning_rate": 5.723560339933038e-07,
      "loss": 0.73043221,
      "num_input_tokens_seen": 135893845,
      "step": 6324,
      "time_per_iteration": 2.6992409229278564
    },
    {
      "auxiliary_loss_clip": 0.01167624,
      "auxiliary_loss_mlp": 0.01207903,
      "balance_loss_clip": 1.04478085,
      "balance_loss_mlp": 1.0007503,
      "epoch": 0.7605362832922503,
      "flos": 29861841363840.0,
      "grad_norm": 6.599972696077122,
      "language_loss": 0.65468371,
      "learning_rate": 5.71810607266513e-07,
      "loss": 0.67843896,
      "num_input_tokens_seen": 135912430,
      "step": 6325,
      "time_per_iteration": 2.7102315425872803
    },
    {
      "auxiliary_loss_clip": 0.01168398,
      "auxiliary_loss_mlp": 0.01020738,
      "balance_loss_clip": 1.04414165,
      "balance_loss_mlp": 1.01330566,
      "epoch": 0.7606565261828895,
      "flos": 13917935278080.0,
      "grad_norm": 1.9237187755352112,
      "language_loss": 0.60038197,
      "learning_rate": 5.712653971949184e-07,
      "loss": 0.62227333,
      "num_input_tokens_seen": 135930550,
      "step": 6326,
      "time_per_iteration": 2.5577335357666016
    },
    {
      "auxiliary_loss_clip": 0.01163906,
      "auxiliary_loss_mlp": 0.01020587,
      "balance_loss_clip": 1.04429078,
      "balance_loss_mlp": 1.01289833,
      "epoch": 0.7607767690735285,
      "flos": 18551273408640.0,
      "grad_norm": 2.335449863906079,
      "language_loss": 0.74955946,
      "learning_rate": 5.707204038612268e-07,
      "loss": 0.77140445,
      "num_input_tokens_seen": 135947980,
      "step": 6327,
      "time_per_iteration": 2.6615164279937744
    },
    {
      "auxiliary_loss_clip": 0.01184984,
      "auxiliary_loss_mlp": 0.01026293,
      "balance_loss_clip": 1.05273032,
      "balance_loss_mlp": 1.01789498,
      "epoch": 0.7608970119641676,
      "flos": 20922993555840.0,
      "grad_norm": 2.4300024795928485,
      "language_loss": 0.73595607,
      "learning_rate": 5.701756273481138e-07,
      "loss": 0.7580688,
      "num_input_tokens_seen": 135965400,
      "step": 6328,
      "time_per_iteration": 2.5960171222686768
    },
    {
      "auxiliary_loss_clip": 0.01176382,
      "auxiliary_loss_mlp": 0.01023353,
      "balance_loss_clip": 1.0445013,
      "balance_loss_mlp": 1.0161736,
      "epoch": 0.7610172548548068,
      "flos": 23807302738560.0,
      "grad_norm": 1.5911906017631305,
      "language_loss": 0.74109119,
      "learning_rate": 5.696310677382212e-07,
      "loss": 0.76308852,
      "num_input_tokens_seen": 135986795,
      "step": 6329,
      "time_per_iteration": 3.5957705974578857
    },
    {
      "auxiliary_loss_clip": 0.01082139,
      "auxiliary_loss_mlp": 0.01003912,
      "balance_loss_clip": 1.0141921,
      "balance_loss_mlp": 1.00321424,
      "epoch": 0.7611374977454458,
      "flos": 66496580426880.0,
      "grad_norm": 0.8762723373901584,
      "language_loss": 0.61707956,
      "learning_rate": 5.690867251141576e-07,
      "loss": 0.63794005,
      "num_input_tokens_seen": 136053450,
      "step": 6330,
      "time_per_iteration": 3.3087315559387207
    },
    {
      "auxiliary_loss_clip": 0.01176354,
      "auxiliary_loss_mlp": 0.01024173,
      "balance_loss_clip": 1.04587626,
      "balance_loss_mlp": 1.01683021,
      "epoch": 0.7612577406360849,
      "flos": 15633136592640.0,
      "grad_norm": 2.584744228016124,
      "language_loss": 0.91635013,
      "learning_rate": 5.685425995585013e-07,
      "loss": 0.93835533,
      "num_input_tokens_seen": 136071375,
      "step": 6331,
      "time_per_iteration": 2.5682711601257324
    },
    {
      "auxiliary_loss_clip": 0.01081086,
      "auxiliary_loss_mlp": 0.01002287,
      "balance_loss_clip": 1.01243901,
      "balance_loss_mlp": 1.00147045,
      "epoch": 0.761377983526724,
      "flos": 60526253237760.0,
      "grad_norm": 0.7611452305696678,
      "language_loss": 0.59006357,
      "learning_rate": 5.679986911537935e-07,
      "loss": 0.6108973,
      "num_input_tokens_seen": 136138905,
      "step": 6332,
      "time_per_iteration": 3.34833025932312
    },
    {
      "auxiliary_loss_clip": 0.01163632,
      "auxiliary_loss_mlp": 0.01020554,
      "balance_loss_clip": 1.04029489,
      "balance_loss_mlp": 1.01322913,
      "epoch": 0.7614982264173631,
      "flos": 35772522019200.0,
      "grad_norm": 2.702282815554429,
      "language_loss": 0.67364287,
      "learning_rate": 5.674549999825462e-07,
      "loss": 0.69548476,
      "num_input_tokens_seen": 136161720,
      "step": 6333,
      "time_per_iteration": 2.7530038356781006
    },
    {
      "auxiliary_loss_clip": 0.0107285,
      "auxiliary_loss_mlp": 0.01001655,
      "balance_loss_clip": 1.01145244,
      "balance_loss_mlp": 1.0008148,
      "epoch": 0.7616184693080021,
      "flos": 67925502345600.0,
      "grad_norm": 0.9182844480183145,
      "language_loss": 0.71406263,
      "learning_rate": 5.669115261272363e-07,
      "loss": 0.73480767,
      "num_input_tokens_seen": 136222040,
      "step": 6334,
      "time_per_iteration": 3.1668615341186523
    },
    {
      "auxiliary_loss_clip": 0.01170392,
      "auxiliary_loss_mlp": 0.01029487,
      "balance_loss_clip": 1.04629982,
      "balance_loss_mlp": 1.02199471,
      "epoch": 0.7617387121986413,
      "flos": 20521979141760.0,
      "grad_norm": 2.9172873296368276,
      "language_loss": 0.72854626,
      "learning_rate": 5.663682696703081e-07,
      "loss": 0.75054502,
      "num_input_tokens_seen": 136240305,
      "step": 6335,
      "time_per_iteration": 3.682600259780884
    },
    {
      "auxiliary_loss_clip": 0.01165834,
      "auxiliary_loss_mlp": 0.010223,
      "balance_loss_clip": 1.04765451,
      "balance_loss_mlp": 1.01506448,
      "epoch": 0.7618589550892804,
      "flos": 18624495283200.0,
      "grad_norm": 6.16989911647509,
      "language_loss": 0.81954336,
      "learning_rate": 5.658252306941746e-07,
      "loss": 0.8414247,
      "num_input_tokens_seen": 136259625,
      "step": 6336,
      "time_per_iteration": 2.554455518722534
    },
    {
      "auxiliary_loss_clip": 0.01179908,
      "auxiliary_loss_mlp": 0.01028719,
      "balance_loss_clip": 1.04374266,
      "balance_loss_mlp": 1.02043426,
      "epoch": 0.7619791979799194,
      "flos": 17453735389440.0,
      "grad_norm": 2.5106674342501765,
      "language_loss": 0.75584543,
      "learning_rate": 5.65282409281212e-07,
      "loss": 0.77793169,
      "num_input_tokens_seen": 136277090,
      "step": 6337,
      "time_per_iteration": 3.601487874984741
    },
    {
      "auxiliary_loss_clip": 0.01170122,
      "auxiliary_loss_mlp": 0.0102339,
      "balance_loss_clip": 1.04390764,
      "balance_loss_mlp": 1.01602888,
      "epoch": 0.7620994408705585,
      "flos": 14137421333760.0,
      "grad_norm": 2.4080393294505695,
      "language_loss": 0.69680941,
      "learning_rate": 5.64739805513768e-07,
      "loss": 0.71874458,
      "num_input_tokens_seen": 136294635,
      "step": 6338,
      "time_per_iteration": 2.618823289871216
    },
    {
      "auxiliary_loss_clip": 0.01067754,
      "auxiliary_loss_mlp": 0.01198659,
      "balance_loss_clip": 1.01145244,
      "balance_loss_mlp": 0.99980462,
      "epoch": 0.7622196837611976,
      "flos": 70708792527360.0,
      "grad_norm": 0.7852521502715052,
      "language_loss": 0.55690789,
      "learning_rate": 5.641974194741541e-07,
      "loss": 0.57957196,
      "num_input_tokens_seen": 136350320,
      "step": 6339,
      "time_per_iteration": 3.086383819580078
    },
    {
      "auxiliary_loss_clip": 0.01073695,
      "auxiliary_loss_mlp": 0.01001627,
      "balance_loss_clip": 1.0209142,
      "balance_loss_mlp": 1.0007863,
      "epoch": 0.7623399266518367,
      "flos": 60684150447360.0,
      "grad_norm": 0.7779198824472037,
      "language_loss": 0.63684607,
      "learning_rate": 5.636552512446502e-07,
      "loss": 0.65759933,
      "num_input_tokens_seen": 136411375,
      "step": 6340,
      "time_per_iteration": 3.105557441711426
    },
    {
      "auxiliary_loss_clip": 0.01163315,
      "auxiliary_loss_mlp": 0.01024972,
      "balance_loss_clip": 1.04396892,
      "balance_loss_mlp": 1.01756024,
      "epoch": 0.7624601695424758,
      "flos": 26468893641600.0,
      "grad_norm": 1.7772819532425743,
      "language_loss": 0.77962017,
      "learning_rate": 5.631133009075027e-07,
      "loss": 0.80150306,
      "num_input_tokens_seen": 136430560,
      "step": 6341,
      "time_per_iteration": 3.6009390354156494
    },
    {
      "auxiliary_loss_clip": 0.01171095,
      "auxiliary_loss_mlp": 0.01207524,
      "balance_loss_clip": 1.04721749,
      "balance_loss_mlp": 1.00066662,
      "epoch": 0.7625804124331149,
      "flos": 19135755515520.0,
      "grad_norm": 1.9803394105465284,
      "language_loss": 0.68613589,
      "learning_rate": 5.625715685449242e-07,
      "loss": 0.70992208,
      "num_input_tokens_seen": 136448665,
      "step": 6342,
      "time_per_iteration": 2.7166495323181152
    },
    {
      "auxiliary_loss_clip": 0.01172937,
      "auxiliary_loss_mlp": 0.01025159,
      "balance_loss_clip": 1.04694223,
      "balance_loss_mlp": 1.01818824,
      "epoch": 0.762700655323754,
      "flos": 26213101914240.0,
      "grad_norm": 2.1259882906874155,
      "language_loss": 0.71673393,
      "learning_rate": 5.620300542390966e-07,
      "loss": 0.73871493,
      "num_input_tokens_seen": 136469710,
      "step": 6343,
      "time_per_iteration": 2.686213493347168
    },
    {
      "auxiliary_loss_clip": 0.01167724,
      "auxiliary_loss_mlp": 0.0102521,
      "balance_loss_clip": 1.04175043,
      "balance_loss_mlp": 1.01845407,
      "epoch": 0.762820898214393,
      "flos": 22382582711040.0,
      "grad_norm": 1.909930340251467,
      "language_loss": 0.85237139,
      "learning_rate": 5.614887580721659e-07,
      "loss": 0.87430072,
      "num_input_tokens_seen": 136489855,
      "step": 6344,
      "time_per_iteration": 2.6571648120880127
    },
    {
      "auxiliary_loss_clip": 0.01167984,
      "auxiliary_loss_mlp": 0.01025598,
      "balance_loss_clip": 1.04473901,
      "balance_loss_mlp": 1.01776266,
      "epoch": 0.7629411411050322,
      "flos": 15700504550400.0,
      "grad_norm": 2.122135663228131,
      "language_loss": 0.73728609,
      "learning_rate": 5.609476801262481e-07,
      "loss": 0.75922191,
      "num_input_tokens_seen": 136504715,
      "step": 6345,
      "time_per_iteration": 2.6686360836029053
    },
    {
      "auxiliary_loss_clip": 0.01173796,
      "auxiliary_loss_mlp": 0.01027906,
      "balance_loss_clip": 1.04570782,
      "balance_loss_mlp": 1.02040195,
      "epoch": 0.7630613839956712,
      "flos": 13770342293760.0,
      "grad_norm": 3.102708216387801,
      "language_loss": 0.64089739,
      "learning_rate": 5.604068204834223e-07,
      "loss": 0.6629144,
      "num_input_tokens_seen": 136521610,
      "step": 6346,
      "time_per_iteration": 2.751007556915283
    },
    {
      "auxiliary_loss_clip": 0.01174543,
      "auxiliary_loss_mlp": 0.01207768,
      "balance_loss_clip": 1.04197562,
      "balance_loss_mlp": 1.00068796,
      "epoch": 0.7631816268863103,
      "flos": 14569569861120.0,
      "grad_norm": 2.198199266916692,
      "language_loss": 0.76447636,
      "learning_rate": 5.598661792257367e-07,
      "loss": 0.78829956,
      "num_input_tokens_seen": 136538655,
      "step": 6347,
      "time_per_iteration": 2.7307608127593994
    },
    {
      "auxiliary_loss_clip": 0.01166867,
      "auxiliary_loss_mlp": 0.01022529,
      "balance_loss_clip": 1.04410863,
      "balance_loss_mlp": 1.01524854,
      "epoch": 0.7633018697769495,
      "flos": 19062210418560.0,
      "grad_norm": 1.813872153739103,
      "language_loss": 0.7548244,
      "learning_rate": 5.593257564352071e-07,
      "loss": 0.77671838,
      "num_input_tokens_seen": 136557095,
      "step": 6348,
      "time_per_iteration": 2.6234920024871826
    },
    {
      "auxiliary_loss_clip": 0.01167443,
      "auxiliary_loss_mlp": 0.01020335,
      "balance_loss_clip": 1.04620802,
      "balance_loss_mlp": 1.01306915,
      "epoch": 0.7634221126675885,
      "flos": 22052958577920.0,
      "grad_norm": 1.5232171728597657,
      "language_loss": 0.75447875,
      "learning_rate": 5.58785552193815e-07,
      "loss": 0.77635658,
      "num_input_tokens_seen": 136577340,
      "step": 6349,
      "time_per_iteration": 2.7287347316741943
    },
    {
      "auxiliary_loss_clip": 0.011677,
      "auxiliary_loss_mlp": 0.01021884,
      "balance_loss_clip": 1.04726291,
      "balance_loss_mlp": 1.01477337,
      "epoch": 0.7635423555582276,
      "flos": 29382720825600.0,
      "grad_norm": 1.765242255230624,
      "language_loss": 0.7542221,
      "learning_rate": 5.582455665835086e-07,
      "loss": 0.77611792,
      "num_input_tokens_seen": 136597635,
      "step": 6350,
      "time_per_iteration": 2.6461715698242188
    },
    {
      "auxiliary_loss_clip": 0.01181716,
      "auxiliary_loss_mlp": 0.01029821,
      "balance_loss_clip": 1.04327643,
      "balance_loss_mlp": 1.02147079,
      "epoch": 0.7636625984488667,
      "flos": 17784903807360.0,
      "grad_norm": 7.410357687706129,
      "language_loss": 0.72768593,
      "learning_rate": 5.577057996862036e-07,
      "loss": 0.74980134,
      "num_input_tokens_seen": 136615260,
      "step": 6351,
      "time_per_iteration": 2.8446524143218994
    },
    {
      "auxiliary_loss_clip": 0.01163498,
      "auxiliary_loss_mlp": 0.01025746,
      "balance_loss_clip": 1.04703069,
      "balance_loss_mlp": 1.01875758,
      "epoch": 0.7637828413395058,
      "flos": 23734583654400.0,
      "grad_norm": 2.335402392911555,
      "language_loss": 0.76147521,
      "learning_rate": 5.571662515837814e-07,
      "loss": 0.78336763,
      "num_input_tokens_seen": 136637220,
      "step": 6352,
      "time_per_iteration": 2.6068661212921143
    },
    {
      "auxiliary_loss_clip": 0.01169383,
      "auxiliary_loss_mlp": 0.01023276,
      "balance_loss_clip": 1.043818,
      "balance_loss_mlp": 1.01587033,
      "epoch": 0.7639030842301449,
      "flos": 36283279461120.0,
      "grad_norm": 1.872516890756218,
      "language_loss": 0.83922005,
      "learning_rate": 5.566269223580926e-07,
      "loss": 0.86114669,
      "num_input_tokens_seen": 136658930,
      "step": 6353,
      "time_per_iteration": 2.8158929347991943
    },
    {
      "auxiliary_loss_clip": 0.01171726,
      "auxiliary_loss_mlp": 0.01022155,
      "balance_loss_clip": 1.04635692,
      "balance_loss_mlp": 1.01514864,
      "epoch": 0.764023327120784,
      "flos": 28878104609280.0,
      "grad_norm": 1.7744167300514788,
      "language_loss": 0.75440478,
      "learning_rate": 5.560878120909511e-07,
      "loss": 0.77634358,
      "num_input_tokens_seen": 136681530,
      "step": 6354,
      "time_per_iteration": 2.6275429725646973
    },
    {
      "auxiliary_loss_clip": 0.01073277,
      "auxiliary_loss_mlp": 0.01000723,
      "balance_loss_clip": 1.0118736,
      "balance_loss_mlp": 0.99988252,
      "epoch": 0.7641435700114231,
      "flos": 64789711067520.0,
      "grad_norm": 0.8480261353761223,
      "language_loss": 0.58561432,
      "learning_rate": 5.55548920864141e-07,
      "loss": 0.60635424,
      "num_input_tokens_seen": 136742185,
      "step": 6355,
      "time_per_iteration": 3.224501848220825
    },
    {
      "auxiliary_loss_clip": 0.01171711,
      "auxiliary_loss_mlp": 0.01022528,
      "balance_loss_clip": 1.04990411,
      "balance_loss_mlp": 1.01583123,
      "epoch": 0.7642638129020621,
      "flos": 16835784785280.0,
      "grad_norm": 2.697302742920487,
      "language_loss": 0.77883327,
      "learning_rate": 5.550102487594113e-07,
      "loss": 0.80077565,
      "num_input_tokens_seen": 136760855,
      "step": 6356,
      "time_per_iteration": 3.509678602218628
    },
    {
      "auxiliary_loss_clip": 0.01181839,
      "auxiliary_loss_mlp": 0.01207346,
      "balance_loss_clip": 1.04118037,
      "balance_loss_mlp": 1.0006032,
      "epoch": 0.7643840557927013,
      "flos": 30408940391040.0,
      "grad_norm": 1.71279578130761,
      "language_loss": 0.71829778,
      "learning_rate": 5.54471795858477e-07,
      "loss": 0.74218965,
      "num_input_tokens_seen": 136780925,
      "step": 6357,
      "time_per_iteration": 2.722381830215454
    },
    {
      "auxiliary_loss_clip": 0.01173222,
      "auxiliary_loss_mlp": 0.01024702,
      "balance_loss_clip": 1.0394814,
      "balance_loss_mlp": 1.01697695,
      "epoch": 0.7645042986833404,
      "flos": 16983234115200.0,
      "grad_norm": 2.2904054806785634,
      "language_loss": 0.83602822,
      "learning_rate": 5.539335622430235e-07,
      "loss": 0.85800755,
      "num_input_tokens_seen": 136799545,
      "step": 6358,
      "time_per_iteration": 2.587219476699829
    },
    {
      "auxiliary_loss_clip": 0.01161834,
      "auxiliary_loss_mlp": 0.01025325,
      "balance_loss_clip": 1.04263437,
      "balance_loss_mlp": 1.01736784,
      "epoch": 0.7646245415739794,
      "flos": 17311493531520.0,
      "grad_norm": 2.7863042063318453,
      "language_loss": 0.74786782,
      "learning_rate": 5.533955479946975e-07,
      "loss": 0.76973945,
      "num_input_tokens_seen": 136818325,
      "step": 6359,
      "time_per_iteration": 2.5918211936950684
    },
    {
      "auxiliary_loss_clip": 0.01080803,
      "auxiliary_loss_mlp": 0.01198539,
      "balance_loss_clip": 1.02076793,
      "balance_loss_mlp": 0.99989361,
      "epoch": 0.7647447844646186,
      "flos": 70402332666240.0,
      "grad_norm": 0.8558997262069951,
      "language_loss": 0.65731251,
      "learning_rate": 5.528577531951173e-07,
      "loss": 0.68010592,
      "num_input_tokens_seen": 136878730,
      "step": 6360,
      "time_per_iteration": 3.214998960494995
    },
    {
      "auxiliary_loss_clip": 0.01178153,
      "auxiliary_loss_mlp": 0.010232,
      "balance_loss_clip": 1.04660106,
      "balance_loss_mlp": 1.01630998,
      "epoch": 0.7648650273552576,
      "flos": 17675914965120.0,
      "grad_norm": 2.253324104467922,
      "language_loss": 0.73507816,
      "learning_rate": 5.523201779258653e-07,
      "loss": 0.7570917,
      "num_input_tokens_seen": 136897705,
      "step": 6361,
      "time_per_iteration": 3.6682865619659424
    },
    {
      "auxiliary_loss_clip": 0.01163807,
      "auxiliary_loss_mlp": 0.01021995,
      "balance_loss_clip": 1.04470325,
      "balance_loss_mlp": 1.01436591,
      "epoch": 0.7649852702458967,
      "flos": 22162019247360.0,
      "grad_norm": 1.9885197671788182,
      "language_loss": 0.8401494,
      "learning_rate": 5.517828222684912e-07,
      "loss": 0.86200738,
      "num_input_tokens_seen": 136918360,
      "step": 6362,
      "time_per_iteration": 2.592308521270752
    },
    {
      "auxiliary_loss_clip": 0.01074648,
      "auxiliary_loss_mlp": 0.01001909,
      "balance_loss_clip": 1.01127326,
      "balance_loss_mlp": 1.00112176,
      "epoch": 0.7651055131365359,
      "flos": 69848338227840.0,
      "grad_norm": 0.7910761002486247,
      "language_loss": 0.59008861,
      "learning_rate": 5.512456863045117e-07,
      "loss": 0.61085415,
      "num_input_tokens_seen": 136979050,
      "step": 6363,
      "time_per_iteration": 4.098569869995117
    },
    {
      "auxiliary_loss_clip": 0.01166705,
      "auxiliary_loss_mlp": 0.01024527,
      "balance_loss_clip": 1.04562616,
      "balance_loss_mlp": 1.01692152,
      "epoch": 0.7652257560271749,
      "flos": 19464014931840.0,
      "grad_norm": 1.9522129655412426,
      "language_loss": 0.74184465,
      "learning_rate": 5.507087701154089e-07,
      "loss": 0.76375699,
      "num_input_tokens_seen": 136998970,
      "step": 6364,
      "time_per_iteration": 2.6192266941070557
    },
    {
      "auxiliary_loss_clip": 0.0117849,
      "auxiliary_loss_mlp": 0.01028421,
      "balance_loss_clip": 1.04129875,
      "balance_loss_mlp": 1.02135181,
      "epoch": 0.765345998917814,
      "flos": 15961108700160.0,
      "grad_norm": 4.291309216817904,
      "language_loss": 0.75102997,
      "learning_rate": 5.50172073782634e-07,
      "loss": 0.77309906,
      "num_input_tokens_seen": 137016950,
      "step": 6365,
      "time_per_iteration": 2.716951370239258
    },
    {
      "auxiliary_loss_clip": 0.01170311,
      "auxiliary_loss_mlp": 0.01024606,
      "balance_loss_clip": 1.04318142,
      "balance_loss_mlp": 1.0173279,
      "epoch": 0.7654662418084531,
      "flos": 23659853408640.0,
      "grad_norm": 1.9669785413696863,
      "language_loss": 0.87638253,
      "learning_rate": 5.496355973876023e-07,
      "loss": 0.8983317,
      "num_input_tokens_seen": 137036205,
      "step": 6366,
      "time_per_iteration": 2.6870715618133545
    },
    {
      "auxiliary_loss_clip": 0.0117239,
      "auxiliary_loss_mlp": 0.01208173,
      "balance_loss_clip": 1.04329872,
      "balance_loss_mlp": 1.00063241,
      "epoch": 0.7655864846990922,
      "flos": 41463608878080.0,
      "grad_norm": 1.8548749437609153,
      "language_loss": 0.70853472,
      "learning_rate": 5.490993410116984e-07,
      "loss": 0.73234034,
      "num_input_tokens_seen": 137059195,
      "step": 6367,
      "time_per_iteration": 2.8159666061401367
    },
    {
      "auxiliary_loss_clip": 0.01166898,
      "auxiliary_loss_mlp": 0.01026805,
      "balance_loss_clip": 1.04201913,
      "balance_loss_mlp": 1.01978397,
      "epoch": 0.7657067275897312,
      "flos": 43142684088960.0,
      "grad_norm": 1.613244231897021,
      "language_loss": 0.698838,
      "learning_rate": 5.485633047362704e-07,
      "loss": 0.72077501,
      "num_input_tokens_seen": 137081200,
      "step": 6368,
      "time_per_iteration": 3.6256439685821533
    },
    {
      "auxiliary_loss_clip": 0.01172881,
      "auxiliary_loss_mlp": 0.01031953,
      "balance_loss_clip": 1.05112934,
      "balance_loss_mlp": 1.02426946,
      "epoch": 0.7658269704803703,
      "flos": 17311780840320.0,
      "grad_norm": 2.63038489548167,
      "language_loss": 0.78668237,
      "learning_rate": 5.480274886426341e-07,
      "loss": 0.80873072,
      "num_input_tokens_seen": 137097840,
      "step": 6369,
      "time_per_iteration": 2.4860212802886963
    },
    {
      "auxiliary_loss_clip": 0.01162325,
      "auxiliary_loss_mlp": 0.0101899,
      "balance_loss_clip": 1.04528618,
      "balance_loss_mlp": 1.01234651,
      "epoch": 0.7659472133710095,
      "flos": 12568160977920.0,
      "grad_norm": 2.485761585412026,
      "language_loss": 0.7811175,
      "learning_rate": 5.474918928120744e-07,
      "loss": 0.80293065,
      "num_input_tokens_seen": 137114335,
      "step": 6370,
      "time_per_iteration": 2.610320568084717
    },
    {
      "auxiliary_loss_clip": 0.01166067,
      "auxiliary_loss_mlp": 0.01020445,
      "balance_loss_clip": 1.04470384,
      "balance_loss_mlp": 1.013659,
      "epoch": 0.7660674562616485,
      "flos": 22707430335360.0,
      "grad_norm": 2.0703557000576365,
      "language_loss": 0.87416434,
      "learning_rate": 5.469565173258392e-07,
      "loss": 0.89602947,
      "num_input_tokens_seen": 137132850,
      "step": 6371,
      "time_per_iteration": 2.6054575443267822
    },
    {
      "auxiliary_loss_clip": 0.01170285,
      "auxiliary_loss_mlp": 0.01026037,
      "balance_loss_clip": 1.04725814,
      "balance_loss_mlp": 1.01788938,
      "epoch": 0.7661876991522876,
      "flos": 17056455989760.0,
      "grad_norm": 1.827880702494449,
      "language_loss": 0.6360718,
      "learning_rate": 5.464213622651454e-07,
      "loss": 0.65803504,
      "num_input_tokens_seen": 137150665,
      "step": 6372,
      "time_per_iteration": 2.672553777694702
    },
    {
      "auxiliary_loss_clip": 0.01179849,
      "auxiliary_loss_mlp": 0.01024201,
      "balance_loss_clip": 1.04338682,
      "balance_loss_mlp": 1.01655936,
      "epoch": 0.7663079420429267,
      "flos": 20084228092800.0,
      "grad_norm": 2.400575097036512,
      "language_loss": 0.84584153,
      "learning_rate": 5.458864277111753e-07,
      "loss": 0.86788201,
      "num_input_tokens_seen": 137168500,
      "step": 6373,
      "time_per_iteration": 2.8424408435821533
    },
    {
      "auxiliary_loss_clip": 0.01165189,
      "auxiliary_loss_mlp": 0.01206868,
      "balance_loss_clip": 1.04242039,
      "balance_loss_mlp": 1.0005995,
      "epoch": 0.7664281849335658,
      "flos": 12677473042560.0,
      "grad_norm": 3.2126280117638895,
      "language_loss": 0.69914806,
      "learning_rate": 5.453517137450769e-07,
      "loss": 0.72286862,
      "num_input_tokens_seen": 137185075,
      "step": 6374,
      "time_per_iteration": 2.7591216564178467
    },
    {
      "auxiliary_loss_clip": 0.01171138,
      "auxiliary_loss_mlp": 0.01023783,
      "balance_loss_clip": 1.04877138,
      "balance_loss_mlp": 1.01596892,
      "epoch": 0.7665484278242048,
      "flos": 22345271458560.0,
      "grad_norm": 1.8327496689102476,
      "language_loss": 0.75612104,
      "learning_rate": 5.448172204479684e-07,
      "loss": 0.77807033,
      "num_input_tokens_seen": 137204355,
      "step": 6375,
      "time_per_iteration": 2.729544162750244
    },
    {
      "auxiliary_loss_clip": 0.01163459,
      "auxiliary_loss_mlp": 0.01023457,
      "balance_loss_clip": 1.04568505,
      "balance_loss_mlp": 1.01602697,
      "epoch": 0.766668670714844,
      "flos": 23617909301760.0,
      "grad_norm": 2.7263758093065373,
      "language_loss": 0.74600875,
      "learning_rate": 5.442829479009294e-07,
      "loss": 0.76787794,
      "num_input_tokens_seen": 137223135,
      "step": 6376,
      "time_per_iteration": 2.6419568061828613
    },
    {
      "auxiliary_loss_clip": 0.011772,
      "auxiliary_loss_mlp": 0.01026949,
      "balance_loss_clip": 1.04704511,
      "balance_loss_mlp": 1.01918817,
      "epoch": 0.7667889136054831,
      "flos": 19427134642560.0,
      "grad_norm": 1.7669138689696993,
      "language_loss": 0.71428031,
      "learning_rate": 5.437488961850103e-07,
      "loss": 0.73632181,
      "num_input_tokens_seen": 137242935,
      "step": 6377,
      "time_per_iteration": 2.6433961391448975
    },
    {
      "auxiliary_loss_clip": 0.01173037,
      "auxiliary_loss_mlp": 0.01023288,
      "balance_loss_clip": 1.04110432,
      "balance_loss_mlp": 1.01632893,
      "epoch": 0.7669091564961221,
      "flos": 26866352609280.0,
      "grad_norm": 1.9206634763813983,
      "language_loss": 0.75775665,
      "learning_rate": 5.432150653812258e-07,
      "loss": 0.77971995,
      "num_input_tokens_seen": 137262970,
      "step": 6378,
      "time_per_iteration": 2.707862138748169
    },
    {
      "auxiliary_loss_clip": 0.011636,
      "auxiliary_loss_mlp": 0.01023105,
      "balance_loss_clip": 1.04474807,
      "balance_loss_mlp": 1.01561856,
      "epoch": 0.7670293993867613,
      "flos": 12385303816320.0,
      "grad_norm": 3.4473340888103934,
      "language_loss": 0.82682824,
      "learning_rate": 5.42681455570557e-07,
      "loss": 0.84869528,
      "num_input_tokens_seen": 137279500,
      "step": 6379,
      "time_per_iteration": 2.6348040103912354
    },
    {
      "auxiliary_loss_clip": 0.01162534,
      "auxiliary_loss_mlp": 0.01022136,
      "balance_loss_clip": 1.04492855,
      "balance_loss_mlp": 1.01489985,
      "epoch": 0.7671496422774003,
      "flos": 21762944167680.0,
      "grad_norm": 4.683424701558295,
      "language_loss": 0.65042573,
      "learning_rate": 5.42148066833954e-07,
      "loss": 0.67227244,
      "num_input_tokens_seen": 137298745,
      "step": 6380,
      "time_per_iteration": 2.640394687652588
    },
    {
      "auxiliary_loss_clip": 0.01163617,
      "auxiliary_loss_mlp": 0.01022748,
      "balance_loss_clip": 1.04626429,
      "balance_loss_mlp": 1.01553249,
      "epoch": 0.7672698851680394,
      "flos": 21069221823360.0,
      "grad_norm": 2.404077702785371,
      "language_loss": 0.75214696,
      "learning_rate": 5.416148992523289e-07,
      "loss": 0.77401054,
      "num_input_tokens_seen": 137317320,
      "step": 6381,
      "time_per_iteration": 2.649289846420288
    },
    {
      "auxiliary_loss_clip": 0.01180594,
      "auxiliary_loss_mlp": 0.01023969,
      "balance_loss_clip": 1.03946674,
      "balance_loss_mlp": 1.0170846,
      "epoch": 0.7673901280586786,
      "flos": 16976697840000.0,
      "grad_norm": 3.645006274893919,
      "language_loss": 0.78561771,
      "learning_rate": 5.410819529065644e-07,
      "loss": 0.80766344,
      "num_input_tokens_seen": 137335275,
      "step": 6382,
      "time_per_iteration": 3.6829652786254883
    },
    {
      "auxiliary_loss_clip": 0.01177278,
      "auxiliary_loss_mlp": 0.01024541,
      "balance_loss_clip": 1.04136801,
      "balance_loss_mlp": 1.01748121,
      "epoch": 0.7675103709493176,
      "flos": 29242669697280.0,
      "grad_norm": 1.9274788562784646,
      "language_loss": 0.6537419,
      "learning_rate": 5.405492278775079e-07,
      "loss": 0.67576003,
      "num_input_tokens_seen": 137355055,
      "step": 6383,
      "time_per_iteration": 2.718745708465576
    },
    {
      "auxiliary_loss_clip": 0.01172409,
      "auxiliary_loss_mlp": 0.01026386,
      "balance_loss_clip": 1.04302001,
      "balance_loss_mlp": 1.01846766,
      "epoch": 0.7676306138399567,
      "flos": 29023004073600.0,
      "grad_norm": 2.366898233015862,
      "language_loss": 0.80409372,
      "learning_rate": 5.400167242459732e-07,
      "loss": 0.82608169,
      "num_input_tokens_seen": 137374015,
      "step": 6384,
      "time_per_iteration": 2.712308645248413
    },
    {
      "auxiliary_loss_clip": 0.01167677,
      "auxiliary_loss_mlp": 0.01025652,
      "balance_loss_clip": 1.04580379,
      "balance_loss_mlp": 1.01845133,
      "epoch": 0.7677508567305958,
      "flos": 22565116650240.0,
      "grad_norm": 1.8972248997664303,
      "language_loss": 0.80733812,
      "learning_rate": 5.394844420927405e-07,
      "loss": 0.82927144,
      "num_input_tokens_seen": 137393625,
      "step": 6385,
      "time_per_iteration": 2.6307899951934814
    },
    {
      "auxiliary_loss_clip": 0.01165694,
      "auxiliary_loss_mlp": 0.01031905,
      "balance_loss_clip": 1.04636693,
      "balance_loss_mlp": 1.02445126,
      "epoch": 0.7678710996212349,
      "flos": 25411432222080.0,
      "grad_norm": 2.24807541445308,
      "language_loss": 0.73561817,
      "learning_rate": 5.389523814985562e-07,
      "loss": 0.75759417,
      "num_input_tokens_seen": 137413045,
      "step": 6386,
      "time_per_iteration": 2.7767369747161865
    },
    {
      "auxiliary_loss_clip": 0.01174419,
      "auxiliary_loss_mlp": 0.0102089,
      "balance_loss_clip": 1.04074502,
      "balance_loss_mlp": 1.01338339,
      "epoch": 0.767991342511874,
      "flos": 26756825063040.0,
      "grad_norm": 2.96522996561583,
      "language_loss": 0.76253033,
      "learning_rate": 5.384205425441344e-07,
      "loss": 0.78448343,
      "num_input_tokens_seen": 137433955,
      "step": 6387,
      "time_per_iteration": 3.008216381072998
    },
    {
      "auxiliary_loss_clip": 0.01171983,
      "auxiliary_loss_mlp": 0.01021071,
      "balance_loss_clip": 1.04197145,
      "balance_loss_mlp": 1.01394844,
      "epoch": 0.7681115854025131,
      "flos": 26359509749760.0,
      "grad_norm": 1.684105485419129,
      "language_loss": 0.84264505,
      "learning_rate": 5.378889253101537e-07,
      "loss": 0.86457562,
      "num_input_tokens_seen": 137454510,
      "step": 6388,
      "time_per_iteration": 3.6785523891448975
    },
    {
      "auxiliary_loss_clip": 0.01169452,
      "auxiliary_loss_mlp": 0.01021257,
      "balance_loss_clip": 1.04513168,
      "balance_loss_mlp": 1.01407731,
      "epoch": 0.7682318282931522,
      "flos": 23257043314560.0,
      "grad_norm": 1.6845395884379066,
      "language_loss": 0.80944902,
      "learning_rate": 5.373575298772617e-07,
      "loss": 0.83135611,
      "num_input_tokens_seen": 137473630,
      "step": 6389,
      "time_per_iteration": 2.6720523834228516
    },
    {
      "auxiliary_loss_clip": 0.01073191,
      "auxiliary_loss_mlp": 0.01001038,
      "balance_loss_clip": 1.01161742,
      "balance_loss_mlp": 1.00025702,
      "epoch": 0.7683520711837912,
      "flos": 70072457137920.0,
      "grad_norm": 0.7532520063333386,
      "language_loss": 0.61314118,
      "learning_rate": 5.368263563260689e-07,
      "loss": 0.63388348,
      "num_input_tokens_seen": 137538765,
      "step": 6390,
      "time_per_iteration": 4.1214776039123535
    },
    {
      "auxiliary_loss_clip": 0.01170143,
      "auxiliary_loss_mlp": 0.0101927,
      "balance_loss_clip": 1.04532933,
      "balance_loss_mlp": 1.0120461,
      "epoch": 0.7684723140744304,
      "flos": 18624890332800.0,
      "grad_norm": 6.1181883604889995,
      "language_loss": 0.64285851,
      "learning_rate": 5.362954047371537e-07,
      "loss": 0.6647526,
      "num_input_tokens_seen": 137557875,
      "step": 6391,
      "time_per_iteration": 2.6201484203338623
    },
    {
      "auxiliary_loss_clip": 0.01175947,
      "auxiliary_loss_mlp": 0.01029912,
      "balance_loss_clip": 1.04764175,
      "balance_loss_mlp": 1.022259,
      "epoch": 0.7685925569650695,
      "flos": 27452989532160.0,
      "grad_norm": 2.1501389978602456,
      "language_loss": 0.72508943,
      "learning_rate": 5.357646751910627e-07,
      "loss": 0.7471481,
      "num_input_tokens_seen": 137579055,
      "step": 6392,
      "time_per_iteration": 2.708760976791382
    },
    {
      "auxiliary_loss_clip": 0.01171556,
      "auxiliary_loss_mlp": 0.0102726,
      "balance_loss_clip": 1.04432893,
      "balance_loss_mlp": 1.01942205,
      "epoch": 0.7687127998557085,
      "flos": 24535714642560.0,
      "grad_norm": 3.305804817885678,
      "language_loss": 0.79718399,
      "learning_rate": 5.352341677683061e-07,
      "loss": 0.81917226,
      "num_input_tokens_seen": 137600355,
      "step": 6393,
      "time_per_iteration": 2.6696770191192627
    },
    {
      "auxiliary_loss_clip": 0.01182562,
      "auxiliary_loss_mlp": 0.0102853,
      "balance_loss_clip": 1.04540646,
      "balance_loss_mlp": 1.02091825,
      "epoch": 0.7688330427463477,
      "flos": 25155963717120.0,
      "grad_norm": 2.2740765151089666,
      "language_loss": 0.79055011,
      "learning_rate": 5.347038825493617e-07,
      "loss": 0.81266105,
      "num_input_tokens_seen": 137621885,
      "step": 6394,
      "time_per_iteration": 3.612889289855957
    },
    {
      "auxiliary_loss_clip": 0.01170175,
      "auxiliary_loss_mlp": 0.01026762,
      "balance_loss_clip": 1.04720688,
      "balance_loss_mlp": 1.01969278,
      "epoch": 0.7689532856369867,
      "flos": 21211284113280.0,
      "grad_norm": 2.154970855611394,
      "language_loss": 0.68615973,
      "learning_rate": 5.341738196146732e-07,
      "loss": 0.70812905,
      "num_input_tokens_seen": 137640230,
      "step": 6395,
      "time_per_iteration": 2.6280319690704346
    },
    {
      "auxiliary_loss_clip": 0.01165354,
      "auxiliary_loss_mlp": 0.01022116,
      "balance_loss_clip": 1.0434953,
      "balance_loss_mlp": 1.01460922,
      "epoch": 0.7690735285276258,
      "flos": 25119083427840.0,
      "grad_norm": 4.597405869369469,
      "language_loss": 0.73576957,
      "learning_rate": 5.336439790446503e-07,
      "loss": 0.7576443,
      "num_input_tokens_seen": 137659330,
      "step": 6396,
      "time_per_iteration": 2.8280391693115234
    },
    {
      "auxiliary_loss_clip": 0.01168742,
      "auxiliary_loss_mlp": 0.01027877,
      "balance_loss_clip": 1.03898096,
      "balance_loss_mlp": 1.01964569,
      "epoch": 0.769193771418265,
      "flos": 54744020640000.0,
      "grad_norm": 1.9134806559346291,
      "language_loss": 0.62999249,
      "learning_rate": 5.331143609196711e-07,
      "loss": 0.6519587,
      "num_input_tokens_seen": 137683145,
      "step": 6397,
      "time_per_iteration": 3.2176432609558105
    },
    {
      "auxiliary_loss_clip": 0.01167089,
      "auxiliary_loss_mlp": 0.01024339,
      "balance_loss_clip": 1.04686213,
      "balance_loss_mlp": 1.01688814,
      "epoch": 0.769314014308904,
      "flos": 37341890115840.0,
      "grad_norm": 2.302673333075366,
      "language_loss": 0.77109444,
      "learning_rate": 5.325849653200758e-07,
      "loss": 0.79300869,
      "num_input_tokens_seen": 137707095,
      "step": 6398,
      "time_per_iteration": 2.7735395431518555
    },
    {
      "auxiliary_loss_clip": 0.01166144,
      "auxiliary_loss_mlp": 0.01022378,
      "balance_loss_clip": 1.04730272,
      "balance_loss_mlp": 1.01498055,
      "epoch": 0.7694342571995431,
      "flos": 20631686256000.0,
      "grad_norm": 1.7752883762039091,
      "language_loss": 0.76539016,
      "learning_rate": 5.32055792326175e-07,
      "loss": 0.78727537,
      "num_input_tokens_seen": 137725520,
      "step": 6399,
      "time_per_iteration": 2.6546895503997803
    },
    {
      "auxiliary_loss_clip": 0.01175766,
      "auxiliary_loss_mlp": 0.01020788,
      "balance_loss_clip": 1.04756427,
      "balance_loss_mlp": 1.01336741,
      "epoch": 0.7695545000901821,
      "flos": 24207706621440.0,
      "grad_norm": 2.5901618617597775,
      "language_loss": 0.72847176,
      "learning_rate": 5.315268420182437e-07,
      "loss": 0.75043732,
      "num_input_tokens_seen": 137744195,
      "step": 6400,
      "time_per_iteration": 2.6772091388702393
    },
    {
      "auxiliary_loss_clip": 0.01179544,
      "auxiliary_loss_mlp": 0.0120751,
      "balance_loss_clip": 1.04353452,
      "balance_loss_mlp": 1.00069618,
      "epoch": 0.7696747429808213,
      "flos": 28001273708160.0,
      "grad_norm": 2.2657721761330616,
      "language_loss": 0.76222974,
      "learning_rate": 5.309981144765221e-07,
      "loss": 0.78610027,
      "num_input_tokens_seen": 137764340,
      "step": 6401,
      "time_per_iteration": 2.7923176288604736
    },
    {
      "auxiliary_loss_clip": 0.01181504,
      "auxiliary_loss_mlp": 0.01022718,
      "balance_loss_clip": 1.0404458,
      "balance_loss_mlp": 1.01573551,
      "epoch": 0.7697949858714603,
      "flos": 11509550323200.0,
      "grad_norm": 2.5391171494209477,
      "language_loss": 0.75322199,
      "learning_rate": 5.304696097812196e-07,
      "loss": 0.7752642,
      "num_input_tokens_seen": 137780940,
      "step": 6402,
      "time_per_iteration": 2.6688120365142822
    },
    {
      "auxiliary_loss_clip": 0.01170298,
      "auxiliary_loss_mlp": 0.01025113,
      "balance_loss_clip": 1.04358196,
      "balance_loss_mlp": 1.01707828,
      "epoch": 0.7699152287620994,
      "flos": 26688271956480.0,
      "grad_norm": 4.166068092657219,
      "language_loss": 0.60338968,
      "learning_rate": 5.299413280125078e-07,
      "loss": 0.62534386,
      "num_input_tokens_seen": 137799250,
      "step": 6403,
      "time_per_iteration": 2.676090717315674
    },
    {
      "auxiliary_loss_clip": 0.01171978,
      "auxiliary_loss_mlp": 0.01030801,
      "balance_loss_clip": 1.04420221,
      "balance_loss_mlp": 1.02366662,
      "epoch": 0.7700354716527386,
      "flos": 16544944362240.0,
      "grad_norm": 1.951378734296583,
      "language_loss": 0.73015714,
      "learning_rate": 5.294132692505284e-07,
      "loss": 0.75218493,
      "num_input_tokens_seen": 137817660,
      "step": 6404,
      "time_per_iteration": 2.6236860752105713
    },
    {
      "auxiliary_loss_clip": 0.01165875,
      "auxiliary_loss_mlp": 0.01025749,
      "balance_loss_clip": 1.03869748,
      "balance_loss_mlp": 1.01803327,
      "epoch": 0.7701557145433776,
      "flos": 19242733196160.0,
      "grad_norm": 2.3067140772146772,
      "language_loss": 0.79475641,
      "learning_rate": 5.288854335753861e-07,
      "loss": 0.81667262,
      "num_input_tokens_seen": 137835920,
      "step": 6405,
      "time_per_iteration": 2.7305610179901123
    },
    {
      "auxiliary_loss_clip": 0.01168661,
      "auxiliary_loss_mlp": 0.01021782,
      "balance_loss_clip": 1.04424059,
      "balance_loss_mlp": 1.01443005,
      "epoch": 0.7702759574340167,
      "flos": 31685744211840.0,
      "grad_norm": 2.8160262193277217,
      "language_loss": 0.75712144,
      "learning_rate": 5.283578210671551e-07,
      "loss": 0.77902591,
      "num_input_tokens_seen": 137858160,
      "step": 6406,
      "time_per_iteration": 2.6918530464172363
    },
    {
      "auxiliary_loss_clip": 0.0117521,
      "auxiliary_loss_mlp": 0.01023229,
      "balance_loss_clip": 1.04549885,
      "balance_loss_mlp": 1.01552761,
      "epoch": 0.7703962003246558,
      "flos": 16800089644800.0,
      "grad_norm": 2.350432616573736,
      "language_loss": 0.7668463,
      "learning_rate": 5.278304318058719e-07,
      "loss": 0.78883064,
      "num_input_tokens_seen": 137876015,
      "step": 6407,
      "time_per_iteration": 2.659578323364258
    },
    {
      "auxiliary_loss_clip": 0.01177127,
      "auxiliary_loss_mlp": 0.01025399,
      "balance_loss_clip": 1.03930235,
      "balance_loss_mlp": 1.01775789,
      "epoch": 0.7705164432152949,
      "flos": 35736072693120.0,
      "grad_norm": 1.932279725322013,
      "language_loss": 0.79056495,
      "learning_rate": 5.273032658715411e-07,
      "loss": 0.81259024,
      "num_input_tokens_seen": 137898825,
      "step": 6408,
      "time_per_iteration": 3.8258681297302246
    },
    {
      "auxiliary_loss_clip": 0.01171592,
      "auxiliary_loss_mlp": 0.01021605,
      "balance_loss_clip": 1.03960991,
      "balance_loss_mlp": 1.014184,
      "epoch": 0.7706366861059339,
      "flos": 23365960329600.0,
      "grad_norm": 2.5586836655147445,
      "language_loss": 0.7681576,
      "learning_rate": 5.267763233441347e-07,
      "loss": 0.79008961,
      "num_input_tokens_seen": 137919455,
      "step": 6409,
      "time_per_iteration": 2.7991225719451904
    },
    {
      "auxiliary_loss_clip": 0.01172506,
      "auxiliary_loss_mlp": 0.01024248,
      "balance_loss_clip": 1.04658508,
      "balance_loss_mlp": 1.01622486,
      "epoch": 0.7707569289965731,
      "flos": 22929897219840.0,
      "grad_norm": 3.36526514250432,
      "language_loss": 0.69707721,
      "learning_rate": 5.26249604303588e-07,
      "loss": 0.71904474,
      "num_input_tokens_seen": 137937960,
      "step": 6410,
      "time_per_iteration": 2.6716063022613525
    },
    {
      "auxiliary_loss_clip": 0.01166884,
      "auxiliary_loss_mlp": 0.01024037,
      "balance_loss_clip": 1.04841483,
      "balance_loss_mlp": 1.01680136,
      "epoch": 0.7708771718872122,
      "flos": 17420661941760.0,
      "grad_norm": 2.3320675224103824,
      "language_loss": 0.78425086,
      "learning_rate": 5.257231088298057e-07,
      "loss": 0.80616003,
      "num_input_tokens_seen": 137956370,
      "step": 6411,
      "time_per_iteration": 2.5534746646881104
    },
    {
      "auxiliary_loss_clip": 0.01085411,
      "auxiliary_loss_mlp": 0.01002469,
      "balance_loss_clip": 1.01537871,
      "balance_loss_mlp": 1.00169969,
      "epoch": 0.7709974147778512,
      "flos": 72241316248320.0,
      "grad_norm": 0.8028477298556367,
      "language_loss": 0.53909123,
      "learning_rate": 5.25196837002655e-07,
      "loss": 0.55997002,
      "num_input_tokens_seen": 138016080,
      "step": 6412,
      "time_per_iteration": 3.2458579540252686
    },
    {
      "auxiliary_loss_clip": 0.01169009,
      "auxiliary_loss_mlp": 0.01033978,
      "balance_loss_clip": 1.04363632,
      "balance_loss_mlp": 1.02634883,
      "epoch": 0.7711176576684904,
      "flos": 39859694876160.0,
      "grad_norm": 1.910526078656238,
      "language_loss": 0.68300706,
      "learning_rate": 5.24670788901971e-07,
      "loss": 0.70503688,
      "num_input_tokens_seen": 138039170,
      "step": 6413,
      "time_per_iteration": 2.822845697402954
    },
    {
      "auxiliary_loss_clip": 0.01172516,
      "auxiliary_loss_mlp": 0.01023822,
      "balance_loss_clip": 1.0461787,
      "balance_loss_mlp": 1.01547098,
      "epoch": 0.7712379005591294,
      "flos": 36976391274240.0,
      "grad_norm": 2.6608623936585603,
      "language_loss": 0.68768048,
      "learning_rate": 5.241449646075557e-07,
      "loss": 0.70964384,
      "num_input_tokens_seen": 138062395,
      "step": 6414,
      "time_per_iteration": 2.8186144828796387
    },
    {
      "auxiliary_loss_clip": 0.01179402,
      "auxiliary_loss_mlp": 0.01026923,
      "balance_loss_clip": 1.04845142,
      "balance_loss_mlp": 1.01941299,
      "epoch": 0.7713581434497685,
      "flos": 22776773541120.0,
      "grad_norm": 2.2667674124659416,
      "language_loss": 0.72482538,
      "learning_rate": 5.236193641991762e-07,
      "loss": 0.74688864,
      "num_input_tokens_seen": 138080325,
      "step": 6415,
      "time_per_iteration": 3.5636367797851562
    },
    {
      "auxiliary_loss_clip": 0.01170963,
      "auxiliary_loss_mlp": 0.01023708,
      "balance_loss_clip": 1.0446502,
      "balance_loss_mlp": 1.01642728,
      "epoch": 0.7714783863404077,
      "flos": 24097460803200.0,
      "grad_norm": 3.2986473704756683,
      "language_loss": 0.70194721,
      "learning_rate": 5.23093987756565e-07,
      "loss": 0.72389394,
      "num_input_tokens_seen": 138099020,
      "step": 6416,
      "time_per_iteration": 3.5791027545928955
    },
    {
      "auxiliary_loss_clip": 0.01178993,
      "auxiliary_loss_mlp": 0.01023218,
      "balance_loss_clip": 1.04141498,
      "balance_loss_mlp": 1.01554704,
      "epoch": 0.7715986292310467,
      "flos": 21063655215360.0,
      "grad_norm": 2.3103717848763843,
      "language_loss": 0.75418222,
      "learning_rate": 5.225688353594217e-07,
      "loss": 0.77620435,
      "num_input_tokens_seen": 138118650,
      "step": 6417,
      "time_per_iteration": 2.8053297996520996
    },
    {
      "auxiliary_loss_clip": 0.01176157,
      "auxiliary_loss_mlp": 0.012072,
      "balance_loss_clip": 1.04596388,
      "balance_loss_mlp": 1.00083387,
      "epoch": 0.7717188721216858,
      "flos": 20594877793920.0,
      "grad_norm": 2.3593042030926954,
      "language_loss": 0.77496022,
      "learning_rate": 5.220439070874108e-07,
      "loss": 0.79879373,
      "num_input_tokens_seen": 138137890,
      "step": 6418,
      "time_per_iteration": 2.6997087001800537
    },
    {
      "auxiliary_loss_clip": 0.01168923,
      "auxiliary_loss_mlp": 0.01027564,
      "balance_loss_clip": 1.04691184,
      "balance_loss_mlp": 1.02021146,
      "epoch": 0.7718391150123249,
      "flos": 26250951870720.0,
      "grad_norm": 1.9282784192944669,
      "language_loss": 0.71491086,
      "learning_rate": 5.215192030201652e-07,
      "loss": 0.73687571,
      "num_input_tokens_seen": 138158880,
      "step": 6419,
      "time_per_iteration": 2.761337995529175
    },
    {
      "auxiliary_loss_clip": 0.01160122,
      "auxiliary_loss_mlp": 0.01023986,
      "balance_loss_clip": 1.03837204,
      "balance_loss_mlp": 1.01638675,
      "epoch": 0.771959357902964,
      "flos": 22049762267520.0,
      "grad_norm": 2.621745915758368,
      "language_loss": 0.86355627,
      "learning_rate": 5.209947232372798e-07,
      "loss": 0.88539743,
      "num_input_tokens_seen": 138176370,
      "step": 6420,
      "time_per_iteration": 2.7223002910614014
    },
    {
      "auxiliary_loss_clip": 0.01172137,
      "auxiliary_loss_mlp": 0.01207545,
      "balance_loss_clip": 1.04459357,
      "balance_loss_mlp": 1.00073576,
      "epoch": 0.772079600793603,
      "flos": 30446000248320.0,
      "grad_norm": 1.735747586173401,
      "language_loss": 0.81174195,
      "learning_rate": 5.204704678183196e-07,
      "loss": 0.83553886,
      "num_input_tokens_seen": 138195105,
      "step": 6421,
      "time_per_iteration": 3.560255289077759
    },
    {
      "auxiliary_loss_clip": 0.01167266,
      "auxiliary_loss_mlp": 0.010238,
      "balance_loss_clip": 1.04780293,
      "balance_loss_mlp": 1.0158844,
      "epoch": 0.7721998436842422,
      "flos": 12969857750400.0,
      "grad_norm": 2.0462529427667584,
      "language_loss": 0.85516256,
      "learning_rate": 5.19946436842813e-07,
      "loss": 0.87707329,
      "num_input_tokens_seen": 138212235,
      "step": 6422,
      "time_per_iteration": 2.6231045722961426
    },
    {
      "auxiliary_loss_clip": 0.0117341,
      "auxiliary_loss_mlp": 0.01019083,
      "balance_loss_clip": 1.04435849,
      "balance_loss_mlp": 1.01199579,
      "epoch": 0.7723200865748813,
      "flos": 32635509678720.0,
      "grad_norm": 2.649228718497482,
      "language_loss": 0.68538195,
      "learning_rate": 5.194226303902546e-07,
      "loss": 0.70730686,
      "num_input_tokens_seen": 138231970,
      "step": 6423,
      "time_per_iteration": 2.834278106689453
    },
    {
      "auxiliary_loss_clip": 0.01169498,
      "auxiliary_loss_mlp": 0.01025014,
      "balance_loss_clip": 1.04326057,
      "balance_loss_mlp": 1.0175097,
      "epoch": 0.7724403294655203,
      "flos": 21105707063040.0,
      "grad_norm": 2.37135230560081,
      "language_loss": 0.70905977,
      "learning_rate": 5.188990485401072e-07,
      "loss": 0.73100483,
      "num_input_tokens_seen": 138251175,
      "step": 6424,
      "time_per_iteration": 2.6532812118530273
    },
    {
      "auxiliary_loss_clip": 0.01171441,
      "auxiliary_loss_mlp": 0.01022218,
      "balance_loss_clip": 1.04720759,
      "balance_loss_mlp": 1.01505351,
      "epoch": 0.7725605723561595,
      "flos": 22090736707200.0,
      "grad_norm": 1.855474880181949,
      "language_loss": 0.86361551,
      "learning_rate": 5.183756913717954e-07,
      "loss": 0.88555211,
      "num_input_tokens_seen": 138270950,
      "step": 6425,
      "time_per_iteration": 2.729114055633545
    },
    {
      "auxiliary_loss_clip": 0.01165345,
      "auxiliary_loss_mlp": 0.01026391,
      "balance_loss_clip": 1.0434581,
      "balance_loss_mlp": 1.01911879,
      "epoch": 0.7726808152467985,
      "flos": 34495610457600.0,
      "grad_norm": 2.09548764302978,
      "language_loss": 0.73714566,
      "learning_rate": 5.178525589647136e-07,
      "loss": 0.75906307,
      "num_input_tokens_seen": 138292590,
      "step": 6426,
      "time_per_iteration": 2.7456490993499756
    },
    {
      "auxiliary_loss_clip": 0.01175818,
      "auxiliary_loss_mlp": 0.01021815,
      "balance_loss_clip": 1.04378438,
      "balance_loss_mlp": 1.01459122,
      "epoch": 0.7728010581374376,
      "flos": 22306344094080.0,
      "grad_norm": 1.7667657588518104,
      "language_loss": 0.78752452,
      "learning_rate": 5.173296513982197e-07,
      "loss": 0.80950087,
      "num_input_tokens_seen": 138311115,
      "step": 6427,
      "time_per_iteration": 2.6793529987335205
    },
    {
      "auxiliary_loss_clip": 0.01183803,
      "auxiliary_loss_mlp": 0.01027445,
      "balance_loss_clip": 1.04550135,
      "balance_loss_mlp": 1.01882958,
      "epoch": 0.7729213010280768,
      "flos": 27126453968640.0,
      "grad_norm": 7.200984640434708,
      "language_loss": 0.65049207,
      "learning_rate": 5.168069687516398e-07,
      "loss": 0.67260456,
      "num_input_tokens_seen": 138330885,
      "step": 6428,
      "time_per_iteration": 2.723310947418213
    },
    {
      "auxiliary_loss_clip": 0.01173137,
      "auxiliary_loss_mlp": 0.01021362,
      "balance_loss_clip": 1.0469681,
      "balance_loss_mlp": 1.01370335,
      "epoch": 0.7730415439187158,
      "flos": 18150223080960.0,
      "grad_norm": 2.130252337136251,
      "language_loss": 0.72019792,
      "learning_rate": 5.16284511104263e-07,
      "loss": 0.74214286,
      "num_input_tokens_seen": 138350020,
      "step": 6429,
      "time_per_iteration": 2.724260091781616
    },
    {
      "auxiliary_loss_clip": 0.01169876,
      "auxiliary_loss_mlp": 0.01027862,
      "balance_loss_clip": 1.04369593,
      "balance_loss_mlp": 1.02016687,
      "epoch": 0.7731617868093549,
      "flos": 11947480940160.0,
      "grad_norm": 2.7641330854034325,
      "language_loss": 0.80582601,
      "learning_rate": 5.157622785353457e-07,
      "loss": 0.82780337,
      "num_input_tokens_seen": 138368135,
      "step": 6430,
      "time_per_iteration": 2.696124792098999
    },
    {
      "auxiliary_loss_clip": 0.01072757,
      "auxiliary_loss_mlp": 0.00998944,
      "balance_loss_clip": 1.01183486,
      "balance_loss_mlp": 0.99816287,
      "epoch": 0.7732820296999939,
      "flos": 64201027069440.0,
      "grad_norm": 0.6516754522579526,
      "language_loss": 0.60360086,
      "learning_rate": 5.152402711241113e-07,
      "loss": 0.62431788,
      "num_input_tokens_seen": 138436040,
      "step": 6431,
      "time_per_iteration": 3.3888003826141357
    },
    {
      "auxiliary_loss_clip": 0.01170361,
      "auxiliary_loss_mlp": 0.01020468,
      "balance_loss_clip": 1.03988683,
      "balance_loss_mlp": 1.01295829,
      "epoch": 0.7734022725906331,
      "flos": 25302191984640.0,
      "grad_norm": 1.5861138465749562,
      "language_loss": 0.828026,
      "learning_rate": 5.147184889497465e-07,
      "loss": 0.84993428,
      "num_input_tokens_seen": 138455510,
      "step": 6432,
      "time_per_iteration": 2.738004207611084
    },
    {
      "auxiliary_loss_clip": 0.01166015,
      "auxiliary_loss_mlp": 0.01022646,
      "balance_loss_clip": 1.04122663,
      "balance_loss_mlp": 1.01492143,
      "epoch": 0.7735225154812722,
      "flos": 17347440067200.0,
      "grad_norm": 2.243192734115847,
      "language_loss": 0.80141616,
      "learning_rate": 5.141969320914072e-07,
      "loss": 0.82330281,
      "num_input_tokens_seen": 138473015,
      "step": 6433,
      "time_per_iteration": 2.6252222061157227
    },
    {
      "auxiliary_loss_clip": 0.01170941,
      "auxiliary_loss_mlp": 0.01025463,
      "balance_loss_clip": 1.04754758,
      "balance_loss_mlp": 1.0176543,
      "epoch": 0.7736427583719112,
      "flos": 32630086725120.0,
      "grad_norm": 3.2542806225641416,
      "language_loss": 0.62483156,
      "learning_rate": 5.136756006282113e-07,
      "loss": 0.64679551,
      "num_input_tokens_seen": 138491680,
      "step": 6434,
      "time_per_iteration": 2.6776182651519775
    },
    {
      "auxiliary_loss_clip": 0.01169155,
      "auxiliary_loss_mlp": 0.01022979,
      "balance_loss_clip": 1.04873109,
      "balance_loss_mlp": 1.01555276,
      "epoch": 0.7737630012625504,
      "flos": 19860073269120.0,
      "grad_norm": 2.652719512719211,
      "language_loss": 0.85170233,
      "learning_rate": 5.131544946392446e-07,
      "loss": 0.87362361,
      "num_input_tokens_seen": 138506960,
      "step": 6435,
      "time_per_iteration": 3.540477752685547
    },
    {
      "auxiliary_loss_clip": 0.01172606,
      "auxiliary_loss_mlp": 0.01024544,
      "balance_loss_clip": 1.04807246,
      "balance_loss_mlp": 1.01687634,
      "epoch": 0.7738832441531894,
      "flos": 36022639397760.0,
      "grad_norm": 2.248083842354211,
      "language_loss": 0.64149368,
      "learning_rate": 5.126336142035592e-07,
      "loss": 0.66346526,
      "num_input_tokens_seen": 138526995,
      "step": 6436,
      "time_per_iteration": 2.8249104022979736
    },
    {
      "auxiliary_loss_clip": 0.01173395,
      "auxiliary_loss_mlp": 0.0102135,
      "balance_loss_clip": 1.04472065,
      "balance_loss_mlp": 1.01352429,
      "epoch": 0.7740034870438285,
      "flos": 13405274415360.0,
      "grad_norm": 2.2069761479652747,
      "language_loss": 0.7202518,
      "learning_rate": 5.121129594001721e-07,
      "loss": 0.7421993,
      "num_input_tokens_seen": 138541260,
      "step": 6437,
      "time_per_iteration": 2.688157081604004
    },
    {
      "auxiliary_loss_clip": 0.01170039,
      "auxiliary_loss_mlp": 0.01023266,
      "balance_loss_clip": 1.04687226,
      "balance_loss_mlp": 1.0153476,
      "epoch": 0.7741237299344677,
      "flos": 22086714384000.0,
      "grad_norm": 1.965586060875561,
      "language_loss": 0.81120014,
      "learning_rate": 5.115925303080661e-07,
      "loss": 0.83313322,
      "num_input_tokens_seen": 138560970,
      "step": 6438,
      "time_per_iteration": 2.590975046157837
    },
    {
      "auxiliary_loss_clip": 0.0117117,
      "auxiliary_loss_mlp": 0.01027637,
      "balance_loss_clip": 1.04427481,
      "balance_loss_mlp": 1.02059162,
      "epoch": 0.7742439728251067,
      "flos": 19864777950720.0,
      "grad_norm": 2.2906012421573747,
      "language_loss": 0.79319555,
      "learning_rate": 5.110723270061899e-07,
      "loss": 0.81518364,
      "num_input_tokens_seen": 138577460,
      "step": 6439,
      "time_per_iteration": 2.6035592555999756
    },
    {
      "auxiliary_loss_clip": 0.0116401,
      "auxiliary_loss_mlp": 0.0102256,
      "balance_loss_clip": 1.04612613,
      "balance_loss_mlp": 1.01581299,
      "epoch": 0.7743642157157458,
      "flos": 16690167048960.0,
      "grad_norm": 1.8786936022817229,
      "language_loss": 0.79937816,
      "learning_rate": 5.105523495734572e-07,
      "loss": 0.82124388,
      "num_input_tokens_seen": 138594860,
      "step": 6440,
      "time_per_iteration": 2.59613299369812
    },
    {
      "auxiliary_loss_clip": 0.0116707,
      "auxiliary_loss_mlp": 0.01023375,
      "balance_loss_clip": 1.04650474,
      "balance_loss_mlp": 1.01523602,
      "epoch": 0.7744844586063849,
      "flos": 20304360593280.0,
      "grad_norm": 1.617242235405505,
      "language_loss": 0.75452065,
      "learning_rate": 5.100325980887499e-07,
      "loss": 0.77642512,
      "num_input_tokens_seen": 138614785,
      "step": 6441,
      "time_per_iteration": 2.5368528366088867
    },
    {
      "auxiliary_loss_clip": 0.01181755,
      "auxiliary_loss_mlp": 0.01022543,
      "balance_loss_clip": 1.04683149,
      "balance_loss_mlp": 1.01508617,
      "epoch": 0.774604701497024,
      "flos": 22966705681920.0,
      "grad_norm": 1.8757952825742177,
      "language_loss": 0.83228934,
      "learning_rate": 5.095130726309116e-07,
      "loss": 0.85433233,
      "num_input_tokens_seen": 138634960,
      "step": 6442,
      "time_per_iteration": 3.7764854431152344
    },
    {
      "auxiliary_loss_clip": 0.01065305,
      "auxiliary_loss_mlp": 0.00999197,
      "balance_loss_clip": 1.01220632,
      "balance_loss_mlp": 0.99839824,
      "epoch": 0.774724944387663,
      "flos": 60288523073280.0,
      "grad_norm": 0.7924942116012972,
      "language_loss": 0.58915341,
      "learning_rate": 5.089937732787559e-07,
      "loss": 0.60979843,
      "num_input_tokens_seen": 138699520,
      "step": 6443,
      "time_per_iteration": 4.160782337188721
    },
    {
      "auxiliary_loss_clip": 0.01173492,
      "auxiliary_loss_mlp": 0.01024605,
      "balance_loss_clip": 1.04232526,
      "balance_loss_mlp": 1.01686203,
      "epoch": 0.7748451872783022,
      "flos": 26761026954240.0,
      "grad_norm": 6.718069436781491,
      "language_loss": 0.66607058,
      "learning_rate": 5.084747001110592e-07,
      "loss": 0.68805152,
      "num_input_tokens_seen": 138719145,
      "step": 6444,
      "time_per_iteration": 2.6808135509490967
    },
    {
      "auxiliary_loss_clip": 0.01164508,
      "auxiliary_loss_mlp": 0.01207319,
      "balance_loss_clip": 1.04766202,
      "balance_loss_mlp": 1.00068378,
      "epoch": 0.7749654301689413,
      "flos": 30338627518080.0,
      "grad_norm": 1.649699015910513,
      "language_loss": 0.70294487,
      "learning_rate": 5.07955853206564e-07,
      "loss": 0.72666317,
      "num_input_tokens_seen": 138743850,
      "step": 6445,
      "time_per_iteration": 2.6740036010742188
    },
    {
      "auxiliary_loss_clip": 0.01173065,
      "auxiliary_loss_mlp": 0.01026901,
      "balance_loss_clip": 1.04619348,
      "balance_loss_mlp": 1.01934278,
      "epoch": 0.7750856730595803,
      "flos": 43179851687040.0,
      "grad_norm": 1.6548743035142648,
      "language_loss": 0.7126857,
      "learning_rate": 5.074372326439807e-07,
      "loss": 0.73468542,
      "num_input_tokens_seen": 138766860,
      "step": 6446,
      "time_per_iteration": 2.823411464691162
    },
    {
      "auxiliary_loss_clip": 0.01175078,
      "auxiliary_loss_mlp": 0.01025899,
      "balance_loss_clip": 1.0423305,
      "balance_loss_mlp": 1.01836514,
      "epoch": 0.7752059159502195,
      "flos": 17640040256640.0,
      "grad_norm": 2.153778451571429,
      "language_loss": 0.73526776,
      "learning_rate": 5.069188385019814e-07,
      "loss": 0.75727749,
      "num_input_tokens_seen": 138784560,
      "step": 6447,
      "time_per_iteration": 3.5522258281707764
    },
    {
      "auxiliary_loss_clip": 0.01185073,
      "auxiliary_loss_mlp": 0.01021692,
      "balance_loss_clip": 1.04215503,
      "balance_loss_mlp": 1.01413107,
      "epoch": 0.7753261588408585,
      "flos": 12677688524160.0,
      "grad_norm": 6.569946599831892,
      "language_loss": 0.61550355,
      "learning_rate": 5.064006708592077e-07,
      "loss": 0.63757122,
      "num_input_tokens_seen": 138800805,
      "step": 6448,
      "time_per_iteration": 2.6695683002471924
    },
    {
      "auxiliary_loss_clip": 0.01165318,
      "auxiliary_loss_mlp": 0.01022325,
      "balance_loss_clip": 1.04513741,
      "balance_loss_mlp": 1.01525927,
      "epoch": 0.7754464017314976,
      "flos": 16690741666560.0,
      "grad_norm": 2.8303541235767273,
      "language_loss": 0.755862,
      "learning_rate": 5.058827297942641e-07,
      "loss": 0.77773851,
      "num_input_tokens_seen": 138815910,
      "step": 6449,
      "time_per_iteration": 2.5907773971557617
    },
    {
      "auxiliary_loss_clip": 0.01179163,
      "auxiliary_loss_mlp": 0.01024317,
      "balance_loss_clip": 1.04549098,
      "balance_loss_mlp": 1.01686943,
      "epoch": 0.7755666446221368,
      "flos": 19718944732800.0,
      "grad_norm": 1.8851657500702381,
      "language_loss": 0.7542609,
      "learning_rate": 5.053650153857237e-07,
      "loss": 0.77629572,
      "num_input_tokens_seen": 138834920,
      "step": 6450,
      "time_per_iteration": 2.6627488136291504
    },
    {
      "auxiliary_loss_clip": 0.01168586,
      "auxiliary_loss_mlp": 0.01027295,
      "balance_loss_clip": 1.04679823,
      "balance_loss_mlp": 1.01989174,
      "epoch": 0.7756868875127758,
      "flos": 18693623007360.0,
      "grad_norm": 1.7685443919302093,
      "language_loss": 0.69981658,
      "learning_rate": 5.048475277121214e-07,
      "loss": 0.72177541,
      "num_input_tokens_seen": 138852135,
      "step": 6451,
      "time_per_iteration": 2.536200523376465
    },
    {
      "auxiliary_loss_clip": 0.01168551,
      "auxiliary_loss_mlp": 0.01020846,
      "balance_loss_clip": 1.0443542,
      "balance_loss_mlp": 1.0135982,
      "epoch": 0.7758071304034149,
      "flos": 28404191543040.0,
      "grad_norm": 3.671404203207576,
      "language_loss": 0.7730397,
      "learning_rate": 5.043302668519598e-07,
      "loss": 0.79493368,
      "num_input_tokens_seen": 138871470,
      "step": 6452,
      "time_per_iteration": 2.673842191696167
    },
    {
      "auxiliary_loss_clip": 0.01172108,
      "auxiliary_loss_mlp": 0.01023166,
      "balance_loss_clip": 1.04498863,
      "balance_loss_mlp": 1.01594758,
      "epoch": 0.775927373294054,
      "flos": 20595344670720.0,
      "grad_norm": 1.8818292805552128,
      "language_loss": 0.72235692,
      "learning_rate": 5.038132328837079e-07,
      "loss": 0.74430966,
      "num_input_tokens_seen": 138889860,
      "step": 6453,
      "time_per_iteration": 2.6567747592926025
    },
    {
      "auxiliary_loss_clip": 0.01170566,
      "auxiliary_loss_mlp": 0.01020179,
      "balance_loss_clip": 1.04535115,
      "balance_loss_mlp": 1.01250815,
      "epoch": 0.7760476161846931,
      "flos": 22526368853760.0,
      "grad_norm": 8.00246472479751,
      "language_loss": 0.74086356,
      "learning_rate": 5.032964258857993e-07,
      "loss": 0.76277101,
      "num_input_tokens_seen": 138909955,
      "step": 6454,
      "time_per_iteration": 2.5846521854400635
    },
    {
      "auxiliary_loss_clip": 0.01166377,
      "auxiliary_loss_mlp": 0.01026472,
      "balance_loss_clip": 1.04181361,
      "balance_loss_mlp": 1.01899481,
      "epoch": 0.7761678590753321,
      "flos": 48651488403840.0,
      "grad_norm": 5.008686599129812,
      "language_loss": 0.68346488,
      "learning_rate": 5.027798459366329e-07,
      "loss": 0.70539331,
      "num_input_tokens_seen": 138935320,
      "step": 6455,
      "time_per_iteration": 2.9233453273773193
    },
    {
      "auxiliary_loss_clip": 0.01174871,
      "auxiliary_loss_mlp": 0.0102842,
      "balance_loss_clip": 1.04633796,
      "balance_loss_mlp": 1.02089834,
      "epoch": 0.7762881019659713,
      "flos": 26177047637760.0,
      "grad_norm": 1.5537245267023667,
      "language_loss": 0.63635218,
      "learning_rate": 5.02263493114573e-07,
      "loss": 0.65838516,
      "num_input_tokens_seen": 138957115,
      "step": 6456,
      "time_per_iteration": 2.6235568523406982
    },
    {
      "auxiliary_loss_clip": 0.01163762,
      "auxiliary_loss_mlp": 0.01026567,
      "balance_loss_clip": 1.04583168,
      "balance_loss_mlp": 1.01859784,
      "epoch": 0.7764083448566104,
      "flos": 20588341518720.0,
      "grad_norm": 2.300712772656312,
      "language_loss": 0.77133453,
      "learning_rate": 5.017473674979502e-07,
      "loss": 0.79323781,
      "num_input_tokens_seen": 138973140,
      "step": 6457,
      "time_per_iteration": 2.5910966396331787
    },
    {
      "auxiliary_loss_clip": 0.01079425,
      "auxiliary_loss_mlp": 0.01002186,
      "balance_loss_clip": 1.01865625,
      "balance_loss_mlp": 1.00142562,
      "epoch": 0.7765285877472494,
      "flos": 67293078560640.0,
      "grad_norm": 0.7444125705632613,
      "language_loss": 0.5833866,
      "learning_rate": 5.01231469165061e-07,
      "loss": 0.60420275,
      "num_input_tokens_seen": 139028965,
      "step": 6458,
      "time_per_iteration": 3.069844961166382
    },
    {
      "auxiliary_loss_clip": 0.01072548,
      "auxiliary_loss_mlp": 0.01000302,
      "balance_loss_clip": 1.01236105,
      "balance_loss_mlp": 0.9995454,
      "epoch": 0.7766488306378886,
      "flos": 61344476121600.0,
      "grad_norm": 0.8328201481633227,
      "language_loss": 0.56861579,
      "learning_rate": 5.007157981941663e-07,
      "loss": 0.58934438,
      "num_input_tokens_seen": 139094325,
      "step": 6459,
      "time_per_iteration": 3.2484350204467773
    },
    {
      "auxiliary_loss_clip": 0.01081455,
      "auxiliary_loss_mlp": 0.0100094,
      "balance_loss_clip": 1.01412916,
      "balance_loss_mlp": 1.00017679,
      "epoch": 0.7767690735285276,
      "flos": 62946199393920.0,
      "grad_norm": 0.8866964613111524,
      "language_loss": 0.67408353,
      "learning_rate": 5.002003546634928e-07,
      "loss": 0.69490749,
      "num_input_tokens_seen": 139150425,
      "step": 6460,
      "time_per_iteration": 3.1630845069885254
    },
    {
      "auxiliary_loss_clip": 0.01174241,
      "auxiliary_loss_mlp": 0.01023446,
      "balance_loss_clip": 1.04412532,
      "balance_loss_mlp": 1.01631725,
      "epoch": 0.7768893164191667,
      "flos": 20886400575360.0,
      "grad_norm": 1.7716940437547548,
      "language_loss": 0.75930476,
      "learning_rate": 4.996851386512331e-07,
      "loss": 0.78128159,
      "num_input_tokens_seen": 139169130,
      "step": 6461,
      "time_per_iteration": 2.7270758152008057
    },
    {
      "auxiliary_loss_clip": 0.01173116,
      "auxiliary_loss_mlp": 0.01026452,
      "balance_loss_clip": 1.04552686,
      "balance_loss_mlp": 1.01801777,
      "epoch": 0.7770095593098058,
      "flos": 20704584908160.0,
      "grad_norm": 1.8319684417783686,
      "language_loss": 0.83018392,
      "learning_rate": 4.991701502355444e-07,
      "loss": 0.85217965,
      "num_input_tokens_seen": 139189595,
      "step": 6462,
      "time_per_iteration": 3.7042884826660156
    },
    {
      "auxiliary_loss_clip": 0.01172712,
      "auxiliary_loss_mlp": 0.01024415,
      "balance_loss_clip": 1.04637527,
      "balance_loss_mlp": 1.01778722,
      "epoch": 0.7771298022004449,
      "flos": 24717709877760.0,
      "grad_norm": 1.9224441674218726,
      "language_loss": 0.76159793,
      "learning_rate": 4.986553894945518e-07,
      "loss": 0.78356916,
      "num_input_tokens_seen": 139210805,
      "step": 6463,
      "time_per_iteration": 2.63187575340271
    },
    {
      "auxiliary_loss_clip": 0.01178046,
      "auxiliary_loss_mlp": 0.01023181,
      "balance_loss_clip": 1.04022992,
      "balance_loss_mlp": 1.01681828,
      "epoch": 0.777250045091084,
      "flos": 25009232659200.0,
      "grad_norm": 1.920392963396874,
      "language_loss": 0.86148965,
      "learning_rate": 4.981408565063416e-07,
      "loss": 0.88350189,
      "num_input_tokens_seen": 139230750,
      "step": 6464,
      "time_per_iteration": 2.835230588912964
    },
    {
      "auxiliary_loss_clip": 0.0116766,
      "auxiliary_loss_mlp": 0.01021731,
      "balance_loss_clip": 1.0473361,
      "balance_loss_mlp": 1.01432157,
      "epoch": 0.777370287981723,
      "flos": 20119887319680.0,
      "grad_norm": 1.8973349755549915,
      "language_loss": 0.7590366,
      "learning_rate": 4.976265513489701e-07,
      "loss": 0.78093046,
      "num_input_tokens_seen": 139250720,
      "step": 6465,
      "time_per_iteration": 2.649580955505371
    },
    {
      "auxiliary_loss_clip": 0.01166805,
      "auxiliary_loss_mlp": 0.01024461,
      "balance_loss_clip": 1.0430367,
      "balance_loss_mlp": 1.0171597,
      "epoch": 0.7774905308723622,
      "flos": 21718809331200.0,
      "grad_norm": 1.935382220910085,
      "language_loss": 0.80641335,
      "learning_rate": 4.971124741004562e-07,
      "loss": 0.82832599,
      "num_input_tokens_seen": 139269720,
      "step": 6466,
      "time_per_iteration": 2.6367604732513428
    },
    {
      "auxiliary_loss_clip": 0.01167727,
      "auxiliary_loss_mlp": 0.01023501,
      "balance_loss_clip": 1.04462326,
      "balance_loss_mlp": 1.01637769,
      "epoch": 0.7776107737630013,
      "flos": 16034115093120.0,
      "grad_norm": 2.5417787178394873,
      "language_loss": 0.76540709,
      "learning_rate": 4.965986248387846e-07,
      "loss": 0.78731936,
      "num_input_tokens_seen": 139288035,
      "step": 6467,
      "time_per_iteration": 2.654968023300171
    },
    {
      "auxiliary_loss_clip": 0.01174642,
      "auxiliary_loss_mlp": 0.01025129,
      "balance_loss_clip": 1.04357958,
      "balance_loss_mlp": 1.01789868,
      "epoch": 0.7777310166536403,
      "flos": 24790895838720.0,
      "grad_norm": 1.7478891640124707,
      "language_loss": 0.77081954,
      "learning_rate": 4.960850036419073e-07,
      "loss": 0.79281723,
      "num_input_tokens_seen": 139307135,
      "step": 6468,
      "time_per_iteration": 3.573497772216797
    },
    {
      "auxiliary_loss_clip": 0.01169237,
      "auxiliary_loss_mlp": 0.01025555,
      "balance_loss_clip": 1.04467058,
      "balance_loss_mlp": 1.01775324,
      "epoch": 0.7778512595442795,
      "flos": 17272530253440.0,
      "grad_norm": 1.944140252019913,
      "language_loss": 0.78773975,
      "learning_rate": 4.955716105877378e-07,
      "loss": 0.80968773,
      "num_input_tokens_seen": 139325905,
      "step": 6469,
      "time_per_iteration": 2.6294827461242676
    },
    {
      "auxiliary_loss_clip": 0.0117271,
      "auxiliary_loss_mlp": 0.01207631,
      "balance_loss_clip": 1.04609227,
      "balance_loss_mlp": 1.00067139,
      "epoch": 0.7779715024349185,
      "flos": 17748418567680.0,
      "grad_norm": 3.988890296438803,
      "language_loss": 0.83092159,
      "learning_rate": 4.950584457541598e-07,
      "loss": 0.854725,
      "num_input_tokens_seen": 139344370,
      "step": 6470,
      "time_per_iteration": 3.478705406188965
    },
    {
      "auxiliary_loss_clip": 0.01171293,
      "auxiliary_loss_mlp": 0.01024382,
      "balance_loss_clip": 1.04581809,
      "balance_loss_mlp": 1.01710176,
      "epoch": 0.7780917453255576,
      "flos": 24316875031680.0,
      "grad_norm": 1.551108903911548,
      "language_loss": 0.82250458,
      "learning_rate": 4.945455092190183e-07,
      "loss": 0.84446132,
      "num_input_tokens_seen": 139365625,
      "step": 6471,
      "time_per_iteration": 2.634427547454834
    },
    {
      "auxiliary_loss_clip": 0.01064949,
      "auxiliary_loss_mlp": 0.01001383,
      "balance_loss_clip": 1.01185989,
      "balance_loss_mlp": 1.00060797,
      "epoch": 0.7782119882161967,
      "flos": 56364601530240.0,
      "grad_norm": 0.6835630340734336,
      "language_loss": 0.5597136,
      "learning_rate": 4.940328010601271e-07,
      "loss": 0.58037686,
      "num_input_tokens_seen": 139430540,
      "step": 6472,
      "time_per_iteration": 3.1489222049713135
    },
    {
      "auxiliary_loss_clip": 0.01184072,
      "auxiliary_loss_mlp": 0.01031023,
      "balance_loss_clip": 1.04985905,
      "balance_loss_mlp": 1.02277386,
      "epoch": 0.7783322311068358,
      "flos": 46789986994560.0,
      "grad_norm": 2.0887774292400167,
      "language_loss": 0.7702859,
      "learning_rate": 4.935203213552621e-07,
      "loss": 0.79243684,
      "num_input_tokens_seen": 139454280,
      "step": 6473,
      "time_per_iteration": 2.8438663482666016
    },
    {
      "auxiliary_loss_clip": 0.01172483,
      "auxiliary_loss_mlp": 0.01022011,
      "balance_loss_clip": 1.044379,
      "balance_loss_mlp": 1.01392293,
      "epoch": 0.7784524739974749,
      "flos": 19057864872960.0,
      "grad_norm": 2.462925330052693,
      "language_loss": 0.66685855,
      "learning_rate": 4.930080701821662e-07,
      "loss": 0.68880349,
      "num_input_tokens_seen": 139471745,
      "step": 6474,
      "time_per_iteration": 3.5904252529144287
    },
    {
      "auxiliary_loss_clip": 0.01170556,
      "auxiliary_loss_mlp": 0.01025076,
      "balance_loss_clip": 1.04286027,
      "balance_loss_mlp": 1.01760459,
      "epoch": 0.778572716888114,
      "flos": 24791111320320.0,
      "grad_norm": 2.0150282218749584,
      "language_loss": 0.77603531,
      "learning_rate": 4.92496047618548e-07,
      "loss": 0.79799163,
      "num_input_tokens_seen": 139491505,
      "step": 6475,
      "time_per_iteration": 2.676276683807373
    },
    {
      "auxiliary_loss_clip": 0.01174103,
      "auxiliary_loss_mlp": 0.01024145,
      "balance_loss_clip": 1.04879451,
      "balance_loss_mlp": 1.0164206,
      "epoch": 0.7786929597787531,
      "flos": 20078086867200.0,
      "grad_norm": 1.8494435717675914,
      "language_loss": 0.77811235,
      "learning_rate": 4.919842537420811e-07,
      "loss": 0.80009484,
      "num_input_tokens_seen": 139508620,
      "step": 6476,
      "time_per_iteration": 2.622852087020874
    },
    {
      "auxiliary_loss_clip": 0.01176259,
      "auxiliary_loss_mlp": 0.01024107,
      "balance_loss_clip": 1.04931355,
      "balance_loss_mlp": 1.01727879,
      "epoch": 0.7788132026693921,
      "flos": 21872220318720.0,
      "grad_norm": 1.760620548167618,
      "language_loss": 0.79593343,
      "learning_rate": 4.91472688630404e-07,
      "loss": 0.81793702,
      "num_input_tokens_seen": 139529360,
      "step": 6477,
      "time_per_iteration": 2.675750494003296
    },
    {
      "auxiliary_loss_clip": 0.01164509,
      "auxiliary_loss_mlp": 0.01023999,
      "balance_loss_clip": 1.04696226,
      "balance_loss_mlp": 1.01702797,
      "epoch": 0.7789334455600313,
      "flos": 11181937351680.0,
      "grad_norm": 2.3649245274145523,
      "language_loss": 0.7412414,
      "learning_rate": 4.909613523611202e-07,
      "loss": 0.76312643,
      "num_input_tokens_seen": 139546240,
      "step": 6478,
      "time_per_iteration": 2.564354419708252
    },
    {
      "auxiliary_loss_clip": 0.01175136,
      "auxiliary_loss_mlp": 0.01207994,
      "balance_loss_clip": 1.03992724,
      "balance_loss_mlp": 1.0007565,
      "epoch": 0.7790536884506704,
      "flos": 28695427015680.0,
      "grad_norm": 1.8605144210284876,
      "language_loss": 0.74587655,
      "learning_rate": 4.904502450117991e-07,
      "loss": 0.76970792,
      "num_input_tokens_seen": 139567200,
      "step": 6479,
      "time_per_iteration": 2.77229380607605
    },
    {
      "auxiliary_loss_clip": 0.01169129,
      "auxiliary_loss_mlp": 0.01025285,
      "balance_loss_clip": 1.04656005,
      "balance_loss_mlp": 1.01805186,
      "epoch": 0.7791739313413094,
      "flos": 11072302064640.0,
      "grad_norm": 2.957762975646124,
      "language_loss": 0.7219193,
      "learning_rate": 4.899393666599762e-07,
      "loss": 0.7438634,
      "num_input_tokens_seen": 139583775,
      "step": 6480,
      "time_per_iteration": 2.5658233165740967
    },
    {
      "auxiliary_loss_clip": 0.01164119,
      "auxiliary_loss_mlp": 0.01020379,
      "balance_loss_clip": 1.04469752,
      "balance_loss_mlp": 1.01317275,
      "epoch": 0.7792941742319486,
      "flos": 14679276975360.0,
      "grad_norm": 2.5470475316926335,
      "language_loss": 0.72497481,
      "learning_rate": 4.894287173831506e-07,
      "loss": 0.74681979,
      "num_input_tokens_seen": 139599735,
      "step": 6481,
      "time_per_iteration": 2.5582635402679443
    },
    {
      "auxiliary_loss_clip": 0.01174361,
      "auxiliary_loss_mlp": 0.01024818,
      "balance_loss_clip": 1.04334664,
      "balance_loss_mlp": 1.0165031,
      "epoch": 0.7794144171225876,
      "flos": 23258874908160.0,
      "grad_norm": 3.173997506941448,
      "language_loss": 0.84157097,
      "learning_rate": 4.889182972587877e-07,
      "loss": 0.8635627,
      "num_input_tokens_seen": 139619030,
      "step": 6482,
      "time_per_iteration": 2.5992791652679443
    },
    {
      "auxiliary_loss_clip": 0.0118294,
      "auxiliary_loss_mlp": 0.01024557,
      "balance_loss_clip": 1.04467225,
      "balance_loss_mlp": 1.01732707,
      "epoch": 0.7795346600132267,
      "flos": 21507080613120.0,
      "grad_norm": 1.9768578822891036,
      "language_loss": 0.66419816,
      "learning_rate": 4.884081063643177e-07,
      "loss": 0.6862731,
      "num_input_tokens_seen": 139637690,
      "step": 6483,
      "time_per_iteration": 2.722670078277588
    },
    {
      "auxiliary_loss_clip": 0.01074097,
      "auxiliary_loss_mlp": 0.01000206,
      "balance_loss_clip": 1.0139327,
      "balance_loss_mlp": 0.99941331,
      "epoch": 0.7796549029038659,
      "flos": 70052273694720.0,
      "grad_norm": 0.8705758934932222,
      "language_loss": 0.52466547,
      "learning_rate": 4.878981447771353e-07,
      "loss": 0.54540849,
      "num_input_tokens_seen": 139692070,
      "step": 6484,
      "time_per_iteration": 3.1829614639282227
    },
    {
      "auxiliary_loss_clip": 0.01170573,
      "auxiliary_loss_mlp": 0.01021566,
      "balance_loss_clip": 1.04328036,
      "balance_loss_mlp": 1.01332593,
      "epoch": 0.7797751457945049,
      "flos": 23989405714560.0,
      "grad_norm": 1.516605592191997,
      "language_loss": 0.72944885,
      "learning_rate": 4.873884125746035e-07,
      "loss": 0.75137025,
      "num_input_tokens_seen": 139713745,
      "step": 6485,
      "time_per_iteration": 2.7138195037841797
    },
    {
      "auxiliary_loss_clip": 0.0116892,
      "auxiliary_loss_mlp": 0.01020976,
      "balance_loss_clip": 1.04400766,
      "balance_loss_mlp": 1.0136621,
      "epoch": 0.779895388685144,
      "flos": 22674751937280.0,
      "grad_norm": 2.0928539423420833,
      "language_loss": 0.72095096,
      "learning_rate": 4.868789098340456e-07,
      "loss": 0.74284995,
      "num_input_tokens_seen": 139731650,
      "step": 6486,
      "time_per_iteration": 2.6316497325897217
    },
    {
      "auxiliary_loss_clip": 0.01176981,
      "auxiliary_loss_mlp": 0.01023948,
      "balance_loss_clip": 1.04332817,
      "balance_loss_mlp": 1.01703691,
      "epoch": 0.7800156315757831,
      "flos": 23768698596480.0,
      "grad_norm": 2.9710525971444204,
      "language_loss": 0.73146307,
      "learning_rate": 4.863696366327543e-07,
      "loss": 0.75347233,
      "num_input_tokens_seen": 139750820,
      "step": 6487,
      "time_per_iteration": 2.7661612033843994
    },
    {
      "auxiliary_loss_clip": 0.01171117,
      "auxiliary_loss_mlp": 0.01026925,
      "balance_loss_clip": 1.04413688,
      "balance_loss_mlp": 1.01939714,
      "epoch": 0.7801358744664222,
      "flos": 26429714881920.0,
      "grad_norm": 2.090129321664799,
      "language_loss": 0.7802372,
      "learning_rate": 4.85860593047986e-07,
      "loss": 0.8022176,
      "num_input_tokens_seen": 139770885,
      "step": 6488,
      "time_per_iteration": 3.679994583129883
    },
    {
      "auxiliary_loss_clip": 0.01167166,
      "auxiliary_loss_mlp": 0.01024549,
      "balance_loss_clip": 1.03846812,
      "balance_loss_mlp": 1.01745343,
      "epoch": 0.7802561173570612,
      "flos": 26322162583680.0,
      "grad_norm": 1.646035299831433,
      "language_loss": 0.75170064,
      "learning_rate": 4.853517791569613e-07,
      "loss": 0.7736178,
      "num_input_tokens_seen": 139793065,
      "step": 6489,
      "time_per_iteration": 2.7182085514068604
    },
    {
      "auxiliary_loss_clip": 0.01178332,
      "auxiliary_loss_mlp": 0.01208377,
      "balance_loss_clip": 1.04453945,
      "balance_loss_mlp": 1.00064301,
      "epoch": 0.7803763602477004,
      "flos": 40333751596800.0,
      "grad_norm": 1.9961382663087235,
      "language_loss": 0.66320324,
      "learning_rate": 4.848431950368684e-07,
      "loss": 0.68707031,
      "num_input_tokens_seen": 139815625,
      "step": 6490,
      "time_per_iteration": 2.8726999759674072
    },
    {
      "auxiliary_loss_clip": 0.01064781,
      "auxiliary_loss_mlp": 0.01198604,
      "balance_loss_clip": 1.01164269,
      "balance_loss_mlp": 0.99988502,
      "epoch": 0.7804966031383395,
      "flos": 67001448038400.0,
      "grad_norm": 0.7147184447359628,
      "language_loss": 0.55724788,
      "learning_rate": 4.843348407648569e-07,
      "loss": 0.57988173,
      "num_input_tokens_seen": 139876905,
      "step": 6491,
      "time_per_iteration": 3.09660005569458
    },
    {
      "auxiliary_loss_clip": 0.01172863,
      "auxiliary_loss_mlp": 0.01022307,
      "balance_loss_clip": 1.04400969,
      "balance_loss_mlp": 1.01424205,
      "epoch": 0.7806168460289785,
      "flos": 17740733057280.0,
      "grad_norm": 10.89110082980647,
      "language_loss": 0.83053601,
      "learning_rate": 4.838267164180457e-07,
      "loss": 0.85248768,
      "num_input_tokens_seen": 139892575,
      "step": 6492,
      "time_per_iteration": 2.564789056777954
    },
    {
      "auxiliary_loss_clip": 0.01169598,
      "auxiliary_loss_mlp": 0.01026416,
      "balance_loss_clip": 1.04794931,
      "balance_loss_mlp": 1.01826262,
      "epoch": 0.7807370889196176,
      "flos": 23946240545280.0,
      "grad_norm": 2.1312211427221643,
      "language_loss": 0.83515924,
      "learning_rate": 4.833188220735156e-07,
      "loss": 0.85711938,
      "num_input_tokens_seen": 139912245,
      "step": 6493,
      "time_per_iteration": 2.5867316722869873
    },
    {
      "auxiliary_loss_clip": 0.01169506,
      "auxiliary_loss_mlp": 0.01025168,
      "balance_loss_clip": 1.04633296,
      "balance_loss_mlp": 1.01777673,
      "epoch": 0.7808573318102567,
      "flos": 18989024457600.0,
      "grad_norm": 2.266123517321665,
      "language_loss": 0.74561387,
      "learning_rate": 4.828111578083152e-07,
      "loss": 0.76756066,
      "num_input_tokens_seen": 139929150,
      "step": 6494,
      "time_per_iteration": 2.5737500190734863
    },
    {
      "auxiliary_loss_clip": 0.01170712,
      "auxiliary_loss_mlp": 0.01026229,
      "balance_loss_clip": 1.04671645,
      "balance_loss_mlp": 1.01828361,
      "epoch": 0.7809775747008958,
      "flos": 23980750536960.0,
      "grad_norm": 2.898647838698878,
      "language_loss": 0.80885583,
      "learning_rate": 4.823037236994556e-07,
      "loss": 0.83082521,
      "num_input_tokens_seen": 139947315,
      "step": 6495,
      "time_per_iteration": 3.5565013885498047
    },
    {
      "auxiliary_loss_clip": 0.01073112,
      "auxiliary_loss_mlp": 0.01002344,
      "balance_loss_clip": 1.0119164,
      "balance_loss_mlp": 1.00158119,
      "epoch": 0.7810978175915348,
      "flos": 68535875180160.0,
      "grad_norm": 0.7139983284922933,
      "language_loss": 0.56291473,
      "learning_rate": 4.817965198239136e-07,
      "loss": 0.5836693,
      "num_input_tokens_seen": 140013775,
      "step": 6496,
      "time_per_iteration": 4.082724571228027
    },
    {
      "auxiliary_loss_clip": 0.01173422,
      "auxiliary_loss_mlp": 0.01026099,
      "balance_loss_clip": 1.04073608,
      "balance_loss_mlp": 1.01790595,
      "epoch": 0.781218060482174,
      "flos": 19642131498240.0,
      "grad_norm": 2.0055411639445766,
      "language_loss": 0.74739778,
      "learning_rate": 4.812895462586331e-07,
      "loss": 0.76939291,
      "num_input_tokens_seen": 140031600,
      "step": 6497,
      "time_per_iteration": 2.6754846572875977
    },
    {
      "auxiliary_loss_clip": 0.01176827,
      "auxiliary_loss_mlp": 0.01024326,
      "balance_loss_clip": 1.04426348,
      "balance_loss_mlp": 1.01745641,
      "epoch": 0.7813383033728131,
      "flos": 25627865621760.0,
      "grad_norm": 1.759375027798739,
      "language_loss": 0.8220489,
      "learning_rate": 4.807828030805207e-07,
      "loss": 0.84406042,
      "num_input_tokens_seen": 140050590,
      "step": 6498,
      "time_per_iteration": 2.7555718421936035
    },
    {
      "auxiliary_loss_clip": 0.01165938,
      "auxiliary_loss_mlp": 0.01032715,
      "balance_loss_clip": 1.0460366,
      "balance_loss_mlp": 1.0250324,
      "epoch": 0.7814585462634521,
      "flos": 20485924865280.0,
      "grad_norm": 2.1333293202908057,
      "language_loss": 0.67715549,
      "learning_rate": 4.802762903664495e-07,
      "loss": 0.69914204,
      "num_input_tokens_seen": 140069770,
      "step": 6499,
      "time_per_iteration": 2.577869176864624
    },
    {
      "auxiliary_loss_clip": 0.01181644,
      "auxiliary_loss_mlp": 0.01028613,
      "balance_loss_clip": 1.04888976,
      "balance_loss_mlp": 1.02033377,
      "epoch": 0.7815787891540913,
      "flos": 22304297018880.0,
      "grad_norm": 2.702490629536837,
      "language_loss": 0.74044693,
      "learning_rate": 4.797700081932565e-07,
      "loss": 0.76254952,
      "num_input_tokens_seen": 140087635,
      "step": 6500,
      "time_per_iteration": 2.720322847366333
    },
    {
      "auxiliary_loss_clip": 0.01174563,
      "auxiliary_loss_mlp": 0.01029693,
      "balance_loss_clip": 1.03805709,
      "balance_loss_mlp": 1.02210498,
      "epoch": 0.7816990320447303,
      "flos": 22600668136320.0,
      "grad_norm": 2.4723102018618994,
      "language_loss": 0.81860471,
      "learning_rate": 4.792639566377442e-07,
      "loss": 0.84064734,
      "num_input_tokens_seen": 140105045,
      "step": 6501,
      "time_per_iteration": 3.6938822269439697
    },
    {
      "auxiliary_loss_clip": 0.01164347,
      "auxiliary_loss_mlp": 0.0102213,
      "balance_loss_clip": 1.04367805,
      "balance_loss_mlp": 1.01429749,
      "epoch": 0.7818192749353694,
      "flos": 24935974871040.0,
      "grad_norm": 1.799363730022981,
      "language_loss": 0.77212846,
      "learning_rate": 4.78758135776681e-07,
      "loss": 0.79399323,
      "num_input_tokens_seen": 140124900,
      "step": 6502,
      "time_per_iteration": 2.6910204887390137
    },
    {
      "auxiliary_loss_clip": 0.01174255,
      "auxiliary_loss_mlp": 0.01027999,
      "balance_loss_clip": 1.04561889,
      "balance_loss_mlp": 1.02073288,
      "epoch": 0.7819395178260086,
      "flos": 23733039369600.0,
      "grad_norm": 6.2792351047330515,
      "language_loss": 0.79151571,
      "learning_rate": 4.782525456867989e-07,
      "loss": 0.81353819,
      "num_input_tokens_seen": 140143755,
      "step": 6503,
      "time_per_iteration": 2.6342482566833496
    },
    {
      "auxiliary_loss_clip": 0.01179004,
      "auxiliary_loss_mlp": 0.01025237,
      "balance_loss_clip": 1.04509664,
      "balance_loss_mlp": 1.01731539,
      "epoch": 0.7820597607166476,
      "flos": 23221671396480.0,
      "grad_norm": 2.1610830573817035,
      "language_loss": 0.83192968,
      "learning_rate": 4.777471864447959e-07,
      "loss": 0.85397208,
      "num_input_tokens_seen": 140164495,
      "step": 6504,
      "time_per_iteration": 2.7206525802612305
    },
    {
      "auxiliary_loss_clip": 0.01173713,
      "auxiliary_loss_mlp": 0.01030419,
      "balance_loss_clip": 1.04363656,
      "balance_loss_mlp": 1.02312958,
      "epoch": 0.7821800036072867,
      "flos": 22309540404480.0,
      "grad_norm": 2.4630752269862173,
      "language_loss": 0.80869102,
      "learning_rate": 4.772420581273344e-07,
      "loss": 0.83073235,
      "num_input_tokens_seen": 140181980,
      "step": 6505,
      "time_per_iteration": 2.649266242980957
    },
    {
      "auxiliary_loss_clip": 0.01165207,
      "auxiliary_loss_mlp": 0.01020758,
      "balance_loss_clip": 1.04559577,
      "balance_loss_mlp": 1.01318204,
      "epoch": 0.7823002464979258,
      "flos": 21544176384000.0,
      "grad_norm": 2.457203422767214,
      "language_loss": 0.76559496,
      "learning_rate": 4.7673716081104134e-07,
      "loss": 0.78745472,
      "num_input_tokens_seen": 140202155,
      "step": 6506,
      "time_per_iteration": 2.5634682178497314
    },
    {
      "auxiliary_loss_clip": 0.01167917,
      "auxiliary_loss_mlp": 0.010271,
      "balance_loss_clip": 1.04721236,
      "balance_loss_mlp": 1.01972699,
      "epoch": 0.7824204893885649,
      "flos": 24535642815360.0,
      "grad_norm": 2.5343584299284307,
      "language_loss": 0.84588957,
      "learning_rate": 4.762324945725109e-07,
      "loss": 0.86783969,
      "num_input_tokens_seen": 140221600,
      "step": 6507,
      "time_per_iteration": 2.65674090385437
    },
    {
      "auxiliary_loss_clip": 0.01167498,
      "auxiliary_loss_mlp": 0.01025995,
      "balance_loss_clip": 1.04667068,
      "balance_loss_mlp": 1.01878917,
      "epoch": 0.782540732279204,
      "flos": 27415211402880.0,
      "grad_norm": 1.7143069839132579,
      "language_loss": 0.76068091,
      "learning_rate": 4.7572805948829844e-07,
      "loss": 0.78261578,
      "num_input_tokens_seen": 140241860,
      "step": 6508,
      "time_per_iteration": 2.62336802482605
    },
    {
      "auxiliary_loss_clip": 0.01182917,
      "auxiliary_loss_mlp": 0.01021956,
      "balance_loss_clip": 1.04320753,
      "balance_loss_mlp": 1.01480329,
      "epoch": 0.7826609751698431,
      "flos": 24353216616960.0,
      "grad_norm": 1.6558037892313142,
      "language_loss": 0.71200216,
      "learning_rate": 4.7522385563492795e-07,
      "loss": 0.73405087,
      "num_input_tokens_seen": 140262160,
      "step": 6509,
      "time_per_iteration": 2.7565231323242188
    },
    {
      "auxiliary_loss_clip": 0.0117819,
      "auxiliary_loss_mlp": 0.01025169,
      "balance_loss_clip": 1.04565907,
      "balance_loss_mlp": 1.01780164,
      "epoch": 0.7827812180604822,
      "flos": 23988543788160.0,
      "grad_norm": 3.356430119020093,
      "language_loss": 0.70741475,
      "learning_rate": 4.747198830888863e-07,
      "loss": 0.72944838,
      "num_input_tokens_seen": 140282030,
      "step": 6510,
      "time_per_iteration": 2.6643340587615967
    },
    {
      "auxiliary_loss_clip": 0.01168659,
      "auxiliary_loss_mlp": 0.01026949,
      "balance_loss_clip": 1.04522479,
      "balance_loss_mlp": 1.01971269,
      "epoch": 0.7829014609511212,
      "flos": 27454318335360.0,
      "grad_norm": 6.591545391807494,
      "language_loss": 0.68449479,
      "learning_rate": 4.742161419266251e-07,
      "loss": 0.70645082,
      "num_input_tokens_seen": 140301190,
      "step": 6511,
      "time_per_iteration": 2.685603618621826
    },
    {
      "auxiliary_loss_clip": 0.01174832,
      "auxiliary_loss_mlp": 0.01028627,
      "balance_loss_clip": 1.04652071,
      "balance_loss_mlp": 1.02032971,
      "epoch": 0.7830217038417604,
      "flos": 29204532432000.0,
      "grad_norm": 2.42527442824437,
      "language_loss": 0.65215039,
      "learning_rate": 4.7371263222456304e-07,
      "loss": 0.67418504,
      "num_input_tokens_seen": 140318510,
      "step": 6512,
      "time_per_iteration": 2.7128052711486816
    },
    {
      "auxiliary_loss_clip": 0.01067156,
      "auxiliary_loss_mlp": 0.01001524,
      "balance_loss_clip": 1.01109743,
      "balance_loss_mlp": 1.00075471,
      "epoch": 0.7831419467323995,
      "flos": 60950895822720.0,
      "grad_norm": 0.8024365106201271,
      "language_loss": 0.61336553,
      "learning_rate": 4.7320935405908004e-07,
      "loss": 0.6340524,
      "num_input_tokens_seen": 140379380,
      "step": 6513,
      "time_per_iteration": 3.147918939590454
    },
    {
      "auxiliary_loss_clip": 0.01170246,
      "auxiliary_loss_mlp": 0.01027923,
      "balance_loss_clip": 1.04774487,
      "balance_loss_mlp": 1.01961398,
      "epoch": 0.7832621896230385,
      "flos": 19682531320320.0,
      "grad_norm": 3.9316723365282487,
      "language_loss": 0.84043211,
      "learning_rate": 4.7270630750652475e-07,
      "loss": 0.86241376,
      "num_input_tokens_seen": 140395335,
      "step": 6514,
      "time_per_iteration": 2.5797007083892822
    },
    {
      "auxiliary_loss_clip": 0.01167576,
      "auxiliary_loss_mlp": 0.01023528,
      "balance_loss_clip": 1.04445219,
      "balance_loss_mlp": 1.01633716,
      "epoch": 0.7833824325136777,
      "flos": 25009232659200.0,
      "grad_norm": 1.8302119215608033,
      "language_loss": 0.80706441,
      "learning_rate": 4.7220349264320746e-07,
      "loss": 0.82897544,
      "num_input_tokens_seen": 140414420,
      "step": 6515,
      "time_per_iteration": 3.608153820037842
    },
    {
      "auxiliary_loss_clip": 0.01071994,
      "auxiliary_loss_mlp": 0.01000232,
      "balance_loss_clip": 1.01241469,
      "balance_loss_mlp": 0.999457,
      "epoch": 0.7835026754043167,
      "flos": 68800142517120.0,
      "grad_norm": 0.7397002973989638,
      "language_loss": 0.54950398,
      "learning_rate": 4.71700909545407e-07,
      "loss": 0.57022625,
      "num_input_tokens_seen": 140477365,
      "step": 6516,
      "time_per_iteration": 3.214388132095337
    },
    {
      "auxiliary_loss_clip": 0.01173045,
      "auxiliary_loss_mlp": 0.01020936,
      "balance_loss_clip": 1.0471561,
      "balance_loss_mlp": 1.01338077,
      "epoch": 0.7836229182949558,
      "flos": 19864598382720.0,
      "grad_norm": 1.8301458786230336,
      "language_loss": 0.77273726,
      "learning_rate": 4.711985582893627e-07,
      "loss": 0.79467708,
      "num_input_tokens_seen": 140495885,
      "step": 6517,
      "time_per_iteration": 2.591128349304199
    },
    {
      "auxiliary_loss_clip": 0.01181077,
      "auxiliary_loss_mlp": 0.01022583,
      "balance_loss_clip": 1.04142475,
      "balance_loss_mlp": 1.01470935,
      "epoch": 0.783743161185595,
      "flos": 22965843755520.0,
      "grad_norm": 2.281063703473845,
      "language_loss": 0.71683824,
      "learning_rate": 4.706964389512811e-07,
      "loss": 0.73887485,
      "num_input_tokens_seen": 140515920,
      "step": 6518,
      "time_per_iteration": 2.7302236557006836
    },
    {
      "auxiliary_loss_clip": 0.01167311,
      "auxiliary_loss_mlp": 0.01021128,
      "balance_loss_clip": 1.04908562,
      "balance_loss_mlp": 1.0138588,
      "epoch": 0.783863404076234,
      "flos": 12458489777280.0,
      "grad_norm": 1.9641289043548917,
      "language_loss": 0.87528229,
      "learning_rate": 4.701945516073345e-07,
      "loss": 0.89716667,
      "num_input_tokens_seen": 140533395,
      "step": 6519,
      "time_per_iteration": 2.5587127208709717
    },
    {
      "auxiliary_loss_clip": 0.01172074,
      "auxiliary_loss_mlp": 0.010205,
      "balance_loss_clip": 1.04355443,
      "balance_loss_mlp": 1.01362729,
      "epoch": 0.7839836469668731,
      "flos": 24243940465920.0,
      "grad_norm": 3.0149585811115758,
      "language_loss": 0.75237727,
      "learning_rate": 4.696928963336577e-07,
      "loss": 0.77430308,
      "num_input_tokens_seen": 140552825,
      "step": 6520,
      "time_per_iteration": 2.7505686283111572
    },
    {
      "auxiliary_loss_clip": 0.01066913,
      "auxiliary_loss_mlp": 0.0100095,
      "balance_loss_clip": 1.01081514,
      "balance_loss_mlp": 1.00019884,
      "epoch": 0.7841038898575122,
      "flos": 62121978938880.0,
      "grad_norm": 0.8549753708570911,
      "language_loss": 0.60972786,
      "learning_rate": 4.6919147320635224e-07,
      "loss": 0.6304065,
      "num_input_tokens_seen": 140615535,
      "step": 6521,
      "time_per_iteration": 4.027263402938843
    },
    {
      "auxiliary_loss_clip": 0.01171772,
      "auxiliary_loss_mlp": 0.01023296,
      "balance_loss_clip": 1.04498386,
      "balance_loss_mlp": 1.01631594,
      "epoch": 0.7842241327481513,
      "flos": 20193899293440.0,
      "grad_norm": 3.963590962184631,
      "language_loss": 0.73206991,
      "learning_rate": 4.6869028230148286e-07,
      "loss": 0.75402051,
      "num_input_tokens_seen": 140633330,
      "step": 6522,
      "time_per_iteration": 2.596656084060669
    },
    {
      "auxiliary_loss_clip": 0.01169291,
      "auxiliary_loss_mlp": 0.01024145,
      "balance_loss_clip": 1.04016304,
      "balance_loss_mlp": 1.0160985,
      "epoch": 0.7843443756387903,
      "flos": 28074531496320.0,
      "grad_norm": 2.550916186846888,
      "language_loss": 0.60084248,
      "learning_rate": 4.6818932369507957e-07,
      "loss": 0.62277681,
      "num_input_tokens_seen": 140652830,
      "step": 6523,
      "time_per_iteration": 3.660430669784546
    },
    {
      "auxiliary_loss_clip": 0.01170677,
      "auxiliary_loss_mlp": 0.01025721,
      "balance_loss_clip": 1.04848516,
      "balance_loss_mlp": 1.01807332,
      "epoch": 0.7844646185294295,
      "flos": 21323397438720.0,
      "grad_norm": 2.5668160670113225,
      "language_loss": 0.89112031,
      "learning_rate": 4.676885974631386e-07,
      "loss": 0.91308427,
      "num_input_tokens_seen": 140671190,
      "step": 6524,
      "time_per_iteration": 2.6406748294830322
    },
    {
      "auxiliary_loss_clip": 0.0117098,
      "auxiliary_loss_mlp": 0.01027,
      "balance_loss_clip": 1.04658866,
      "balance_loss_mlp": 1.01972532,
      "epoch": 0.7845848614200686,
      "flos": 23656585271040.0,
      "grad_norm": 2.315117436145503,
      "language_loss": 0.81103921,
      "learning_rate": 4.67188103681619e-07,
      "loss": 0.83301902,
      "num_input_tokens_seen": 140690975,
      "step": 6525,
      "time_per_iteration": 2.6171154975891113
    },
    {
      "auxiliary_loss_clip": 0.01165497,
      "auxiliary_loss_mlp": 0.01207342,
      "balance_loss_clip": 1.04694152,
      "balance_loss_mlp": 1.00061035,
      "epoch": 0.7847051043107076,
      "flos": 23402194174080.0,
      "grad_norm": 1.9172665909687905,
      "language_loss": 0.69072944,
      "learning_rate": 4.666878424264453e-07,
      "loss": 0.71445787,
      "num_input_tokens_seen": 140710930,
      "step": 6526,
      "time_per_iteration": 2.7607576847076416
    },
    {
      "auxiliary_loss_clip": 0.01163585,
      "auxiliary_loss_mlp": 0.01018311,
      "balance_loss_clip": 1.04395962,
      "balance_loss_mlp": 1.0118171,
      "epoch": 0.7848253472013467,
      "flos": 19022277473280.0,
      "grad_norm": 1.6891007874288224,
      "language_loss": 0.73657542,
      "learning_rate": 4.661878137735069e-07,
      "loss": 0.75839436,
      "num_input_tokens_seen": 140729120,
      "step": 6527,
      "time_per_iteration": 2.6216330528259277
    },
    {
      "auxiliary_loss_clip": 0.01172974,
      "auxiliary_loss_mlp": 0.01023738,
      "balance_loss_clip": 1.04566789,
      "balance_loss_mlp": 1.01628411,
      "epoch": 0.7849455900919858,
      "flos": 21179180332800.0,
      "grad_norm": 1.9573082070660044,
      "language_loss": 0.75147164,
      "learning_rate": 4.656880177986571e-07,
      "loss": 0.77343881,
      "num_input_tokens_seen": 140747665,
      "step": 6528,
      "time_per_iteration": 3.6047022342681885
    },
    {
      "auxiliary_loss_clip": 0.01176018,
      "auxiliary_loss_mlp": 0.01024045,
      "balance_loss_clip": 1.04396081,
      "balance_loss_mlp": 1.01603413,
      "epoch": 0.7850658329826249,
      "flos": 19536482620800.0,
      "grad_norm": 1.963162727210389,
      "language_loss": 0.81897211,
      "learning_rate": 4.6518845457771607e-07,
      "loss": 0.84097272,
      "num_input_tokens_seen": 140766525,
      "step": 6529,
      "time_per_iteration": 2.711357593536377
    },
    {
      "auxiliary_loss_clip": 0.0116309,
      "auxiliary_loss_mlp": 0.01207515,
      "balance_loss_clip": 1.04505551,
      "balance_loss_mlp": 1.00071681,
      "epoch": 0.7851860758732639,
      "flos": 12495334152960.0,
      "grad_norm": 16.607935798970797,
      "language_loss": 0.79551172,
      "learning_rate": 4.646891241864652e-07,
      "loss": 0.8192178,
      "num_input_tokens_seen": 140785090,
      "step": 6530,
      "time_per_iteration": 2.624974489212036
    },
    {
      "auxiliary_loss_clip": 0.01170196,
      "auxiliary_loss_mlp": 0.0102883,
      "balance_loss_clip": 1.04577565,
      "balance_loss_mlp": 1.02057469,
      "epoch": 0.7853063187639031,
      "flos": 22960959505920.0,
      "grad_norm": 2.028154536709927,
      "language_loss": 0.73059249,
      "learning_rate": 4.6419002670065397e-07,
      "loss": 0.75258279,
      "num_input_tokens_seen": 140804670,
      "step": 6531,
      "time_per_iteration": 2.6259076595306396
    },
    {
      "auxiliary_loss_clip": 0.01180254,
      "auxiliary_loss_mlp": 0.01024782,
      "balance_loss_clip": 1.04560041,
      "balance_loss_mlp": 1.01682448,
      "epoch": 0.7854265616545422,
      "flos": 17347260499200.0,
      "grad_norm": 2.0306563025324706,
      "language_loss": 0.86713409,
      "learning_rate": 4.6369116219599445e-07,
      "loss": 0.88918447,
      "num_input_tokens_seen": 140820655,
      "step": 6532,
      "time_per_iteration": 2.601271390914917
    },
    {
      "auxiliary_loss_clip": 0.01171234,
      "auxiliary_loss_mlp": 0.01022793,
      "balance_loss_clip": 1.04174674,
      "balance_loss_mlp": 1.0156821,
      "epoch": 0.7855468045451812,
      "flos": 23838293197440.0,
      "grad_norm": 1.6509282406865242,
      "language_loss": 0.79247248,
      "learning_rate": 4.631925307481637e-07,
      "loss": 0.81441283,
      "num_input_tokens_seen": 140840470,
      "step": 6533,
      "time_per_iteration": 2.7158803939819336
    },
    {
      "auxiliary_loss_clip": 0.01170164,
      "auxiliary_loss_mlp": 0.01022985,
      "balance_loss_clip": 1.04607224,
      "balance_loss_mlp": 1.01595783,
      "epoch": 0.7856670474358204,
      "flos": 25666792986240.0,
      "grad_norm": 2.6165542315901096,
      "language_loss": 0.75878251,
      "learning_rate": 4.6269413243280533e-07,
      "loss": 0.78071404,
      "num_input_tokens_seen": 140859890,
      "step": 6534,
      "time_per_iteration": 2.662635564804077
    },
    {
      "auxiliary_loss_clip": 0.0117732,
      "auxiliary_loss_mlp": 0.01021703,
      "balance_loss_clip": 1.04769087,
      "balance_loss_mlp": 1.01367402,
      "epoch": 0.7857872903264594,
      "flos": 18144656472960.0,
      "grad_norm": 3.360230891328211,
      "language_loss": 0.73716378,
      "learning_rate": 4.621959673255236e-07,
      "loss": 0.75915396,
      "num_input_tokens_seen": 140876190,
      "step": 6535,
      "time_per_iteration": 2.5847251415252686
    },
    {
      "auxiliary_loss_clip": 0.01174162,
      "auxiliary_loss_mlp": 0.01027999,
      "balance_loss_clip": 1.04107714,
      "balance_loss_mlp": 1.02053642,
      "epoch": 0.7859075332170985,
      "flos": 14386138081920.0,
      "grad_norm": 2.078731869740105,
      "language_loss": 0.90793693,
      "learning_rate": 4.6169803550189135e-07,
      "loss": 0.92995852,
      "num_input_tokens_seen": 140891885,
      "step": 6536,
      "time_per_iteration": 2.5839943885803223
    },
    {
      "auxiliary_loss_clip": 0.01168665,
      "auxiliary_loss_mlp": 0.01027402,
      "balance_loss_clip": 1.04188561,
      "balance_loss_mlp": 1.01937342,
      "epoch": 0.7860277761077377,
      "flos": 19864059678720.0,
      "grad_norm": 3.4465727160971236,
      "language_loss": 0.77554536,
      "learning_rate": 4.6120033703744355e-07,
      "loss": 0.79750603,
      "num_input_tokens_seen": 140910780,
      "step": 6537,
      "time_per_iteration": 2.6750049591064453
    },
    {
      "auxiliary_loss_clip": 0.01161187,
      "auxiliary_loss_mlp": 0.01023023,
      "balance_loss_clip": 1.04265797,
      "balance_loss_mlp": 1.01556671,
      "epoch": 0.7861480189983767,
      "flos": 26396174557440.0,
      "grad_norm": 1.9949014315794964,
      "language_loss": 0.78589779,
      "learning_rate": 4.607028720076822e-07,
      "loss": 0.80773997,
      "num_input_tokens_seen": 140927460,
      "step": 6538,
      "time_per_iteration": 2.6300582885742188
    },
    {
      "auxiliary_loss_clip": 0.01170673,
      "auxiliary_loss_mlp": 0.01025339,
      "balance_loss_clip": 1.04694605,
      "balance_loss_mlp": 1.01777232,
      "epoch": 0.7862682618890158,
      "flos": 24236578177920.0,
      "grad_norm": 2.06736122023583,
      "language_loss": 0.73517478,
      "learning_rate": 4.6020564048807074e-07,
      "loss": 0.75713491,
      "num_input_tokens_seen": 140945135,
      "step": 6539,
      "time_per_iteration": 2.6831717491149902
    },
    {
      "auxiliary_loss_clip": 0.01173185,
      "auxiliary_loss_mlp": 0.01024786,
      "balance_loss_clip": 1.04673815,
      "balance_loss_mlp": 1.01700163,
      "epoch": 0.7863885047796549,
      "flos": 47551508259840.0,
      "grad_norm": 2.5218485401821056,
      "language_loss": 0.72124577,
      "learning_rate": 4.5970864255403883e-07,
      "loss": 0.74322546,
      "num_input_tokens_seen": 140966660,
      "step": 6540,
      "time_per_iteration": 2.832057476043701
    },
    {
      "auxiliary_loss_clip": 0.01158005,
      "auxiliary_loss_mlp": 0.01021872,
      "balance_loss_clip": 1.04390657,
      "balance_loss_mlp": 1.01488042,
      "epoch": 0.786508747670294,
      "flos": 24389234979840.0,
      "grad_norm": 3.2463478382029582,
      "language_loss": 0.8222124,
      "learning_rate": 4.59211878280982e-07,
      "loss": 0.84401119,
      "num_input_tokens_seen": 140986175,
      "step": 6541,
      "time_per_iteration": 3.605722427368164
    },
    {
      "auxiliary_loss_clip": 0.01173939,
      "auxiliary_loss_mlp": 0.01022027,
      "balance_loss_clip": 1.04521847,
      "balance_loss_mlp": 1.01438582,
      "epoch": 0.786628990560933,
      "flos": 18041234238720.0,
      "grad_norm": 2.2647150993034595,
      "language_loss": 0.70298374,
      "learning_rate": 4.587153477442578e-07,
      "loss": 0.7249434,
      "num_input_tokens_seen": 141002490,
      "step": 6542,
      "time_per_iteration": 2.629347324371338
    },
    {
      "auxiliary_loss_clip": 0.01171144,
      "auxiliary_loss_mlp": 0.01027039,
      "balance_loss_clip": 1.0489403,
      "balance_loss_mlp": 1.01883125,
      "epoch": 0.7867492334515722,
      "flos": 25848860048640.0,
      "grad_norm": 2.670049253423373,
      "language_loss": 0.81677711,
      "learning_rate": 4.582190510191899e-07,
      "loss": 0.83875889,
      "num_input_tokens_seen": 141021150,
      "step": 6543,
      "time_per_iteration": 2.658003330230713
    },
    {
      "auxiliary_loss_clip": 0.01172108,
      "auxiliary_loss_mlp": 0.01021169,
      "balance_loss_clip": 1.04524565,
      "balance_loss_mlp": 1.01404071,
      "epoch": 0.7868694763422113,
      "flos": 16580819070720.0,
      "grad_norm": 2.2896198947969215,
      "language_loss": 0.87019336,
      "learning_rate": 4.5772298818106625e-07,
      "loss": 0.8921262,
      "num_input_tokens_seen": 141036940,
      "step": 6544,
      "time_per_iteration": 2.6813879013061523
    },
    {
      "auxiliary_loss_clip": 0.01181127,
      "auxiliary_loss_mlp": 0.01026385,
      "balance_loss_clip": 1.04568529,
      "balance_loss_mlp": 1.01846981,
      "epoch": 0.7869897192328503,
      "flos": 29386276272000.0,
      "grad_norm": 2.4008304279996437,
      "language_loss": 0.72229064,
      "learning_rate": 4.572271593051384e-07,
      "loss": 0.74436575,
      "num_input_tokens_seen": 141054295,
      "step": 6545,
      "time_per_iteration": 2.6877801418304443
    },
    {
      "auxiliary_loss_clip": 0.01169901,
      "auxiliary_loss_mlp": 0.0102024,
      "balance_loss_clip": 1.0418818,
      "balance_loss_mlp": 1.01343286,
      "epoch": 0.7871099621234895,
      "flos": 17128923678720.0,
      "grad_norm": 1.917798038157214,
      "language_loss": 0.78409344,
      "learning_rate": 4.567315644666245e-07,
      "loss": 0.80599487,
      "num_input_tokens_seen": 141073090,
      "step": 6546,
      "time_per_iteration": 2.672377109527588
    },
    {
      "auxiliary_loss_clip": 0.01168233,
      "auxiliary_loss_mlp": 0.01021945,
      "balance_loss_clip": 1.0439992,
      "balance_loss_mlp": 1.01472676,
      "epoch": 0.7872302050141285,
      "flos": 23440187784960.0,
      "grad_norm": 2.079943063921063,
      "language_loss": 0.84855855,
      "learning_rate": 4.5623620374070507e-07,
      "loss": 0.87046033,
      "num_input_tokens_seen": 141092405,
      "step": 6547,
      "time_per_iteration": 3.5789923667907715
    },
    {
      "auxiliary_loss_clip": 0.01084028,
      "auxiliary_loss_mlp": 0.01000508,
      "balance_loss_clip": 1.01252568,
      "balance_loss_mlp": 0.9996376,
      "epoch": 0.7873504479047676,
      "flos": 65959752689280.0,
      "grad_norm": 0.7614926118050894,
      "language_loss": 0.58328938,
      "learning_rate": 4.557410772025263e-07,
      "loss": 0.60413474,
      "num_input_tokens_seen": 141154355,
      "step": 6548,
      "time_per_iteration": 3.3254053592681885
    },
    {
      "auxiliary_loss_clip": 0.01167681,
      "auxiliary_loss_mlp": 0.01024946,
      "balance_loss_clip": 1.0430001,
      "balance_loss_mlp": 1.01749551,
      "epoch": 0.7874706907954068,
      "flos": 23258336204160.0,
      "grad_norm": 1.7732988903257672,
      "language_loss": 0.66530204,
      "learning_rate": 4.5524618492719803e-07,
      "loss": 0.68722832,
      "num_input_tokens_seen": 141173575,
      "step": 6549,
      "time_per_iteration": 2.653273820877075
    },
    {
      "auxiliary_loss_clip": 0.01169702,
      "auxiliary_loss_mlp": 0.01024109,
      "balance_loss_clip": 1.04550207,
      "balance_loss_mlp": 1.0168606,
      "epoch": 0.7875909336860458,
      "flos": 28767786963840.0,
      "grad_norm": 1.557424296171864,
      "language_loss": 0.79210609,
      "learning_rate": 4.54751526989795e-07,
      "loss": 0.81404418,
      "num_input_tokens_seen": 141195415,
      "step": 6550,
      "time_per_iteration": 3.5885162353515625
    },
    {
      "auxiliary_loss_clip": 0.01173068,
      "auxiliary_loss_mlp": 0.0102515,
      "balance_loss_clip": 1.04704821,
      "balance_loss_mlp": 1.01739264,
      "epoch": 0.7877111765766849,
      "flos": 18697286194560.0,
      "grad_norm": 2.00538601279826,
      "language_loss": 0.78820252,
      "learning_rate": 4.5425710346535775e-07,
      "loss": 0.81018472,
      "num_input_tokens_seen": 141213360,
      "step": 6551,
      "time_per_iteration": 2.6702680587768555
    },
    {
      "auxiliary_loss_clip": 0.01170561,
      "auxiliary_loss_mlp": 0.01024373,
      "balance_loss_clip": 1.04552078,
      "balance_loss_mlp": 1.01621318,
      "epoch": 0.787831419467324,
      "flos": 27592968833280.0,
      "grad_norm": 2.1242047702304134,
      "language_loss": 0.82026136,
      "learning_rate": 4.537629144288877e-07,
      "loss": 0.84221065,
      "num_input_tokens_seen": 141230815,
      "step": 6552,
      "time_per_iteration": 2.6761057376861572
    },
    {
      "auxiliary_loss_clip": 0.0118418,
      "auxiliary_loss_mlp": 0.01022919,
      "balance_loss_clip": 1.0404644,
      "balance_loss_mlp": 1.01541162,
      "epoch": 0.7879516623579631,
      "flos": 18150187167360.0,
      "grad_norm": 3.459739954508104,
      "language_loss": 0.74935353,
      "learning_rate": 4.5326895995535477e-07,
      "loss": 0.77142453,
      "num_input_tokens_seen": 141249715,
      "step": 6553,
      "time_per_iteration": 2.744115114212036
    },
    {
      "auxiliary_loss_clip": 0.0116839,
      "auxiliary_loss_mlp": 0.01026426,
      "balance_loss_clip": 1.04687023,
      "balance_loss_mlp": 1.01886845,
      "epoch": 0.7880719052486022,
      "flos": 20339193807360.0,
      "grad_norm": 2.45199001254466,
      "language_loss": 0.845191,
      "learning_rate": 4.527752401196907e-07,
      "loss": 0.86713916,
      "num_input_tokens_seen": 141267730,
      "step": 6554,
      "time_per_iteration": 3.541487693786621
    },
    {
      "auxiliary_loss_clip": 0.01165575,
      "auxiliary_loss_mlp": 0.01025821,
      "balance_loss_clip": 1.04366899,
      "balance_loss_mlp": 1.01802444,
      "epoch": 0.7881921481392413,
      "flos": 21653237053440.0,
      "grad_norm": 1.7254613134108077,
      "language_loss": 0.67028475,
      "learning_rate": 4.5228175499679254e-07,
      "loss": 0.69219869,
      "num_input_tokens_seen": 141287315,
      "step": 6555,
      "time_per_iteration": 2.671135663986206
    },
    {
      "auxiliary_loss_clip": 0.01071961,
      "auxiliary_loss_mlp": 0.01000917,
      "balance_loss_clip": 1.01186991,
      "balance_loss_mlp": 1.00014257,
      "epoch": 0.7883123910298804,
      "flos": 68565860058240.0,
      "grad_norm": 0.8281659505018739,
      "language_loss": 0.544438,
      "learning_rate": 4.5178850466152174e-07,
      "loss": 0.56516677,
      "num_input_tokens_seen": 141346145,
      "step": 6556,
      "time_per_iteration": 3.1695685386657715
    },
    {
      "auxiliary_loss_clip": 0.01166849,
      "auxiliary_loss_mlp": 0.01019763,
      "balance_loss_clip": 1.04165578,
      "balance_loss_mlp": 1.01243722,
      "epoch": 0.7884326339205194,
      "flos": 19318217627520.0,
      "grad_norm": 2.1699686017854214,
      "language_loss": 0.82345998,
      "learning_rate": 4.512954891887031e-07,
      "loss": 0.84532607,
      "num_input_tokens_seen": 141364445,
      "step": 6557,
      "time_per_iteration": 2.660862922668457
    },
    {
      "auxiliary_loss_clip": 0.01166676,
      "auxiliary_loss_mlp": 0.01028737,
      "balance_loss_clip": 1.0445298,
      "balance_loss_mlp": 1.02057087,
      "epoch": 0.7885528768111585,
      "flos": 17784903807360.0,
      "grad_norm": 2.6334349299746185,
      "language_loss": 0.83555835,
      "learning_rate": 4.5080270865312806e-07,
      "loss": 0.85751247,
      "num_input_tokens_seen": 141381640,
      "step": 6558,
      "time_per_iteration": 2.6273717880249023
    },
    {
      "auxiliary_loss_clip": 0.0116806,
      "auxiliary_loss_mlp": 0.01021752,
      "balance_loss_clip": 1.04506445,
      "balance_loss_mlp": 1.01460552,
      "epoch": 0.7886731197017977,
      "flos": 18807639753600.0,
      "grad_norm": 2.272980770865802,
      "language_loss": 0.71267641,
      "learning_rate": 4.5031016312954985e-07,
      "loss": 0.7345745,
      "num_input_tokens_seen": 141399955,
      "step": 6559,
      "time_per_iteration": 2.548375129699707
    },
    {
      "auxiliary_loss_clip": 0.0118062,
      "auxiliary_loss_mlp": 0.01025127,
      "balance_loss_clip": 1.04983735,
      "balance_loss_mlp": 1.01752758,
      "epoch": 0.7887933625924367,
      "flos": 33365358126720.0,
      "grad_norm": 3.59390791228607,
      "language_loss": 0.74861968,
      "learning_rate": 4.498178526926886e-07,
      "loss": 0.77067709,
      "num_input_tokens_seen": 141420820,
      "step": 6560,
      "time_per_iteration": 2.7595534324645996
    },
    {
      "auxiliary_loss_clip": 0.01166735,
      "auxiliary_loss_mlp": 0.01027736,
      "balance_loss_clip": 1.04824352,
      "balance_loss_mlp": 1.02066064,
      "epoch": 0.7889136054830758,
      "flos": 17019360218880.0,
      "grad_norm": 2.070406072963789,
      "language_loss": 0.72271812,
      "learning_rate": 4.4932577741722635e-07,
      "loss": 0.74466276,
      "num_input_tokens_seen": 141439350,
      "step": 6561,
      "time_per_iteration": 2.538759469985962
    },
    {
      "auxiliary_loss_clip": 0.01169778,
      "auxiliary_loss_mlp": 0.01029309,
      "balance_loss_clip": 1.04459167,
      "balance_loss_mlp": 1.02135181,
      "epoch": 0.7890338483737149,
      "flos": 29424629018880.0,
      "grad_norm": 1.8239361717136282,
      "language_loss": 0.74122012,
      "learning_rate": 4.4883393737780985e-07,
      "loss": 0.76321101,
      "num_input_tokens_seen": 141460300,
      "step": 6562,
      "time_per_iteration": 2.6677842140197754
    },
    {
      "auxiliary_loss_clip": 0.01163992,
      "auxiliary_loss_mlp": 0.01024274,
      "balance_loss_clip": 1.04444361,
      "balance_loss_mlp": 1.01684141,
      "epoch": 0.789154091264354,
      "flos": 19971576063360.0,
      "grad_norm": 1.940187672384791,
      "language_loss": 0.78615886,
      "learning_rate": 4.4834233264905254e-07,
      "loss": 0.80804151,
      "num_input_tokens_seen": 141477315,
      "step": 6563,
      "time_per_iteration": 2.585413694381714
    },
    {
      "auxiliary_loss_clip": 0.0116951,
      "auxiliary_loss_mlp": 0.01029205,
      "balance_loss_clip": 1.04180944,
      "balance_loss_mlp": 1.02153957,
      "epoch": 0.789274334154993,
      "flos": 14537825216640.0,
      "grad_norm": 2.8421302391617553,
      "language_loss": 0.72208297,
      "learning_rate": 4.478509633055294e-07,
      "loss": 0.74407005,
      "num_input_tokens_seen": 141495025,
      "step": 6564,
      "time_per_iteration": 2.6390185356140137
    },
    {
      "auxiliary_loss_clip": 0.0117652,
      "auxiliary_loss_mlp": 0.01028295,
      "balance_loss_clip": 1.04564714,
      "balance_loss_mlp": 1.02021253,
      "epoch": 0.7893945770456322,
      "flos": 21827403123840.0,
      "grad_norm": 2.5105543243313635,
      "language_loss": 0.79919481,
      "learning_rate": 4.473598294217813e-07,
      "loss": 0.82124305,
      "num_input_tokens_seen": 141510450,
      "step": 6565,
      "time_per_iteration": 2.694789409637451
    },
    {
      "auxiliary_loss_clip": 0.01167427,
      "auxiliary_loss_mlp": 0.01022198,
      "balance_loss_clip": 1.04629064,
      "balance_loss_mlp": 1.01481867,
      "epoch": 0.7895148199362713,
      "flos": 20740639184640.0,
      "grad_norm": 2.502095463236926,
      "language_loss": 0.71678638,
      "learning_rate": 4.468689310723124e-07,
      "loss": 0.73868263,
      "num_input_tokens_seen": 141528265,
      "step": 6566,
      "time_per_iteration": 2.5935893058776855
    },
    {
      "auxiliary_loss_clip": 0.0117741,
      "auxiliary_loss_mlp": 0.01026235,
      "balance_loss_clip": 1.04311037,
      "balance_loss_mlp": 1.01890922,
      "epoch": 0.7896350628269103,
      "flos": 16690669839360.0,
      "grad_norm": 2.2455513003333962,
      "language_loss": 0.78483307,
      "learning_rate": 4.463782683315913e-07,
      "loss": 0.80686951,
      "num_input_tokens_seen": 141547270,
      "step": 6567,
      "time_per_iteration": 2.6782357692718506
    },
    {
      "auxiliary_loss_clip": 0.01163583,
      "auxiliary_loss_mlp": 0.01023679,
      "balance_loss_clip": 1.04606438,
      "balance_loss_mlp": 1.01650226,
      "epoch": 0.7897553057175495,
      "flos": 22638374438400.0,
      "grad_norm": 1.8966113793537493,
      "language_loss": 0.73370957,
      "learning_rate": 4.458878412740523e-07,
      "loss": 0.75558221,
      "num_input_tokens_seen": 141566050,
      "step": 6568,
      "time_per_iteration": 3.507697820663452
    },
    {
      "auxiliary_loss_clip": 0.01165499,
      "auxiliary_loss_mlp": 0.01023652,
      "balance_loss_clip": 1.04555643,
      "balance_loss_mlp": 1.01607585,
      "epoch": 0.7898755486081885,
      "flos": 14537573821440.0,
      "grad_norm": 3.4771460803805434,
      "language_loss": 0.77718019,
      "learning_rate": 4.453976499740919e-07,
      "loss": 0.79907167,
      "num_input_tokens_seen": 141583695,
      "step": 6569,
      "time_per_iteration": 2.6687421798706055
    },
    {
      "auxiliary_loss_clip": 0.01165546,
      "auxiliary_loss_mlp": 0.01023688,
      "balance_loss_clip": 1.04614317,
      "balance_loss_mlp": 1.01623785,
      "epoch": 0.7899957914988276,
      "flos": 17238487138560.0,
      "grad_norm": 2.3704515764815874,
      "language_loss": 0.77887559,
      "learning_rate": 4.4490769450607215e-07,
      "loss": 0.8007679,
      "num_input_tokens_seen": 141601320,
      "step": 6570,
      "time_per_iteration": 2.591688871383667
    },
    {
      "auxiliary_loss_clip": 0.01171958,
      "auxiliary_loss_mlp": 0.01022829,
      "balance_loss_clip": 1.03960383,
      "balance_loss_mlp": 1.01531267,
      "epoch": 0.7901160343894668,
      "flos": 41279351086080.0,
      "grad_norm": 2.5195583702452673,
      "language_loss": 0.72722363,
      "learning_rate": 4.4441797494431845e-07,
      "loss": 0.7491715,
      "num_input_tokens_seen": 141623125,
      "step": 6571,
      "time_per_iteration": 2.8545632362365723
    },
    {
      "auxiliary_loss_clip": 0.01169223,
      "auxiliary_loss_mlp": 0.01021856,
      "balance_loss_clip": 1.04753649,
      "balance_loss_mlp": 1.01389265,
      "epoch": 0.7902362772801058,
      "flos": 16837005847680.0,
      "grad_norm": 2.3661285801756,
      "language_loss": 0.78363538,
      "learning_rate": 4.439284913631207e-07,
      "loss": 0.80554616,
      "num_input_tokens_seen": 141640335,
      "step": 6572,
      "time_per_iteration": 2.635474920272827
    },
    {
      "auxiliary_loss_clip": 0.01180653,
      "auxiliary_loss_mlp": 0.01029676,
      "balance_loss_clip": 1.04705858,
      "balance_loss_mlp": 1.02226102,
      "epoch": 0.7903565201707449,
      "flos": 27125987091840.0,
      "grad_norm": 2.3660044478958815,
      "language_loss": 0.8355757,
      "learning_rate": 4.434392438367347e-07,
      "loss": 0.85767901,
      "num_input_tokens_seen": 141659760,
      "step": 6573,
      "time_per_iteration": 2.711404800415039
    },
    {
      "auxiliary_loss_clip": 0.01174457,
      "auxiliary_loss_mlp": 0.01020387,
      "balance_loss_clip": 1.04502964,
      "balance_loss_mlp": 1.01256979,
      "epoch": 0.790476763061384,
      "flos": 31025167142400.0,
      "grad_norm": 3.114955659314269,
      "language_loss": 0.74352992,
      "learning_rate": 4.4295023243937677e-07,
      "loss": 0.76547837,
      "num_input_tokens_seen": 141679965,
      "step": 6574,
      "time_per_iteration": 3.630094528198242
    },
    {
      "auxiliary_loss_clip": 0.011715,
      "auxiliary_loss_mlp": 0.01024991,
      "balance_loss_clip": 1.04820931,
      "balance_loss_mlp": 1.01740348,
      "epoch": 0.7905970059520231,
      "flos": 22089084681600.0,
      "grad_norm": 1.6100252736987375,
      "language_loss": 0.80110872,
      "learning_rate": 4.4246145724523123e-07,
      "loss": 0.82307363,
      "num_input_tokens_seen": 141697710,
      "step": 6575,
      "time_per_iteration": 2.6072909832000732
    },
    {
      "auxiliary_loss_clip": 0.01173592,
      "auxiliary_loss_mlp": 0.01024377,
      "balance_loss_clip": 1.04418302,
      "balance_loss_mlp": 1.01691449,
      "epoch": 0.7907172488426621,
      "flos": 20558141159040.0,
      "grad_norm": 10.354254962736983,
      "language_loss": 0.7737208,
      "learning_rate": 4.41972918328444e-07,
      "loss": 0.79570049,
      "num_input_tokens_seen": 141715145,
      "step": 6576,
      "time_per_iteration": 2.7064578533172607
    },
    {
      "auxiliary_loss_clip": 0.01168881,
      "auxiliary_loss_mlp": 0.01029483,
      "balance_loss_clip": 1.04690742,
      "balance_loss_mlp": 1.0215888,
      "epoch": 0.7908374917333013,
      "flos": 30081542901120.0,
      "grad_norm": 2.45083497179853,
      "language_loss": 0.77491373,
      "learning_rate": 4.4148461576312646e-07,
      "loss": 0.79689735,
      "num_input_tokens_seen": 141734810,
      "step": 6577,
      "time_per_iteration": 3.5588059425354004
    },
    {
      "auxiliary_loss_clip": 0.01171314,
      "auxiliary_loss_mlp": 0.01023438,
      "balance_loss_clip": 1.04957533,
      "balance_loss_mlp": 1.01622581,
      "epoch": 0.7909577346239404,
      "flos": 20996359084800.0,
      "grad_norm": 1.5090088018256735,
      "language_loss": 0.74793565,
      "learning_rate": 4.4099654962335343e-07,
      "loss": 0.76988316,
      "num_input_tokens_seen": 141755260,
      "step": 6578,
      "time_per_iteration": 2.6448988914489746
    },
    {
      "auxiliary_loss_clip": 0.01180226,
      "auxiliary_loss_mlp": 0.01024676,
      "balance_loss_clip": 1.04829752,
      "balance_loss_mlp": 1.01711249,
      "epoch": 0.7910779775145794,
      "flos": 26247935128320.0,
      "grad_norm": 2.068923757827151,
      "language_loss": 0.75031453,
      "learning_rate": 4.405087199831636e-07,
      "loss": 0.77236354,
      "num_input_tokens_seen": 141775500,
      "step": 6579,
      "time_per_iteration": 2.773796796798706
    },
    {
      "auxiliary_loss_clip": 0.01173979,
      "auxiliary_loss_mlp": 0.01207487,
      "balance_loss_clip": 1.04344106,
      "balance_loss_mlp": 1.0005821,
      "epoch": 0.7911982204052186,
      "flos": 22564434291840.0,
      "grad_norm": 2.0051794030176624,
      "language_loss": 0.67336917,
      "learning_rate": 4.400211269165619e-07,
      "loss": 0.69718385,
      "num_input_tokens_seen": 141791955,
      "step": 6580,
      "time_per_iteration": 2.683145761489868
    },
    {
      "auxiliary_loss_clip": 0.01171985,
      "auxiliary_loss_mlp": 0.01022418,
      "balance_loss_clip": 1.0524385,
      "balance_loss_mlp": 1.01586497,
      "epoch": 0.7913184632958576,
      "flos": 23112538899840.0,
      "grad_norm": 4.390721759622264,
      "language_loss": 0.76908886,
      "learning_rate": 4.3953377049751416e-07,
      "loss": 0.79103285,
      "num_input_tokens_seen": 141812380,
      "step": 6581,
      "time_per_iteration": 3.542029857635498
    },
    {
      "auxiliary_loss_clip": 0.01178131,
      "auxiliary_loss_mlp": 0.01028386,
      "balance_loss_clip": 1.04717958,
      "balance_loss_mlp": 1.02093577,
      "epoch": 0.7914387061864967,
      "flos": 12311758719360.0,
      "grad_norm": 2.2612113804267633,
      "language_loss": 0.783328,
      "learning_rate": 4.390466507999537e-07,
      "loss": 0.80539316,
      "num_input_tokens_seen": 141828130,
      "step": 6582,
      "time_per_iteration": 2.601274251937866
    },
    {
      "auxiliary_loss_clip": 0.01171463,
      "auxiliary_loss_mlp": 0.01029601,
      "balance_loss_clip": 1.04107118,
      "balance_loss_mlp": 1.02232313,
      "epoch": 0.7915589490771359,
      "flos": 17603267708160.0,
      "grad_norm": 2.7699965601278005,
      "language_loss": 0.76321721,
      "learning_rate": 4.385597678977748e-07,
      "loss": 0.78522784,
      "num_input_tokens_seen": 141846965,
      "step": 6583,
      "time_per_iteration": 2.5887181758880615
    },
    {
      "auxiliary_loss_clip": 0.01169304,
      "auxiliary_loss_mlp": 0.0102215,
      "balance_loss_clip": 1.04271984,
      "balance_loss_mlp": 1.01455605,
      "epoch": 0.7916791919677749,
      "flos": 25591272641280.0,
      "grad_norm": 1.7210085362956768,
      "language_loss": 0.75557971,
      "learning_rate": 4.3807312186483726e-07,
      "loss": 0.77749425,
      "num_input_tokens_seen": 141867685,
      "step": 6584,
      "time_per_iteration": 2.6753039360046387
    },
    {
      "auxiliary_loss_clip": 0.01167722,
      "auxiliary_loss_mlp": 0.01023721,
      "balance_loss_clip": 1.04830503,
      "balance_loss_mlp": 1.01610374,
      "epoch": 0.791799434858414,
      "flos": 18844340474880.0,
      "grad_norm": 2.24121746183905,
      "language_loss": 0.78279984,
      "learning_rate": 4.375867127749655e-07,
      "loss": 0.80471426,
      "num_input_tokens_seen": 141885960,
      "step": 6585,
      "time_per_iteration": 2.596837043762207
    },
    {
      "auxiliary_loss_clip": 0.01175724,
      "auxiliary_loss_mlp": 0.01025763,
      "balance_loss_clip": 1.04645944,
      "balance_loss_mlp": 1.01797831,
      "epoch": 0.7919196777490531,
      "flos": 25812015672960.0,
      "grad_norm": 1.9132174926895604,
      "language_loss": 0.67267352,
      "learning_rate": 4.3710054070194744e-07,
      "loss": 0.69468838,
      "num_input_tokens_seen": 141905655,
      "step": 6586,
      "time_per_iteration": 2.672914505004883
    },
    {
      "auxiliary_loss_clip": 0.0116977,
      "auxiliary_loss_mlp": 0.01207775,
      "balance_loss_clip": 1.04816639,
      "balance_loss_mlp": 1.00065577,
      "epoch": 0.7920399206396922,
      "flos": 11947624594560.0,
      "grad_norm": 3.923703873457609,
      "language_loss": 0.67011905,
      "learning_rate": 4.3661460571953455e-07,
      "loss": 0.69389451,
      "num_input_tokens_seen": 141922390,
      "step": 6587,
      "time_per_iteration": 2.6504580974578857
    },
    {
      "auxiliary_loss_clip": 0.01169133,
      "auxiliary_loss_mlp": 0.01020962,
      "balance_loss_clip": 1.04302049,
      "balance_loss_mlp": 1.01368999,
      "epoch": 0.7921601635303313,
      "flos": 21579907438080.0,
      "grad_norm": 2.5758754385807006,
      "language_loss": 0.68491691,
      "learning_rate": 4.36128907901443e-07,
      "loss": 0.70681787,
      "num_input_tokens_seen": 141941985,
      "step": 6588,
      "time_per_iteration": 2.6966681480407715
    },
    {
      "auxiliary_loss_clip": 0.0117675,
      "auxiliary_loss_mlp": 0.01023666,
      "balance_loss_clip": 1.04219651,
      "balance_loss_mlp": 1.01566124,
      "epoch": 0.7922804064209703,
      "flos": 18113989236480.0,
      "grad_norm": 2.074244473508614,
      "language_loss": 0.72526455,
      "learning_rate": 4.356434473213519e-07,
      "loss": 0.74726874,
      "num_input_tokens_seen": 141959435,
      "step": 6589,
      "time_per_iteration": 2.6296918392181396
    },
    {
      "auxiliary_loss_clip": 0.01172851,
      "auxiliary_loss_mlp": 0.01023793,
      "balance_loss_clip": 1.04741955,
      "balance_loss_mlp": 1.01667035,
      "epoch": 0.7924006493116095,
      "flos": 21652806090240.0,
      "grad_norm": 1.6427088223833894,
      "language_loss": 0.79653943,
      "learning_rate": 4.351582240529068e-07,
      "loss": 0.81850582,
      "num_input_tokens_seen": 141980265,
      "step": 6590,
      "time_per_iteration": 2.691801071166992
    },
    {
      "auxiliary_loss_clip": 0.0108047,
      "auxiliary_loss_mlp": 0.01000338,
      "balance_loss_clip": 1.01103354,
      "balance_loss_mlp": 0.99952185,
      "epoch": 0.7925208922022485,
      "flos": 64242755694720.0,
      "grad_norm": 0.6803799460560298,
      "language_loss": 0.58145285,
      "learning_rate": 4.346732381697149e-07,
      "loss": 0.60226095,
      "num_input_tokens_seen": 142044395,
      "step": 6591,
      "time_per_iteration": 3.2295644283294678
    },
    {
      "auxiliary_loss_clip": 0.01166188,
      "auxiliary_loss_mlp": 0.01025528,
      "balance_loss_clip": 1.04495072,
      "balance_loss_mlp": 1.01796389,
      "epoch": 0.7926411350928876,
      "flos": 16941541403520.0,
      "grad_norm": 1.9919521787071008,
      "language_loss": 0.81529325,
      "learning_rate": 4.3418848974534825e-07,
      "loss": 0.83721042,
      "num_input_tokens_seen": 142061335,
      "step": 6592,
      "time_per_iteration": 2.5816056728363037
    },
    {
      "auxiliary_loss_clip": 0.01180479,
      "auxiliary_loss_mlp": 0.01024684,
      "balance_loss_clip": 1.04557431,
      "balance_loss_mlp": 1.01749611,
      "epoch": 0.7927613779835267,
      "flos": 34459987144320.0,
      "grad_norm": 1.8718606730139564,
      "language_loss": 0.69069922,
      "learning_rate": 4.3370397885334276e-07,
      "loss": 0.71275091,
      "num_input_tokens_seen": 142081965,
      "step": 6593,
      "time_per_iteration": 2.729557991027832
    },
    {
      "auxiliary_loss_clip": 0.01164877,
      "auxiliary_loss_mlp": 0.0102944,
      "balance_loss_clip": 1.04757953,
      "balance_loss_mlp": 1.02189064,
      "epoch": 0.7928816208741658,
      "flos": 18951174501120.0,
      "grad_norm": 2.3463307371698217,
      "language_loss": 0.75947464,
      "learning_rate": 4.3321970556719777e-07,
      "loss": 0.78141785,
      "num_input_tokens_seen": 142100260,
      "step": 6594,
      "time_per_iteration": 2.6617608070373535
    },
    {
      "auxiliary_loss_clip": 0.0116767,
      "auxiliary_loss_mlp": 0.01028914,
      "balance_loss_clip": 1.04848945,
      "balance_loss_mlp": 1.02099216,
      "epoch": 0.7930018637648049,
      "flos": 18623022825600.0,
      "grad_norm": 3.07624501318581,
      "language_loss": 0.72286493,
      "learning_rate": 4.3273566996037856e-07,
      "loss": 0.74483073,
      "num_input_tokens_seen": 142116955,
      "step": 6595,
      "time_per_iteration": 3.458833694458008
    },
    {
      "auxiliary_loss_clip": 0.0117167,
      "auxiliary_loss_mlp": 0.01025401,
      "balance_loss_clip": 1.04567695,
      "balance_loss_mlp": 1.01818025,
      "epoch": 0.793122106655444,
      "flos": 24530650824960.0,
      "grad_norm": 4.395914605693892,
      "language_loss": 0.80686283,
      "learning_rate": 4.322518721063113e-07,
      "loss": 0.82883352,
      "num_input_tokens_seen": 142135505,
      "step": 6596,
      "time_per_iteration": 2.6680843830108643
    },
    {
      "auxiliary_loss_clip": 0.0117251,
      "auxiliary_loss_mlp": 0.01024561,
      "balance_loss_clip": 1.04921937,
      "balance_loss_mlp": 1.01705647,
      "epoch": 0.7932423495460831,
      "flos": 34421203434240.0,
      "grad_norm": 2.3650534789378868,
      "language_loss": 0.70470631,
      "learning_rate": 4.3176831207838906e-07,
      "loss": 0.72667694,
      "num_input_tokens_seen": 142158915,
      "step": 6597,
      "time_per_iteration": 2.7404673099517822
    },
    {
      "auxiliary_loss_clip": 0.01170006,
      "auxiliary_loss_mlp": 0.01023643,
      "balance_loss_clip": 1.04945314,
      "balance_loss_mlp": 1.01604629,
      "epoch": 0.7933625924367221,
      "flos": 26980333441920.0,
      "grad_norm": 2.8782047655457212,
      "language_loss": 0.74515057,
      "learning_rate": 4.3128498994996685e-07,
      "loss": 0.7670871,
      "num_input_tokens_seen": 142178390,
      "step": 6598,
      "time_per_iteration": 2.655848264694214
    },
    {
      "auxiliary_loss_clip": 0.01175743,
      "auxiliary_loss_mlp": 0.01024963,
      "balance_loss_clip": 1.0481019,
      "balance_loss_mlp": 1.01684523,
      "epoch": 0.7934828353273613,
      "flos": 29568630643200.0,
      "grad_norm": 2.188837474721115,
      "language_loss": 0.71503377,
      "learning_rate": 4.308019057943646e-07,
      "loss": 0.73704088,
      "num_input_tokens_seen": 142200115,
      "step": 6599,
      "time_per_iteration": 2.7150821685791016
    },
    {
      "auxiliary_loss_clip": 0.01187953,
      "auxiliary_loss_mlp": 0.01028942,
      "balance_loss_clip": 1.04491687,
      "balance_loss_mlp": 1.02171803,
      "epoch": 0.7936030782180004,
      "flos": 28615381557120.0,
      "grad_norm": 1.6404944412904792,
      "language_loss": 0.74646503,
      "learning_rate": 4.3031905968486535e-07,
      "loss": 0.76863402,
      "num_input_tokens_seen": 142220945,
      "step": 6600,
      "time_per_iteration": 2.6736507415771484
    },
    {
      "auxiliary_loss_clip": 0.01177262,
      "auxiliary_loss_mlp": 0.01021604,
      "balance_loss_clip": 1.04553008,
      "balance_loss_mlp": 1.01406956,
      "epoch": 0.7937233211086394,
      "flos": 16392574869120.0,
      "grad_norm": 2.181944164503043,
      "language_loss": 0.69021642,
      "learning_rate": 4.298364516947162e-07,
      "loss": 0.71220505,
      "num_input_tokens_seen": 142238175,
      "step": 6601,
      "time_per_iteration": 3.6228151321411133
    },
    {
      "auxiliary_loss_clip": 0.01173336,
      "auxiliary_loss_mlp": 0.01021175,
      "balance_loss_clip": 1.04071534,
      "balance_loss_mlp": 1.01409054,
      "epoch": 0.7938435639992786,
      "flos": 22013420682240.0,
      "grad_norm": 2.632144921444883,
      "language_loss": 0.6573813,
      "learning_rate": 4.293540818971295e-07,
      "loss": 0.67932642,
      "num_input_tokens_seen": 142255980,
      "step": 6602,
      "time_per_iteration": 2.668024778366089
    },
    {
      "auxiliary_loss_clip": 0.01176998,
      "auxiliary_loss_mlp": 0.01019266,
      "balance_loss_clip": 1.04812241,
      "balance_loss_mlp": 1.0117681,
      "epoch": 0.7939638068899176,
      "flos": 22197032029440.0,
      "grad_norm": 3.2770162915039336,
      "language_loss": 0.77165639,
      "learning_rate": 4.2887195036527934e-07,
      "loss": 0.79361904,
      "num_input_tokens_seen": 142274785,
      "step": 6603,
      "time_per_iteration": 3.640345811843872
    },
    {
      "auxiliary_loss_clip": 0.01158537,
      "auxiliary_loss_mlp": 0.01023021,
      "balance_loss_clip": 1.04111981,
      "balance_loss_mlp": 1.01546943,
      "epoch": 0.7940840497805567,
      "flos": 17745186343680.0,
      "grad_norm": 2.4657053613270405,
      "language_loss": 0.73922497,
      "learning_rate": 4.28390057172306e-07,
      "loss": 0.76104057,
      "num_input_tokens_seen": 142291290,
      "step": 6604,
      "time_per_iteration": 2.726361036300659
    },
    {
      "auxiliary_loss_clip": 0.0116917,
      "auxiliary_loss_mlp": 0.01027874,
      "balance_loss_clip": 1.03975248,
      "balance_loss_mlp": 1.01947558,
      "epoch": 0.7942042926711959,
      "flos": 23805435231360.0,
      "grad_norm": 2.800563067208125,
      "language_loss": 0.71975297,
      "learning_rate": 4.279084023913111e-07,
      "loss": 0.74172342,
      "num_input_tokens_seen": 142309165,
      "step": 6605,
      "time_per_iteration": 2.621253490447998
    },
    {
      "auxiliary_loss_clip": 0.01169258,
      "auxiliary_loss_mlp": 0.01026014,
      "balance_loss_clip": 1.04781187,
      "balance_loss_mlp": 1.01884937,
      "epoch": 0.7943245355618349,
      "flos": 19244959839360.0,
      "grad_norm": 1.9670903016779904,
      "language_loss": 0.69676673,
      "learning_rate": 4.2742698609536096e-07,
      "loss": 0.71871948,
      "num_input_tokens_seen": 142327475,
      "step": 6606,
      "time_per_iteration": 2.659015417098999
    },
    {
      "auxiliary_loss_clip": 0.01176668,
      "auxiliary_loss_mlp": 0.01024346,
      "balance_loss_clip": 1.04613173,
      "balance_loss_mlp": 1.01683009,
      "epoch": 0.794444778452474,
      "flos": 25007616547200.0,
      "grad_norm": 2.036758139837038,
      "language_loss": 0.79016209,
      "learning_rate": 4.2694580835748706e-07,
      "loss": 0.81217223,
      "num_input_tokens_seen": 142347335,
      "step": 6607,
      "time_per_iteration": 2.6567633152008057
    },
    {
      "auxiliary_loss_clip": 0.01171851,
      "auxiliary_loss_mlp": 0.01024009,
      "balance_loss_clip": 1.04430258,
      "balance_loss_mlp": 1.0161984,
      "epoch": 0.7945650213431131,
      "flos": 23221491828480.0,
      "grad_norm": 2.4195957745285757,
      "language_loss": 0.74276519,
      "learning_rate": 4.264648692506836e-07,
      "loss": 0.76472378,
      "num_input_tokens_seen": 142366125,
      "step": 6608,
      "time_per_iteration": 3.5441601276397705
    },
    {
      "auxiliary_loss_clip": 0.01164798,
      "auxiliary_loss_mlp": 0.01025044,
      "balance_loss_clip": 1.04113746,
      "balance_loss_mlp": 1.01687253,
      "epoch": 0.7946852642337522,
      "flos": 26062887237120.0,
      "grad_norm": 1.9250691109188174,
      "language_loss": 0.72253478,
      "learning_rate": 4.2598416884790824e-07,
      "loss": 0.74443316,
      "num_input_tokens_seen": 142385175,
      "step": 6609,
      "time_per_iteration": 2.7143218517303467
    },
    {
      "auxiliary_loss_clip": 0.01181791,
      "auxiliary_loss_mlp": 0.01024934,
      "balance_loss_clip": 1.04470623,
      "balance_loss_mlp": 1.01653004,
      "epoch": 0.7948055071243912,
      "flos": 23769704177280.0,
      "grad_norm": 4.691234612668338,
      "language_loss": 0.81173384,
      "learning_rate": 4.255037072220828e-07,
      "loss": 0.83380109,
      "num_input_tokens_seen": 142406545,
      "step": 6610,
      "time_per_iteration": 2.722666025161743
    },
    {
      "auxiliary_loss_clip": 0.01165511,
      "auxiliary_loss_mlp": 0.01022502,
      "balance_loss_clip": 1.04802656,
      "balance_loss_mlp": 1.01557028,
      "epoch": 0.7949257500150304,
      "flos": 21980814111360.0,
      "grad_norm": 1.6144260250363147,
      "language_loss": 0.71683383,
      "learning_rate": 4.2502348444609293e-07,
      "loss": 0.73871392,
      "num_input_tokens_seen": 142426165,
      "step": 6611,
      "time_per_iteration": 2.6315345764160156
    },
    {
      "auxiliary_loss_clip": 0.01174895,
      "auxiliary_loss_mlp": 0.01022412,
      "balance_loss_clip": 1.03935432,
      "balance_loss_mlp": 1.01550937,
      "epoch": 0.7950459929056695,
      "flos": 25774129802880.0,
      "grad_norm": 1.922136028270983,
      "language_loss": 0.69515616,
      "learning_rate": 4.2454350059278844e-07,
      "loss": 0.71712923,
      "num_input_tokens_seen": 142447225,
      "step": 6612,
      "time_per_iteration": 2.7825767993927
    },
    {
      "auxiliary_loss_clip": 0.01165396,
      "auxiliary_loss_mlp": 0.01025712,
      "balance_loss_clip": 1.04014254,
      "balance_loss_mlp": 1.01845205,
      "epoch": 0.7951662357963085,
      "flos": 22158068751360.0,
      "grad_norm": 1.7925381936236653,
      "language_loss": 0.84630066,
      "learning_rate": 4.240637557349824e-07,
      "loss": 0.86821175,
      "num_input_tokens_seen": 142464440,
      "step": 6613,
      "time_per_iteration": 2.6912691593170166
    },
    {
      "auxiliary_loss_clip": 0.01159275,
      "auxiliary_loss_mlp": 0.01023231,
      "balance_loss_clip": 1.04364252,
      "balance_loss_mlp": 1.01569104,
      "epoch": 0.7952864786869477,
      "flos": 24641938137600.0,
      "grad_norm": 3.8226918248707222,
      "language_loss": 0.66622806,
      "learning_rate": 4.235842499454516e-07,
      "loss": 0.68805313,
      "num_input_tokens_seen": 142484355,
      "step": 6614,
      "time_per_iteration": 2.646528482437134
    },
    {
      "auxiliary_loss_clip": 0.01175265,
      "auxiliary_loss_mlp": 0.01029374,
      "balance_loss_clip": 1.04657018,
      "balance_loss_mlp": 1.02185202,
      "epoch": 0.7954067215775867,
      "flos": 21830922656640.0,
      "grad_norm": 1.7548953850962372,
      "language_loss": 0.83073062,
      "learning_rate": 4.2310498329693687e-07,
      "loss": 0.852777,
      "num_input_tokens_seen": 142505255,
      "step": 6615,
      "time_per_iteration": 2.688822031021118
    },
    {
      "auxiliary_loss_clip": 0.01173064,
      "auxiliary_loss_mlp": 0.01024322,
      "balance_loss_clip": 1.0477438,
      "balance_loss_mlp": 1.01590014,
      "epoch": 0.7955269644682258,
      "flos": 24060652341120.0,
      "grad_norm": 1.697701996106827,
      "language_loss": 0.8081336,
      "learning_rate": 4.2262595586214164e-07,
      "loss": 0.83010745,
      "num_input_tokens_seen": 142526350,
      "step": 6616,
      "time_per_iteration": 2.651703119277954
    },
    {
      "auxiliary_loss_clip": 0.01175119,
      "auxiliary_loss_mlp": 0.01027313,
      "balance_loss_clip": 1.04814601,
      "balance_loss_mlp": 1.01945758,
      "epoch": 0.795647207358865,
      "flos": 25010741030400.0,
      "grad_norm": 2.160205325819151,
      "language_loss": 0.76926428,
      "learning_rate": 4.221471677137358e-07,
      "loss": 0.79128861,
      "num_input_tokens_seen": 142547165,
      "step": 6617,
      "time_per_iteration": 2.6942739486694336
    },
    {
      "auxiliary_loss_clip": 0.01162857,
      "auxiliary_loss_mlp": 0.01022471,
      "balance_loss_clip": 1.04479277,
      "balance_loss_mlp": 1.01566136,
      "epoch": 0.795767450249504,
      "flos": 14648358343680.0,
      "grad_norm": 3.0548166200552296,
      "language_loss": 0.70357984,
      "learning_rate": 4.216686189243492e-07,
      "loss": 0.72543311,
      "num_input_tokens_seen": 142565955,
      "step": 6618,
      "time_per_iteration": 2.7126059532165527
    },
    {
      "auxiliary_loss_clip": 0.01171411,
      "auxiliary_loss_mlp": 0.01023852,
      "balance_loss_clip": 1.04371393,
      "balance_loss_mlp": 1.01596045,
      "epoch": 0.7958876931401431,
      "flos": 18547897530240.0,
      "grad_norm": 2.030217280667337,
      "language_loss": 0.72804409,
      "learning_rate": 4.211903095665785e-07,
      "loss": 0.74999666,
      "num_input_tokens_seen": 142585340,
      "step": 6619,
      "time_per_iteration": 2.6789071559906006
    },
    {
      "auxiliary_loss_clip": 0.01163243,
      "auxiliary_loss_mlp": 0.01026377,
      "balance_loss_clip": 1.04421234,
      "balance_loss_mlp": 1.01896524,
      "epoch": 0.7960079360307821,
      "flos": 21543960902400.0,
      "grad_norm": 1.7698085300934363,
      "language_loss": 0.75147438,
      "learning_rate": 4.2071223971298277e-07,
      "loss": 0.77337062,
      "num_input_tokens_seen": 142602525,
      "step": 6620,
      "time_per_iteration": 2.57309889793396
    },
    {
      "auxiliary_loss_clip": 0.01170733,
      "auxiliary_loss_mlp": 0.01024651,
      "balance_loss_clip": 1.04596674,
      "balance_loss_mlp": 1.01636553,
      "epoch": 0.7961281789214213,
      "flos": 25481745095040.0,
      "grad_norm": 3.0509516892860358,
      "language_loss": 0.61345762,
      "learning_rate": 4.2023440943608433e-07,
      "loss": 0.63541144,
      "num_input_tokens_seen": 142622490,
      "step": 6621,
      "time_per_iteration": 3.6122052669525146
    },
    {
      "auxiliary_loss_clip": 0.01167303,
      "auxiliary_loss_mlp": 0.01018261,
      "balance_loss_clip": 1.04361618,
      "balance_loss_mlp": 1.01097715,
      "epoch": 0.7962484218120603,
      "flos": 21944436612480.0,
      "grad_norm": 1.8718444140011798,
      "language_loss": 0.77811873,
      "learning_rate": 4.1975681880837023e-07,
      "loss": 0.79997438,
      "num_input_tokens_seen": 142642495,
      "step": 6622,
      "time_per_iteration": 2.616297960281372
    },
    {
      "auxiliary_loss_clip": 0.01169029,
      "auxiliary_loss_mlp": 0.01022273,
      "balance_loss_clip": 1.03956532,
      "balance_loss_mlp": 1.01487017,
      "epoch": 0.7963686647026994,
      "flos": 18876264687360.0,
      "grad_norm": 1.6493124932063732,
      "language_loss": 0.83046436,
      "learning_rate": 4.192794679022895e-07,
      "loss": 0.8523773,
      "num_input_tokens_seen": 142660820,
      "step": 6623,
      "time_per_iteration": 2.620328664779663
    },
    {
      "auxiliary_loss_clip": 0.01171071,
      "auxiliary_loss_mlp": 0.01028134,
      "balance_loss_clip": 1.04541242,
      "balance_loss_mlp": 1.02137148,
      "epoch": 0.7964889075933386,
      "flos": 29716582763520.0,
      "grad_norm": 1.8641907784891392,
      "language_loss": 0.72136885,
      "learning_rate": 4.1880235679025743e-07,
      "loss": 0.74336088,
      "num_input_tokens_seen": 142680915,
      "step": 6624,
      "time_per_iteration": 2.7183454036712646
    },
    {
      "auxiliary_loss_clip": 0.01182172,
      "auxiliary_loss_mlp": 0.0103322,
      "balance_loss_clip": 1.04046059,
      "balance_loss_mlp": 1.0256238,
      "epoch": 0.7966091504839776,
      "flos": 29491458272640.0,
      "grad_norm": 2.304502245373731,
      "language_loss": 0.63232148,
      "learning_rate": 4.1832548554464986e-07,
      "loss": 0.65447545,
      "num_input_tokens_seen": 142699210,
      "step": 6625,
      "time_per_iteration": 2.7528417110443115
    },
    {
      "auxiliary_loss_clip": 0.0106771,
      "auxiliary_loss_mlp": 0.01000125,
      "balance_loss_clip": 1.01142335,
      "balance_loss_mlp": 0.99931413,
      "epoch": 0.7967293933746167,
      "flos": 67288697101440.0,
      "grad_norm": 0.7380233387743003,
      "language_loss": 0.58642507,
      "learning_rate": 4.178488542378098e-07,
      "loss": 0.60710341,
      "num_input_tokens_seen": 142756790,
      "step": 6626,
      "time_per_iteration": 3.104029893875122
    },
    {
      "auxiliary_loss_clip": 0.01171593,
      "auxiliary_loss_mlp": 0.01027401,
      "balance_loss_clip": 1.04958367,
      "balance_loss_mlp": 1.01967609,
      "epoch": 0.7968496362652558,
      "flos": 25554679660800.0,
      "grad_norm": 1.9808375826874989,
      "language_loss": 0.88912964,
      "learning_rate": 4.173724629420401e-07,
      "loss": 0.91111958,
      "num_input_tokens_seen": 142778150,
      "step": 6627,
      "time_per_iteration": 3.61665415763855
    },
    {
      "auxiliary_loss_clip": 0.01180512,
      "auxiliary_loss_mlp": 0.0102644,
      "balance_loss_clip": 1.04722655,
      "balance_loss_mlp": 1.01851857,
      "epoch": 0.7969698791558949,
      "flos": 14501088581760.0,
      "grad_norm": 2.9726816498657076,
      "language_loss": 0.68735731,
      "learning_rate": 4.168963117296087e-07,
      "loss": 0.70942688,
      "num_input_tokens_seen": 142795485,
      "step": 6628,
      "time_per_iteration": 2.5806479454040527
    },
    {
      "auxiliary_loss_clip": 0.0116718,
      "auxiliary_loss_mlp": 0.01019932,
      "balance_loss_clip": 1.04818583,
      "balance_loss_mlp": 1.01262462,
      "epoch": 0.797090122046534,
      "flos": 22127545169280.0,
      "grad_norm": 2.5345445411444687,
      "language_loss": 0.75929898,
      "learning_rate": 4.1642040067274876e-07,
      "loss": 0.78117007,
      "num_input_tokens_seen": 142815155,
      "step": 6629,
      "time_per_iteration": 2.639376640319824
    },
    {
      "auxiliary_loss_clip": 0.01175304,
      "auxiliary_loss_mlp": 0.01020875,
      "balance_loss_clip": 1.04517651,
      "balance_loss_mlp": 1.01342452,
      "epoch": 0.7972103649371731,
      "flos": 19897671830400.0,
      "grad_norm": 1.6013157500257882,
      "language_loss": 0.72626853,
      "learning_rate": 4.1594472984365493e-07,
      "loss": 0.74823034,
      "num_input_tokens_seen": 142833840,
      "step": 6630,
      "time_per_iteration": 3.51484751701355
    },
    {
      "auxiliary_loss_clip": 0.01163143,
      "auxiliary_loss_mlp": 0.01025529,
      "balance_loss_clip": 1.04430163,
      "balance_loss_mlp": 1.01846862,
      "epoch": 0.7973306078278122,
      "flos": 36058621847040.0,
      "grad_norm": 2.2278546532870362,
      "language_loss": 0.77715528,
      "learning_rate": 4.154692993144862e-07,
      "loss": 0.79904199,
      "num_input_tokens_seen": 142853610,
      "step": 6631,
      "time_per_iteration": 2.7208378314971924
    },
    {
      "auxiliary_loss_clip": 0.01166241,
      "auxiliary_loss_mlp": 0.01208017,
      "balance_loss_clip": 1.047472,
      "balance_loss_mlp": 1.00070679,
      "epoch": 0.7974508507184512,
      "flos": 21360600950400.0,
      "grad_norm": 2.138271106402303,
      "language_loss": 0.71523112,
      "learning_rate": 4.1499410915736476e-07,
      "loss": 0.73897374,
      "num_input_tokens_seen": 142872540,
      "step": 6632,
      "time_per_iteration": 2.6432478427886963
    },
    {
      "auxiliary_loss_clip": 0.01071576,
      "auxiliary_loss_mlp": 0.01000811,
      "balance_loss_clip": 1.01128888,
      "balance_loss_mlp": 0.99997616,
      "epoch": 0.7975710936090904,
      "flos": 68253115317120.0,
      "grad_norm": 0.7686151328000085,
      "language_loss": 0.64291728,
      "learning_rate": 4.145191594443762e-07,
      "loss": 0.66364115,
      "num_input_tokens_seen": 142936895,
      "step": 6633,
      "time_per_iteration": 3.3491098880767822
    },
    {
      "auxiliary_loss_clip": 0.01169151,
      "auxiliary_loss_mlp": 0.01026443,
      "balance_loss_clip": 1.04206848,
      "balance_loss_mlp": 1.01884925,
      "epoch": 0.7976913364997295,
      "flos": 22492433479680.0,
      "grad_norm": 2.035483932554791,
      "language_loss": 0.70770478,
      "learning_rate": 4.140444502475713e-07,
      "loss": 0.72966069,
      "num_input_tokens_seen": 142956445,
      "step": 6634,
      "time_per_iteration": 2.704519748687744
    },
    {
      "auxiliary_loss_clip": 0.01165309,
      "auxiliary_loss_mlp": 0.01028591,
      "balance_loss_clip": 1.04375625,
      "balance_loss_mlp": 1.02113438,
      "epoch": 0.7978115793903685,
      "flos": 15263220378240.0,
      "grad_norm": 2.8113886641527404,
      "language_loss": 0.70450318,
      "learning_rate": 4.1356998163896216e-07,
      "loss": 0.72644222,
      "num_input_tokens_seen": 142973495,
      "step": 6635,
      "time_per_iteration": 3.4839236736297607
    },
    {
      "auxiliary_loss_clip": 0.01178303,
      "auxiliary_loss_mlp": 0.01022502,
      "balance_loss_clip": 1.04459548,
      "balance_loss_mlp": 1.01521802,
      "epoch": 0.7979318222810077,
      "flos": 19719232041600.0,
      "grad_norm": 2.066235442353944,
      "language_loss": 0.75179827,
      "learning_rate": 4.130957536905255e-07,
      "loss": 0.77380633,
      "num_input_tokens_seen": 142991510,
      "step": 6636,
      "time_per_iteration": 2.6418087482452393
    },
    {
      "auxiliary_loss_clip": 0.01182109,
      "auxiliary_loss_mlp": 0.0102925,
      "balance_loss_clip": 1.04633534,
      "balance_loss_mlp": 1.02103686,
      "epoch": 0.7980520651716467,
      "flos": 15560273854080.0,
      "grad_norm": 2.5246249057617463,
      "language_loss": 0.70659113,
      "learning_rate": 4.1262176647420134e-07,
      "loss": 0.72870469,
      "num_input_tokens_seen": 143009675,
      "step": 6637,
      "time_per_iteration": 2.602303981781006
    },
    {
      "auxiliary_loss_clip": 0.01180237,
      "auxiliary_loss_mlp": 0.01025572,
      "balance_loss_clip": 1.04777718,
      "balance_loss_mlp": 1.01799679,
      "epoch": 0.7981723080622858,
      "flos": 22309432663680.0,
      "grad_norm": 1.732063233592969,
      "language_loss": 0.80011177,
      "learning_rate": 4.121480200618923e-07,
      "loss": 0.8221699,
      "num_input_tokens_seen": 143029330,
      "step": 6638,
      "time_per_iteration": 2.6594483852386475
    },
    {
      "auxiliary_loss_clip": 0.01168797,
      "auxiliary_loss_mlp": 0.0102301,
      "balance_loss_clip": 1.04506207,
      "balance_loss_mlp": 1.01592314,
      "epoch": 0.798292550952925,
      "flos": 22929573997440.0,
      "grad_norm": 1.9362486735165148,
      "language_loss": 0.8020848,
      "learning_rate": 4.116745145254674e-07,
      "loss": 0.82400286,
      "num_input_tokens_seen": 143048865,
      "step": 6639,
      "time_per_iteration": 2.634941816329956
    },
    {
      "auxiliary_loss_clip": 0.01073636,
      "auxiliary_loss_mlp": 0.01003026,
      "balance_loss_clip": 1.01162291,
      "balance_loss_mlp": 1.00221527,
      "epoch": 0.798412793843564,
      "flos": 64497936890880.0,
      "grad_norm": 0.7609698565069287,
      "language_loss": 0.57930744,
      "learning_rate": 4.1120124993675476e-07,
      "loss": 0.60007405,
      "num_input_tokens_seen": 143113295,
      "step": 6640,
      "time_per_iteration": 3.182257652282715
    },
    {
      "auxiliary_loss_clip": 0.0118273,
      "auxiliary_loss_mlp": 0.0102465,
      "balance_loss_clip": 1.04645312,
      "balance_loss_mlp": 1.01631975,
      "epoch": 0.7985330367342031,
      "flos": 13586910514560.0,
      "grad_norm": 4.959646573123011,
      "language_loss": 0.62355638,
      "learning_rate": 4.107282263675498e-07,
      "loss": 0.64563012,
      "num_input_tokens_seen": 143130965,
      "step": 6641,
      "time_per_iteration": 2.7472054958343506
    },
    {
      "auxiliary_loss_clip": 0.01073848,
      "auxiliary_loss_mlp": 0.01198476,
      "balance_loss_clip": 1.01368999,
      "balance_loss_mlp": 0.9999426,
      "epoch": 0.7986532796248422,
      "flos": 67698797656320.0,
      "grad_norm": 0.7604969243913271,
      "language_loss": 0.52466059,
      "learning_rate": 4.1025544388960907e-07,
      "loss": 0.54738379,
      "num_input_tokens_seen": 143192005,
      "step": 6642,
      "time_per_iteration": 3.131976366043091
    },
    {
      "auxiliary_loss_clip": 0.01170118,
      "auxiliary_loss_mlp": 0.01025588,
      "balance_loss_clip": 1.04786158,
      "balance_loss_mlp": 1.01808953,
      "epoch": 0.7987735225154813,
      "flos": 22455373622400.0,
      "grad_norm": 3.3684874126302544,
      "language_loss": 0.71842849,
      "learning_rate": 4.097829025746538e-07,
      "loss": 0.74038553,
      "num_input_tokens_seen": 143213550,
      "step": 6643,
      "time_per_iteration": 2.7617835998535156
    },
    {
      "auxiliary_loss_clip": 0.01070036,
      "auxiliary_loss_mlp": 0.0100034,
      "balance_loss_clip": 1.01129365,
      "balance_loss_mlp": 0.9995594,
      "epoch": 0.7988937654061203,
      "flos": 68864098682880.0,
      "grad_norm": 0.655460678290196,
      "language_loss": 0.60976446,
      "learning_rate": 4.0931060249436757e-07,
      "loss": 0.63046813,
      "num_input_tokens_seen": 143277390,
      "step": 6644,
      "time_per_iteration": 3.151561737060547
    },
    {
      "auxiliary_loss_clip": 0.01169832,
      "auxiliary_loss_mlp": 0.0102814,
      "balance_loss_clip": 1.04815626,
      "balance_loss_mlp": 1.02023089,
      "epoch": 0.7990140082967595,
      "flos": 20806893820800.0,
      "grad_norm": 2.3267078022861596,
      "language_loss": 0.69585401,
      "learning_rate": 4.088385437203978e-07,
      "loss": 0.71783376,
      "num_input_tokens_seen": 143294400,
      "step": 6645,
      "time_per_iteration": 2.641932725906372
    },
    {
      "auxiliary_loss_clip": 0.01167284,
      "auxiliary_loss_mlp": 0.01024405,
      "balance_loss_clip": 1.04619443,
      "balance_loss_mlp": 1.01694798,
      "epoch": 0.7991342511873986,
      "flos": 18985289443200.0,
      "grad_norm": 2.645858259939203,
      "language_loss": 0.77596235,
      "learning_rate": 4.083667263243564e-07,
      "loss": 0.79787922,
      "num_input_tokens_seen": 143312745,
      "step": 6646,
      "time_per_iteration": 2.6484313011169434
    },
    {
      "auxiliary_loss_clip": 0.01167271,
      "auxiliary_loss_mlp": 0.01026982,
      "balance_loss_clip": 1.04795206,
      "balance_loss_mlp": 1.01986217,
      "epoch": 0.7992544940780376,
      "flos": 20816805974400.0,
      "grad_norm": 1.6103218979034026,
      "language_loss": 0.71557283,
      "learning_rate": 4.0789515037781653e-07,
      "loss": 0.73751533,
      "num_input_tokens_seen": 143333470,
      "step": 6647,
      "time_per_iteration": 2.5843517780303955
    },
    {
      "auxiliary_loss_clip": 0.01172739,
      "auxiliary_loss_mlp": 0.01020604,
      "balance_loss_clip": 1.04655635,
      "balance_loss_mlp": 1.0132246,
      "epoch": 0.7993747369686768,
      "flos": 12640772321280.0,
      "grad_norm": 2.0903487769512488,
      "language_loss": 0.82870108,
      "learning_rate": 4.0742381595231755e-07,
      "loss": 0.85063452,
      "num_input_tokens_seen": 143350195,
      "step": 6648,
      "time_per_iteration": 3.5173511505126953
    },
    {
      "auxiliary_loss_clip": 0.01181452,
      "auxiliary_loss_mlp": 0.01026494,
      "balance_loss_clip": 1.0466032,
      "balance_loss_mlp": 1.01954401,
      "epoch": 0.7994949798593158,
      "flos": 20078769225600.0,
      "grad_norm": 1.6661530469909216,
      "language_loss": 0.78344792,
      "learning_rate": 4.06952723119359e-07,
      "loss": 0.80552739,
      "num_input_tokens_seen": 143370070,
      "step": 6649,
      "time_per_iteration": 2.6593544483184814
    },
    {
      "auxiliary_loss_clip": 0.01163898,
      "auxiliary_loss_mlp": 0.01025303,
      "balance_loss_clip": 1.04369879,
      "balance_loss_mlp": 1.01760805,
      "epoch": 0.7996152227499549,
      "flos": 38654209509120.0,
      "grad_norm": 2.0811055740263082,
      "language_loss": 0.67345226,
      "learning_rate": 4.0648187195040504e-07,
      "loss": 0.69534427,
      "num_input_tokens_seen": 143392275,
      "step": 6650,
      "time_per_iteration": 2.7000927925109863
    },
    {
      "auxiliary_loss_clip": 0.01067451,
      "auxiliary_loss_mlp": 0.0100021,
      "balance_loss_clip": 1.01114905,
      "balance_loss_mlp": 0.99939293,
      "epoch": 0.799735465640594,
      "flos": 70243821947520.0,
      "grad_norm": 0.818559791052477,
      "language_loss": 0.67486107,
      "learning_rate": 4.060112625168848e-07,
      "loss": 0.69553769,
      "num_input_tokens_seen": 143457385,
      "step": 6651,
      "time_per_iteration": 3.2490696907043457
    },
    {
      "auxiliary_loss_clip": 0.01168569,
      "auxiliary_loss_mlp": 0.01027249,
      "balance_loss_clip": 1.04878497,
      "balance_loss_mlp": 1.01920223,
      "epoch": 0.7998557085312331,
      "flos": 24240995550720.0,
      "grad_norm": 1.8929300545257433,
      "language_loss": 0.74086303,
      "learning_rate": 4.055408948901886e-07,
      "loss": 0.7628212,
      "num_input_tokens_seen": 143478785,
      "step": 6652,
      "time_per_iteration": 2.5839457511901855
    },
    {
      "auxiliary_loss_clip": 0.01174425,
      "auxiliary_loss_mlp": 0.01023238,
      "balance_loss_clip": 1.04728389,
      "balance_loss_mlp": 1.01477456,
      "epoch": 0.7999759514218722,
      "flos": 27564025449600.0,
      "grad_norm": 1.6751547061017278,
      "language_loss": 0.71137571,
      "learning_rate": 4.050707691416708e-07,
      "loss": 0.73335242,
      "num_input_tokens_seen": 143500095,
      "step": 6653,
      "time_per_iteration": 2.68112850189209
    },
    {
      "auxiliary_loss_clip": 0.01067737,
      "auxiliary_loss_mlp": 0.01000138,
      "balance_loss_clip": 1.01134086,
      "balance_loss_mlp": 0.99932164,
      "epoch": 0.8000961943125112,
      "flos": 67337428878720.0,
      "grad_norm": 0.6685941831908978,
      "language_loss": 0.59693611,
      "learning_rate": 4.046008853426495e-07,
      "loss": 0.61761487,
      "num_input_tokens_seen": 143563410,
      "step": 6654,
      "time_per_iteration": 4.202452898025513
    },
    {
      "auxiliary_loss_clip": 0.01170894,
      "auxiliary_loss_mlp": 0.01026887,
      "balance_loss_clip": 1.04262161,
      "balance_loss_mlp": 1.01907825,
      "epoch": 0.8002164372031504,
      "flos": 28733815676160.0,
      "grad_norm": 1.9201417560457137,
      "language_loss": 0.62899488,
      "learning_rate": 4.0413124356440464e-07,
      "loss": 0.65097266,
      "num_input_tokens_seen": 143587455,
      "step": 6655,
      "time_per_iteration": 2.7420694828033447
    },
    {
      "auxiliary_loss_clip": 0.01179085,
      "auxiliary_loss_mlp": 0.01023839,
      "balance_loss_clip": 1.04008126,
      "balance_loss_mlp": 1.01657319,
      "epoch": 0.8003366800937894,
      "flos": 17639429725440.0,
      "grad_norm": 2.0354644477897392,
      "language_loss": 0.82352239,
      "learning_rate": 4.0366184387818223e-07,
      "loss": 0.84555161,
      "num_input_tokens_seen": 143605915,
      "step": 6656,
      "time_per_iteration": 2.6752591133117676
    },
    {
      "auxiliary_loss_clip": 0.011737,
      "auxiliary_loss_mlp": 0.01023145,
      "balance_loss_clip": 1.04936481,
      "balance_loss_mlp": 1.01487839,
      "epoch": 0.8004569229844285,
      "flos": 25995303797760.0,
      "grad_norm": 1.764393942215559,
      "language_loss": 0.85110384,
      "learning_rate": 4.0319268635518797e-07,
      "loss": 0.87307233,
      "num_input_tokens_seen": 143626490,
      "step": 6657,
      "time_per_iteration": 2.6789233684539795
    },
    {
      "auxiliary_loss_clip": 0.01168419,
      "auxiliary_loss_mlp": 0.01021621,
      "balance_loss_clip": 1.04515898,
      "balance_loss_mlp": 1.01435208,
      "epoch": 0.8005771658750677,
      "flos": 20812352688000.0,
      "grad_norm": 1.642635576196585,
      "language_loss": 0.74995285,
      "learning_rate": 4.027237710665943e-07,
      "loss": 0.77185327,
      "num_input_tokens_seen": 143644955,
      "step": 6658,
      "time_per_iteration": 2.8378477096557617
    },
    {
      "auxiliary_loss_clip": 0.0117741,
      "auxiliary_loss_mlp": 0.01023834,
      "balance_loss_clip": 1.04116344,
      "balance_loss_mlp": 1.01601946,
      "epoch": 0.8006974087657067,
      "flos": 25812626204160.0,
      "grad_norm": 2.199453702486151,
      "language_loss": 0.69215387,
      "learning_rate": 4.022550980835344e-07,
      "loss": 0.71416628,
      "num_input_tokens_seen": 143667200,
      "step": 6659,
      "time_per_iteration": 2.75274920463562
    },
    {
      "auxiliary_loss_clip": 0.01172609,
      "auxiliary_loss_mlp": 0.01021126,
      "balance_loss_clip": 1.04003298,
      "balance_loss_mlp": 1.01379418,
      "epoch": 0.8008176516563458,
      "flos": 17164690646400.0,
      "grad_norm": 2.1044062432283708,
      "language_loss": 0.79666471,
      "learning_rate": 4.017866674771051e-07,
      "loss": 0.81860209,
      "num_input_tokens_seen": 143684685,
      "step": 6660,
      "time_per_iteration": 2.669738531112671
    },
    {
      "auxiliary_loss_clip": 0.01167643,
      "auxiliary_loss_mlp": 0.01029158,
      "balance_loss_clip": 1.03785503,
      "balance_loss_mlp": 1.02120996,
      "epoch": 0.8009378945469849,
      "flos": 24207311571840.0,
      "grad_norm": 1.6900629468556327,
      "language_loss": 0.74430847,
      "learning_rate": 4.013184793183688e-07,
      "loss": 0.76627648,
      "num_input_tokens_seen": 143706780,
      "step": 6661,
      "time_per_iteration": 3.706589460372925
    },
    {
      "auxiliary_loss_clip": 0.01167005,
      "auxiliary_loss_mlp": 0.01024611,
      "balance_loss_clip": 1.04365921,
      "balance_loss_mlp": 1.01732707,
      "epoch": 0.801058137437624,
      "flos": 19787318271360.0,
      "grad_norm": 1.8592910095006858,
      "language_loss": 0.72699189,
      "learning_rate": 4.008505336783472e-07,
      "loss": 0.7489081,
      "num_input_tokens_seen": 143724505,
      "step": 6662,
      "time_per_iteration": 2.647056818008423
    },
    {
      "auxiliary_loss_clip": 0.01159346,
      "auxiliary_loss_mlp": 0.01024514,
      "balance_loss_clip": 1.04327309,
      "balance_loss_mlp": 1.01738548,
      "epoch": 0.801178380328263,
      "flos": 18659400324480.0,
      "grad_norm": 1.884097022871924,
      "language_loss": 0.80826128,
      "learning_rate": 4.003828306280284e-07,
      "loss": 0.83009982,
      "num_input_tokens_seen": 143742180,
      "step": 6663,
      "time_per_iteration": 2.6322929859161377
    },
    {
      "auxiliary_loss_clip": 0.01170668,
      "auxiliary_loss_mlp": 0.01024332,
      "balance_loss_clip": 1.04613495,
      "balance_loss_mlp": 1.01750135,
      "epoch": 0.8012986232189022,
      "flos": 15706573948800.0,
      "grad_norm": 2.0838782414049026,
      "language_loss": 0.77891409,
      "learning_rate": 3.999153702383626e-07,
      "loss": 0.8008641,
      "num_input_tokens_seen": 143760070,
      "step": 6664,
      "time_per_iteration": 2.6146252155303955
    },
    {
      "auxiliary_loss_clip": 0.01171876,
      "auxiliary_loss_mlp": 0.01024101,
      "balance_loss_clip": 1.04541206,
      "balance_loss_mlp": 1.01604283,
      "epoch": 0.8014188661095413,
      "flos": 28584139703040.0,
      "grad_norm": 2.862135946914504,
      "language_loss": 0.74023259,
      "learning_rate": 3.9944815258026263e-07,
      "loss": 0.76219243,
      "num_input_tokens_seen": 143781890,
      "step": 6665,
      "time_per_iteration": 2.6942405700683594
    },
    {
      "auxiliary_loss_clip": 0.01175063,
      "auxiliary_loss_mlp": 0.01025061,
      "balance_loss_clip": 1.04862213,
      "balance_loss_mlp": 1.01712799,
      "epoch": 0.8015391090001803,
      "flos": 29310360877440.0,
      "grad_norm": 3.074918458802541,
      "language_loss": 0.83123368,
      "learning_rate": 3.989811777246057e-07,
      "loss": 0.85323495,
      "num_input_tokens_seen": 143802060,
      "step": 6666,
      "time_per_iteration": 2.697521209716797
    },
    {
      "auxiliary_loss_clip": 0.01064028,
      "auxiliary_loss_mlp": 0.01001688,
      "balance_loss_clip": 1.01091361,
      "balance_loss_mlp": 1.00088942,
      "epoch": 0.8016593518908195,
      "flos": 70397340675840.0,
      "grad_norm": 0.8442553520694063,
      "language_loss": 0.66144049,
      "learning_rate": 3.985144457422305e-07,
      "loss": 0.68209755,
      "num_input_tokens_seen": 143856345,
      "step": 6667,
      "time_per_iteration": 3.1265206336975098
    },
    {
      "auxiliary_loss_clip": 0.01169864,
      "auxiliary_loss_mlp": 0.01023192,
      "balance_loss_clip": 1.04961801,
      "balance_loss_mlp": 1.01550293,
      "epoch": 0.8017795947814585,
      "flos": 26026114688640.0,
      "grad_norm": 2.4363919377022025,
      "language_loss": 0.77100813,
      "learning_rate": 3.9804795670394096e-07,
      "loss": 0.79293871,
      "num_input_tokens_seen": 143876470,
      "step": 6668,
      "time_per_iteration": 2.6564881801605225
    },
    {
      "auxiliary_loss_clip": 0.01165204,
      "auxiliary_loss_mlp": 0.01025864,
      "balance_loss_clip": 1.04408383,
      "balance_loss_mlp": 1.0183953,
      "epoch": 0.8018998376720976,
      "flos": 22087181260800.0,
      "grad_norm": 1.862538221681838,
      "language_loss": 0.70643735,
      "learning_rate": 3.975817106805022e-07,
      "loss": 0.72834802,
      "num_input_tokens_seen": 143895170,
      "step": 6669,
      "time_per_iteration": 2.7202157974243164
    },
    {
      "auxiliary_loss_clip": 0.01177827,
      "auxiliary_loss_mlp": 0.01030415,
      "balance_loss_clip": 1.04472685,
      "balance_loss_mlp": 1.02258325,
      "epoch": 0.8020200805627368,
      "flos": 34568545023360.0,
      "grad_norm": 2.077930811067231,
      "language_loss": 0.65286022,
      "learning_rate": 3.97115707742645e-07,
      "loss": 0.67494261,
      "num_input_tokens_seen": 143915845,
      "step": 6670,
      "time_per_iteration": 2.7777140140533447
    },
    {
      "auxiliary_loss_clip": 0.01175089,
      "auxiliary_loss_mlp": 0.01022545,
      "balance_loss_clip": 1.0470891,
      "balance_loss_mlp": 1.01524639,
      "epoch": 0.8021403234533758,
      "flos": 20120354196480.0,
      "grad_norm": 2.012373775904294,
      "language_loss": 0.65360188,
      "learning_rate": 3.966499479610599e-07,
      "loss": 0.67557824,
      "num_input_tokens_seen": 143933940,
      "step": 6671,
      "time_per_iteration": 2.6281206607818604
    },
    {
      "auxiliary_loss_clip": 0.01171436,
      "auxiliary_loss_mlp": 0.01023408,
      "balance_loss_clip": 1.04426074,
      "balance_loss_mlp": 1.01620793,
      "epoch": 0.8022605663440149,
      "flos": 27746200252800.0,
      "grad_norm": 2.0157719926878297,
      "language_loss": 0.64709604,
      "learning_rate": 3.9618443140640225e-07,
      "loss": 0.66904438,
      "num_input_tokens_seen": 143952850,
      "step": 6672,
      "time_per_iteration": 2.7553040981292725
    },
    {
      "auxiliary_loss_clip": 0.01089934,
      "auxiliary_loss_mlp": 0.01001198,
      "balance_loss_clip": 1.01296306,
      "balance_loss_mlp": 1.00034523,
      "epoch": 0.802380809234654,
      "flos": 60244998768000.0,
      "grad_norm": 0.6853766389280874,
      "language_loss": 0.51244158,
      "learning_rate": 3.957191581492918e-07,
      "loss": 0.53335291,
      "num_input_tokens_seen": 144013610,
      "step": 6673,
      "time_per_iteration": 3.2018346786499023
    },
    {
      "auxiliary_loss_clip": 0.0117079,
      "auxiliary_loss_mlp": 0.01024111,
      "balance_loss_clip": 1.04507232,
      "balance_loss_mlp": 1.01613855,
      "epoch": 0.8025010521252931,
      "flos": 15080722352640.0,
      "grad_norm": 4.213059045351225,
      "language_loss": 0.70903277,
      "learning_rate": 3.952541282603097e-07,
      "loss": 0.73098177,
      "num_input_tokens_seen": 144028715,
      "step": 6674,
      "time_per_iteration": 3.5525569915771484
    },
    {
      "auxiliary_loss_clip": 0.01165665,
      "auxiliary_loss_mlp": 0.01023306,
      "balance_loss_clip": 1.04486227,
      "balance_loss_mlp": 1.01589084,
      "epoch": 0.8026212950159322,
      "flos": 22163527618560.0,
      "grad_norm": 2.9690543436395487,
      "language_loss": 0.83785051,
      "learning_rate": 3.9478934181000013e-07,
      "loss": 0.8597402,
      "num_input_tokens_seen": 144048740,
      "step": 6675,
      "time_per_iteration": 2.6163170337677
    },
    {
      "auxiliary_loss_clip": 0.01173719,
      "auxiliary_loss_mlp": 0.01023749,
      "balance_loss_clip": 1.04970789,
      "balance_loss_mlp": 1.01606596,
      "epoch": 0.8027415379065713,
      "flos": 17675986792320.0,
      "grad_norm": 2.5382594289670792,
      "language_loss": 0.84448349,
      "learning_rate": 3.943247988688714e-07,
      "loss": 0.86645818,
      "num_input_tokens_seen": 144067435,
      "step": 6676,
      "time_per_iteration": 2.5781869888305664
    },
    {
      "auxiliary_loss_clip": 0.01168864,
      "auxiliary_loss_mlp": 0.01021621,
      "balance_loss_clip": 1.0452826,
      "balance_loss_mlp": 1.01497197,
      "epoch": 0.8028617807972104,
      "flos": 21979593048960.0,
      "grad_norm": 1.7738929998321296,
      "language_loss": 0.72188848,
      "learning_rate": 3.938604995073933e-07,
      "loss": 0.74379331,
      "num_input_tokens_seen": 144085905,
      "step": 6677,
      "time_per_iteration": 2.6614320278167725
    },
    {
      "auxiliary_loss_clip": 0.01176056,
      "auxiliary_loss_mlp": 0.01024695,
      "balance_loss_clip": 1.04499221,
      "balance_loss_mlp": 1.01749444,
      "epoch": 0.8029820236878494,
      "flos": 26428457905920.0,
      "grad_norm": 1.6947088317755006,
      "language_loss": 0.65485889,
      "learning_rate": 3.9339644379600157e-07,
      "loss": 0.67686641,
      "num_input_tokens_seen": 144105735,
      "step": 6678,
      "time_per_iteration": 2.6466991901397705
    },
    {
      "auxiliary_loss_clip": 0.01175084,
      "auxiliary_loss_mlp": 0.01023807,
      "balance_loss_clip": 1.04779649,
      "balance_loss_mlp": 1.01651716,
      "epoch": 0.8031022665784886,
      "flos": 17676489582720.0,
      "grad_norm": 2.5217344045518297,
      "language_loss": 0.71389234,
      "learning_rate": 3.929326318050907e-07,
      "loss": 0.73588121,
      "num_input_tokens_seen": 144123405,
      "step": 6679,
      "time_per_iteration": 2.6939902305603027
    },
    {
      "auxiliary_loss_clip": 0.01164201,
      "auxiliary_loss_mlp": 0.01025236,
      "balance_loss_clip": 1.04551005,
      "balance_loss_mlp": 1.01794052,
      "epoch": 0.8032225094691277,
      "flos": 15450279431040.0,
      "grad_norm": 1.9015812093944282,
      "language_loss": 0.7900725,
      "learning_rate": 3.924690636050225e-07,
      "loss": 0.81196684,
      "num_input_tokens_seen": 144140815,
      "step": 6680,
      "time_per_iteration": 2.573732376098633
    },
    {
      "auxiliary_loss_clip": 0.01171618,
      "auxiliary_loss_mlp": 0.01027073,
      "balance_loss_clip": 1.04691291,
      "balance_loss_mlp": 1.01819754,
      "epoch": 0.8033427523597667,
      "flos": 26179202453760.0,
      "grad_norm": 2.2169440756792445,
      "language_loss": 0.73140836,
      "learning_rate": 3.9200573926611915e-07,
      "loss": 0.75339526,
      "num_input_tokens_seen": 144162230,
      "step": 6681,
      "time_per_iteration": 3.6922616958618164
    },
    {
      "auxiliary_loss_clip": 0.011678,
      "auxiliary_loss_mlp": 0.01024796,
      "balance_loss_clip": 1.04763389,
      "balance_loss_mlp": 1.01726174,
      "epoch": 0.8034629952504058,
      "flos": 21324905809920.0,
      "grad_norm": 2.837777906411097,
      "language_loss": 0.73002732,
      "learning_rate": 3.9154265885866613e-07,
      "loss": 0.75195324,
      "num_input_tokens_seen": 144181540,
      "step": 6682,
      "time_per_iteration": 2.581240177154541
    },
    {
      "auxiliary_loss_clip": 0.0116807,
      "auxiliary_loss_mlp": 0.0102396,
      "balance_loss_clip": 1.04689908,
      "balance_loss_mlp": 1.01565731,
      "epoch": 0.8035832381410449,
      "flos": 21651585027840.0,
      "grad_norm": 2.553420435545492,
      "language_loss": 0.74709928,
      "learning_rate": 3.9107982245291394e-07,
      "loss": 0.76901954,
      "num_input_tokens_seen": 144199665,
      "step": 6683,
      "time_per_iteration": 3.5670952796936035
    },
    {
      "auxiliary_loss_clip": 0.0117606,
      "auxiliary_loss_mlp": 0.01027463,
      "balance_loss_clip": 1.04519987,
      "balance_loss_mlp": 1.01944923,
      "epoch": 0.803703481031684,
      "flos": 20518818744960.0,
      "grad_norm": 1.9947594627757177,
      "language_loss": 0.77330422,
      "learning_rate": 3.9061723011907245e-07,
      "loss": 0.79533947,
      "num_input_tokens_seen": 144219020,
      "step": 6684,
      "time_per_iteration": 2.6679511070251465
    },
    {
      "auxiliary_loss_clip": 0.01173218,
      "auxiliary_loss_mlp": 0.01024309,
      "balance_loss_clip": 1.04440045,
      "balance_loss_mlp": 1.01684618,
      "epoch": 0.803823723922323,
      "flos": 22854807838080.0,
      "grad_norm": 2.1051304241758655,
      "language_loss": 0.79300791,
      "learning_rate": 3.901548819273179e-07,
      "loss": 0.81498325,
      "num_input_tokens_seen": 144239035,
      "step": 6685,
      "time_per_iteration": 2.5783941745758057
    },
    {
      "auxiliary_loss_clip": 0.01172627,
      "auxiliary_loss_mlp": 0.01026658,
      "balance_loss_clip": 1.04838204,
      "balance_loss_mlp": 1.01892161,
      "epoch": 0.8039439668129622,
      "flos": 21362145235200.0,
      "grad_norm": 2.2745210318324696,
      "language_loss": 0.69277436,
      "learning_rate": 3.896927779477881e-07,
      "loss": 0.71476722,
      "num_input_tokens_seen": 144258295,
      "step": 6686,
      "time_per_iteration": 2.643084764480591
    },
    {
      "auxiliary_loss_clip": 0.01174578,
      "auxiliary_loss_mlp": 0.0101944,
      "balance_loss_clip": 1.04249573,
      "balance_loss_mlp": 1.01147103,
      "epoch": 0.8040642097036013,
      "flos": 23802382575360.0,
      "grad_norm": 2.092374224740493,
      "language_loss": 0.67043507,
      "learning_rate": 3.892309182505833e-07,
      "loss": 0.6923753,
      "num_input_tokens_seen": 144276110,
      "step": 6687,
      "time_per_iteration": 2.6893680095672607
    },
    {
      "auxiliary_loss_clip": 0.01165891,
      "auxiliary_loss_mlp": 0.01022319,
      "balance_loss_clip": 1.04597771,
      "balance_loss_mlp": 1.01510668,
      "epoch": 0.8041844525942403,
      "flos": 25922046009600.0,
      "grad_norm": 2.427256921300212,
      "language_loss": 0.85879117,
      "learning_rate": 3.887693029057675e-07,
      "loss": 0.88067323,
      "num_input_tokens_seen": 144295620,
      "step": 6688,
      "time_per_iteration": 3.503814935684204
    },
    {
      "auxiliary_loss_clip": 0.01171452,
      "auxiliary_loss_mlp": 0.01022879,
      "balance_loss_clip": 1.04465675,
      "balance_loss_mlp": 1.0156405,
      "epoch": 0.8043046954848795,
      "flos": 25191120153600.0,
      "grad_norm": 1.7787689214081288,
      "language_loss": 0.8152265,
      "learning_rate": 3.8830793198336684e-07,
      "loss": 0.83716983,
      "num_input_tokens_seen": 144315210,
      "step": 6689,
      "time_per_iteration": 2.6322500705718994
    },
    {
      "auxiliary_loss_clip": 0.01175339,
      "auxiliary_loss_mlp": 0.01027939,
      "balance_loss_clip": 1.04603851,
      "balance_loss_mlp": 1.02063131,
      "epoch": 0.8044249383755185,
      "flos": 41719185123840.0,
      "grad_norm": 2.391135836564404,
      "language_loss": 0.70352507,
      "learning_rate": 3.878468055533721e-07,
      "loss": 0.72555786,
      "num_input_tokens_seen": 144337750,
      "step": 6690,
      "time_per_iteration": 2.8096349239349365
    },
    {
      "auxiliary_loss_clip": 0.01180913,
      "auxiliary_loss_mlp": 0.0102902,
      "balance_loss_clip": 1.04553127,
      "balance_loss_mlp": 1.0211699,
      "epoch": 0.8045451812661576,
      "flos": 20631434860800.0,
      "grad_norm": 3.112036154433585,
      "language_loss": 0.84251046,
      "learning_rate": 3.8738592368573464e-07,
      "loss": 0.86460984,
      "num_input_tokens_seen": 144355305,
      "step": 6691,
      "time_per_iteration": 2.6784749031066895
    },
    {
      "auxiliary_loss_clip": 0.01166128,
      "auxiliary_loss_mlp": 0.01024706,
      "balance_loss_clip": 1.04310536,
      "balance_loss_mlp": 1.01699352,
      "epoch": 0.8046654241567968,
      "flos": 29711806254720.0,
      "grad_norm": 2.1052675354230383,
      "language_loss": 0.87999701,
      "learning_rate": 3.8692528645037137e-07,
      "loss": 0.9019053,
      "num_input_tokens_seen": 144374485,
      "step": 6692,
      "time_per_iteration": 2.7116668224334717
    },
    {
      "auxiliary_loss_clip": 0.01167503,
      "auxiliary_loss_mlp": 0.01026558,
      "balance_loss_clip": 1.04808044,
      "balance_loss_mlp": 1.01938176,
      "epoch": 0.8047856670474358,
      "flos": 17671389851520.0,
      "grad_norm": 2.178436406941529,
      "language_loss": 0.78039002,
      "learning_rate": 3.8646489391715907e-07,
      "loss": 0.80233061,
      "num_input_tokens_seen": 144388780,
      "step": 6693,
      "time_per_iteration": 2.668724298477173
    },
    {
      "auxiliary_loss_clip": 0.01170959,
      "auxiliary_loss_mlp": 0.01026911,
      "balance_loss_clip": 1.04431367,
      "balance_loss_mlp": 1.01937151,
      "epoch": 0.8049059099380749,
      "flos": 17120699464320.0,
      "grad_norm": 2.4457288389702203,
      "language_loss": 0.88210469,
      "learning_rate": 3.8600474615593903e-07,
      "loss": 0.90408343,
      "num_input_tokens_seen": 144403395,
      "step": 6694,
      "time_per_iteration": 2.609631061553955
    },
    {
      "auxiliary_loss_clip": 0.0108443,
      "auxiliary_loss_mlp": 0.01000563,
      "balance_loss_clip": 1.01157713,
      "balance_loss_mlp": 0.99975204,
      "epoch": 0.805026152828714,
      "flos": 62212903240320.0,
      "grad_norm": 0.7912457433807089,
      "language_loss": 0.59641522,
      "learning_rate": 3.8554484323651605e-07,
      "loss": 0.61726516,
      "num_input_tokens_seen": 144465265,
      "step": 6695,
      "time_per_iteration": 3.293560028076172
    },
    {
      "auxiliary_loss_clip": 0.01170326,
      "auxiliary_loss_mlp": 0.01207843,
      "balance_loss_clip": 1.04816437,
      "balance_loss_mlp": 1.00071192,
      "epoch": 0.8051463957193531,
      "flos": 21688608971520.0,
      "grad_norm": 1.8718514930570807,
      "language_loss": 0.79552543,
      "learning_rate": 3.85085185228657e-07,
      "loss": 0.81930709,
      "num_input_tokens_seen": 144484235,
      "step": 6696,
      "time_per_iteration": 2.6334495544433594
    },
    {
      "auxiliary_loss_clip": 0.01169121,
      "auxiliary_loss_mlp": 0.01030883,
      "balance_loss_clip": 1.04487634,
      "balance_loss_mlp": 1.02316475,
      "epoch": 0.8052666386099921,
      "flos": 32051458535040.0,
      "grad_norm": 3.4128771981304613,
      "language_loss": 0.73386717,
      "learning_rate": 3.8462577220209114e-07,
      "loss": 0.75586724,
      "num_input_tokens_seen": 144504610,
      "step": 6697,
      "time_per_iteration": 2.6970648765563965
    },
    {
      "auxiliary_loss_clip": 0.01064852,
      "auxiliary_loss_mlp": 0.01002298,
      "balance_loss_clip": 1.01172471,
      "balance_loss_mlp": 1.00148141,
      "epoch": 0.8053868815006313,
      "flos": 67157875768320.0,
      "grad_norm": 0.7073321430997962,
      "language_loss": 0.58983165,
      "learning_rate": 3.8416660422651127e-07,
      "loss": 0.6105032,
      "num_input_tokens_seen": 144574260,
      "step": 6698,
      "time_per_iteration": 3.332368850708008
    },
    {
      "auxiliary_loss_clip": 0.0118021,
      "auxiliary_loss_mlp": 0.01028575,
      "balance_loss_clip": 1.04375303,
      "balance_loss_mlp": 1.02049828,
      "epoch": 0.8055071243912704,
      "flos": 23837000307840.0,
      "grad_norm": 9.440703180429878,
      "language_loss": 0.6818819,
      "learning_rate": 3.837076813715723e-07,
      "loss": 0.70396978,
      "num_input_tokens_seen": 144594145,
      "step": 6699,
      "time_per_iteration": 2.687457799911499
    },
    {
      "auxiliary_loss_clip": 0.01173668,
      "auxiliary_loss_mlp": 0.01022916,
      "balance_loss_clip": 1.04185569,
      "balance_loss_mlp": 1.01507843,
      "epoch": 0.8056273672819094,
      "flos": 21324510760320.0,
      "grad_norm": 4.06028403393629,
      "language_loss": 0.75630307,
      "learning_rate": 3.832490037068941e-07,
      "loss": 0.77826893,
      "num_input_tokens_seen": 144612935,
      "step": 6700,
      "time_per_iteration": 2.7430267333984375
    },
    {
      "auxiliary_loss_clip": 0.01176984,
      "auxiliary_loss_mlp": 0.01022884,
      "balance_loss_clip": 1.0395422,
      "balance_loss_mlp": 1.0155344,
      "epoch": 0.8057476101725486,
      "flos": 25768383626880.0,
      "grad_norm": 2.1824109694091627,
      "language_loss": 0.76577008,
      "learning_rate": 3.827905713020554e-07,
      "loss": 0.78776872,
      "num_input_tokens_seen": 144630580,
      "step": 6701,
      "time_per_iteration": 3.68807315826416
    },
    {
      "auxiliary_loss_clip": 0.01178881,
      "auxiliary_loss_mlp": 0.01029572,
      "balance_loss_clip": 1.04194903,
      "balance_loss_mlp": 1.02107835,
      "epoch": 0.8058678530631876,
      "flos": 24535283679360.0,
      "grad_norm": 2.0981102449719127,
      "language_loss": 0.69052368,
      "learning_rate": 3.823323842266017e-07,
      "loss": 0.71260822,
      "num_input_tokens_seen": 144649975,
      "step": 6702,
      "time_per_iteration": 2.723151922225952
    },
    {
      "auxiliary_loss_clip": 0.0117081,
      "auxiliary_loss_mlp": 0.01025369,
      "balance_loss_clip": 1.04358649,
      "balance_loss_mlp": 1.01777554,
      "epoch": 0.8059880959538267,
      "flos": 24753728240640.0,
      "grad_norm": 7.205815095717573,
      "language_loss": 0.72633713,
      "learning_rate": 3.818744425500393e-07,
      "loss": 0.74829888,
      "num_input_tokens_seen": 144667990,
      "step": 6703,
      "time_per_iteration": 2.637205123901367
    },
    {
      "auxiliary_loss_clip": 0.01169751,
      "auxiliary_loss_mlp": 0.01027511,
      "balance_loss_clip": 1.04137564,
      "balance_loss_mlp": 1.01865339,
      "epoch": 0.8061083388444659,
      "flos": 22196349671040.0,
      "grad_norm": 2.274548330485803,
      "language_loss": 0.80518436,
      "learning_rate": 3.8141674634183675e-07,
      "loss": 0.82715702,
      "num_input_tokens_seen": 144687020,
      "step": 6704,
      "time_per_iteration": 2.706831693649292
    },
    {
      "auxiliary_loss_clip": 0.01176219,
      "auxiliary_loss_mlp": 0.01022077,
      "balance_loss_clip": 1.04328322,
      "balance_loss_mlp": 1.01491547,
      "epoch": 0.8062285817351049,
      "flos": 30044195735040.0,
      "grad_norm": 1.732787424166437,
      "language_loss": 0.66242576,
      "learning_rate": 3.809592956714278e-07,
      "loss": 0.68440872,
      "num_input_tokens_seen": 144710255,
      "step": 6705,
      "time_per_iteration": 2.7855770587921143
    },
    {
      "auxiliary_loss_clip": 0.01175395,
      "auxiliary_loss_mlp": 0.01026809,
      "balance_loss_clip": 1.04841173,
      "balance_loss_mlp": 1.01936722,
      "epoch": 0.806348824625744,
      "flos": 22782591544320.0,
      "grad_norm": 2.2019929606122997,
      "language_loss": 0.74600989,
      "learning_rate": 3.805020906082057e-07,
      "loss": 0.7680319,
      "num_input_tokens_seen": 144728830,
      "step": 6706,
      "time_per_iteration": 2.690951347351074
    },
    {
      "auxiliary_loss_clip": 0.01176063,
      "auxiliary_loss_mlp": 0.01025143,
      "balance_loss_clip": 1.04494917,
      "balance_loss_mlp": 1.01710844,
      "epoch": 0.8064690675163831,
      "flos": 23404600385280.0,
      "grad_norm": 2.341591289231824,
      "language_loss": 0.8069014,
      "learning_rate": 3.8004513122152917e-07,
      "loss": 0.82891345,
      "num_input_tokens_seen": 144747140,
      "step": 6707,
      "time_per_iteration": 3.6679742336273193
    },
    {
      "auxiliary_loss_clip": 0.01162409,
      "auxiliary_loss_mlp": 0.01028749,
      "balance_loss_clip": 1.04577422,
      "balance_loss_mlp": 1.02147102,
      "epoch": 0.8065893104070222,
      "flos": 24060903736320.0,
      "grad_norm": 1.9165551982818292,
      "language_loss": 0.66959321,
      "learning_rate": 3.79588417580718e-07,
      "loss": 0.69150484,
      "num_input_tokens_seen": 144765250,
      "step": 6708,
      "time_per_iteration": 2.636263370513916
    },
    {
      "auxiliary_loss_clip": 0.01173951,
      "auxiliary_loss_mlp": 0.01024122,
      "balance_loss_clip": 1.04880667,
      "balance_loss_mlp": 1.01708889,
      "epoch": 0.8067095532976613,
      "flos": 22305410340480.0,
      "grad_norm": 1.9959328658483027,
      "language_loss": 0.76554245,
      "learning_rate": 3.791319497550558e-07,
      "loss": 0.78752315,
      "num_input_tokens_seen": 144783080,
      "step": 6709,
      "time_per_iteration": 2.593865394592285
    },
    {
      "auxiliary_loss_clip": 0.01181615,
      "auxiliary_loss_mlp": 0.01207298,
      "balance_loss_clip": 1.04525995,
      "balance_loss_mlp": 1.00062346,
      "epoch": 0.8068297961883004,
      "flos": 17129498296320.0,
      "grad_norm": 1.8803569008922474,
      "language_loss": 0.71473402,
      "learning_rate": 3.78675727813788e-07,
      "loss": 0.73862314,
      "num_input_tokens_seen": 144800645,
      "step": 6710,
      "time_per_iteration": 3.4768171310424805
    },
    {
      "auxiliary_loss_clip": 0.01174786,
      "auxiliary_loss_mlp": 0.01020883,
      "balance_loss_clip": 1.04640365,
      "balance_loss_mlp": 1.01346183,
      "epoch": 0.8069500390789395,
      "flos": 22018843635840.0,
      "grad_norm": 1.6572509848275032,
      "language_loss": 0.73659027,
      "learning_rate": 3.782197518261225e-07,
      "loss": 0.75854695,
      "num_input_tokens_seen": 144820085,
      "step": 6711,
      "time_per_iteration": 2.574448347091675
    },
    {
      "auxiliary_loss_clip": 0.01178794,
      "auxiliary_loss_mlp": 0.01027606,
      "balance_loss_clip": 1.04653645,
      "balance_loss_mlp": 1.01985788,
      "epoch": 0.8070702819695785,
      "flos": 19244241567360.0,
      "grad_norm": 2.266475427243482,
      "language_loss": 0.95728797,
      "learning_rate": 3.777640218612319e-07,
      "loss": 0.979352,
      "num_input_tokens_seen": 144838070,
      "step": 6712,
      "time_per_iteration": 2.5431222915649414
    },
    {
      "auxiliary_loss_clip": 0.01161672,
      "auxiliary_loss_mlp": 0.01023134,
      "balance_loss_clip": 1.04395437,
      "balance_loss_mlp": 1.01589179,
      "epoch": 0.8071905248602176,
      "flos": 21544320038400.0,
      "grad_norm": 2.5273801171092374,
      "language_loss": 0.72135222,
      "learning_rate": 3.773085379882488e-07,
      "loss": 0.7432003,
      "num_input_tokens_seen": 144857125,
      "step": 6713,
      "time_per_iteration": 2.480625867843628
    },
    {
      "auxiliary_loss_clip": 0.01170349,
      "auxiliary_loss_mlp": 0.01208092,
      "balance_loss_clip": 1.04438281,
      "balance_loss_mlp": 1.00062966,
      "epoch": 0.8073107677508568,
      "flos": 37268309105280.0,
      "grad_norm": 1.8556861548645984,
      "language_loss": 0.76132929,
      "learning_rate": 3.768533002762715e-07,
      "loss": 0.78511369,
      "num_input_tokens_seen": 144880660,
      "step": 6714,
      "time_per_iteration": 2.724076986312866
    },
    {
      "auxiliary_loss_clip": 0.01171582,
      "auxiliary_loss_mlp": 0.01021468,
      "balance_loss_clip": 1.04177904,
      "balance_loss_mlp": 1.01444983,
      "epoch": 0.8074310106414958,
      "flos": 28366269759360.0,
      "grad_norm": 2.222179540025052,
      "language_loss": 0.77214295,
      "learning_rate": 3.763983087943572e-07,
      "loss": 0.79407346,
      "num_input_tokens_seen": 144900050,
      "step": 6715,
      "time_per_iteration": 3.5574207305908203
    },
    {
      "auxiliary_loss_clip": 0.01160562,
      "auxiliary_loss_mlp": 0.01207947,
      "balance_loss_clip": 1.04320598,
      "balance_loss_mlp": 1.00063729,
      "epoch": 0.8075512535321349,
      "flos": 24281646768000.0,
      "grad_norm": 1.9061254507943381,
      "language_loss": 0.81028247,
      "learning_rate": 3.759435636115282e-07,
      "loss": 0.83396763,
      "num_input_tokens_seen": 144920835,
      "step": 6716,
      "time_per_iteration": 2.642876625061035
    },
    {
      "auxiliary_loss_clip": 0.011753,
      "auxiliary_loss_mlp": 0.01207053,
      "balance_loss_clip": 1.04228544,
      "balance_loss_mlp": 1.00065541,
      "epoch": 0.807671496422774,
      "flos": 26030855283840.0,
      "grad_norm": 1.737425879406121,
      "language_loss": 0.73102927,
      "learning_rate": 3.7548906479676967e-07,
      "loss": 0.75485277,
      "num_input_tokens_seen": 144940430,
      "step": 6717,
      "time_per_iteration": 2.7390642166137695
    },
    {
      "auxiliary_loss_clip": 0.01173731,
      "auxiliary_loss_mlp": 0.01021702,
      "balance_loss_clip": 1.0448668,
      "balance_loss_mlp": 1.01431131,
      "epoch": 0.8077917393134131,
      "flos": 23730740899200.0,
      "grad_norm": 1.8812804462053685,
      "language_loss": 0.71961868,
      "learning_rate": 3.7503481241902855e-07,
      "loss": 0.74157304,
      "num_input_tokens_seen": 144960405,
      "step": 6718,
      "time_per_iteration": 2.657344102859497
    },
    {
      "auxiliary_loss_clip": 0.01174363,
      "auxiliary_loss_mlp": 0.01207587,
      "balance_loss_clip": 1.04500544,
      "balance_loss_mlp": 1.00069094,
      "epoch": 0.8079119822040521,
      "flos": 18402028398720.0,
      "grad_norm": 2.762717313147742,
      "language_loss": 0.80347049,
      "learning_rate": 3.745808065472145e-07,
      "loss": 0.82728994,
      "num_input_tokens_seen": 144977700,
      "step": 6719,
      "time_per_iteration": 2.598454475402832
    },
    {
      "auxiliary_loss_clip": 0.01165925,
      "auxiliary_loss_mlp": 0.01030914,
      "balance_loss_clip": 1.04953885,
      "balance_loss_mlp": 1.0236249,
      "epoch": 0.8080322250946913,
      "flos": 23621787970560.0,
      "grad_norm": 1.571346453128701,
      "language_loss": 0.76374727,
      "learning_rate": 3.741270472501994e-07,
      "loss": 0.7857157,
      "num_input_tokens_seen": 144998340,
      "step": 6720,
      "time_per_iteration": 2.662609815597534
    },
    {
      "auxiliary_loss_clip": 0.01172411,
      "auxiliary_loss_mlp": 0.01022813,
      "balance_loss_clip": 1.04736137,
      "balance_loss_mlp": 1.01614666,
      "epoch": 0.8081524679853304,
      "flos": 22820692896000.0,
      "grad_norm": 1.683091817488238,
      "language_loss": 0.72878468,
      "learning_rate": 3.736735345968183e-07,
      "loss": 0.75073695,
      "num_input_tokens_seen": 145017950,
      "step": 6721,
      "time_per_iteration": 2.6471471786499023
    },
    {
      "auxiliary_loss_clip": 0.01170729,
      "auxiliary_loss_mlp": 0.01022665,
      "balance_loss_clip": 1.04692304,
      "balance_loss_mlp": 1.01575136,
      "epoch": 0.8082727108759694,
      "flos": 17640004343040.0,
      "grad_norm": 1.8274714800894407,
      "language_loss": 0.78710628,
      "learning_rate": 3.7322026865586986e-07,
      "loss": 0.80904019,
      "num_input_tokens_seen": 145036985,
      "step": 6722,
      "time_per_iteration": 2.570369005203247
    },
    {
      "auxiliary_loss_clip": 0.01177858,
      "auxiliary_loss_mlp": 0.01024055,
      "balance_loss_clip": 1.04979336,
      "balance_loss_mlp": 1.01656282,
      "epoch": 0.8083929537666086,
      "flos": 25958172113280.0,
      "grad_norm": 2.1550978371169838,
      "language_loss": 0.73617995,
      "learning_rate": 3.7276724949611206e-07,
      "loss": 0.7581991,
      "num_input_tokens_seen": 145057095,
      "step": 6723,
      "time_per_iteration": 2.7035880088806152
    },
    {
      "auxiliary_loss_clip": 0.01177506,
      "auxiliary_loss_mlp": 0.0102405,
      "balance_loss_clip": 1.04619527,
      "balance_loss_mlp": 1.01590168,
      "epoch": 0.8085131966572476,
      "flos": 27089178629760.0,
      "grad_norm": 3.1089262985012724,
      "language_loss": 0.74948126,
      "learning_rate": 3.723144771862694e-07,
      "loss": 0.77149677,
      "num_input_tokens_seen": 145077735,
      "step": 6724,
      "time_per_iteration": 2.65008807182312
    },
    {
      "auxiliary_loss_clip": 0.01178665,
      "auxiliary_loss_mlp": 0.01023799,
      "balance_loss_clip": 1.04264808,
      "balance_loss_mlp": 1.01628244,
      "epoch": 0.8086334395478867,
      "flos": 23988543788160.0,
      "grad_norm": 1.5961311808483967,
      "language_loss": 0.77266073,
      "learning_rate": 3.718619517950263e-07,
      "loss": 0.79468536,
      "num_input_tokens_seen": 145098330,
      "step": 6725,
      "time_per_iteration": 2.74000883102417
    },
    {
      "auxiliary_loss_clip": 0.0116706,
      "auxiliary_loss_mlp": 0.01029606,
      "balance_loss_clip": 1.04849291,
      "balance_loss_mlp": 1.02234876,
      "epoch": 0.8087536824385259,
      "flos": 20405879406720.0,
      "grad_norm": 22.975249555433773,
      "language_loss": 0.7699948,
      "learning_rate": 3.714096733910301e-07,
      "loss": 0.79196143,
      "num_input_tokens_seen": 145115855,
      "step": 6726,
      "time_per_iteration": 2.565951347351074
    },
    {
      "auxiliary_loss_clip": 0.01178317,
      "auxiliary_loss_mlp": 0.01022023,
      "balance_loss_clip": 1.04782009,
      "balance_loss_mlp": 1.01366055,
      "epoch": 0.8088739253291649,
      "flos": 25919639798400.0,
      "grad_norm": 3.190792869588463,
      "language_loss": 0.70603198,
      "learning_rate": 3.709576420428926e-07,
      "loss": 0.72803539,
      "num_input_tokens_seen": 145136655,
      "step": 6727,
      "time_per_iteration": 3.5737802982330322
    },
    {
      "auxiliary_loss_clip": 0.01173129,
      "auxiliary_loss_mlp": 0.01023312,
      "balance_loss_clip": 1.04267573,
      "balance_loss_mlp": 1.01605785,
      "epoch": 0.808994168219804,
      "flos": 28402072640640.0,
      "grad_norm": 2.620042875028036,
      "language_loss": 0.73236251,
      "learning_rate": 3.7050585781918463e-07,
      "loss": 0.75432682,
      "num_input_tokens_seen": 145156955,
      "step": 6728,
      "time_per_iteration": 2.696553945541382
    },
    {
      "auxiliary_loss_clip": 0.01175261,
      "auxiliary_loss_mlp": 0.01024542,
      "balance_loss_clip": 1.04699659,
      "balance_loss_mlp": 1.01648068,
      "epoch": 0.8091144111104431,
      "flos": 17421056991360.0,
      "grad_norm": 3.168706203572221,
      "language_loss": 0.69077826,
      "learning_rate": 3.700543207884428e-07,
      "loss": 0.7127763,
      "num_input_tokens_seen": 145173865,
      "step": 6729,
      "time_per_iteration": 2.627462387084961
    },
    {
      "auxiliary_loss_clip": 0.01169408,
      "auxiliary_loss_mlp": 0.01025506,
      "balance_loss_clip": 1.04672849,
      "balance_loss_mlp": 1.0183717,
      "epoch": 0.8092346540010822,
      "flos": 32153803361280.0,
      "grad_norm": 1.7167579684930125,
      "language_loss": 0.70934272,
      "learning_rate": 3.6960303101916466e-07,
      "loss": 0.73129183,
      "num_input_tokens_seen": 145193780,
      "step": 6730,
      "time_per_iteration": 2.715994119644165
    },
    {
      "auxiliary_loss_clip": 0.0106317,
      "auxiliary_loss_mlp": 0.01198625,
      "balance_loss_clip": 1.01036692,
      "balance_loss_mlp": 0.99982911,
      "epoch": 0.8093548968917212,
      "flos": 58035093390720.0,
      "grad_norm": 0.7436167923153255,
      "language_loss": 0.55502141,
      "learning_rate": 3.6915198857981047e-07,
      "loss": 0.57763934,
      "num_input_tokens_seen": 145258980,
      "step": 6731,
      "time_per_iteration": 3.184420347213745
    },
    {
      "auxiliary_loss_clip": 0.01171828,
      "auxiliary_loss_mlp": 0.01022284,
      "balance_loss_clip": 1.04325938,
      "balance_loss_mlp": 1.01403737,
      "epoch": 0.8094751397823604,
      "flos": 27381599251200.0,
      "grad_norm": 5.549622601750562,
      "language_loss": 0.68197417,
      "learning_rate": 3.687011935388027e-07,
      "loss": 0.70391524,
      "num_input_tokens_seen": 145281875,
      "step": 6732,
      "time_per_iteration": 2.712188243865967
    },
    {
      "auxiliary_loss_clip": 0.01169514,
      "auxiliary_loss_mlp": 0.01022689,
      "balance_loss_clip": 1.04640412,
      "balance_loss_mlp": 1.01555109,
      "epoch": 0.8095953826729995,
      "flos": 24061083304320.0,
      "grad_norm": 1.9675379432838138,
      "language_loss": 0.72661644,
      "learning_rate": 3.6825064596452646e-07,
      "loss": 0.74853849,
      "num_input_tokens_seen": 145302220,
      "step": 6733,
      "time_per_iteration": 2.6016952991485596
    },
    {
      "auxiliary_loss_clip": 0.01170159,
      "auxiliary_loss_mlp": 0.01023143,
      "balance_loss_clip": 1.04510832,
      "balance_loss_mlp": 1.01589489,
      "epoch": 0.8097156255636385,
      "flos": 23951412103680.0,
      "grad_norm": 2.431510823480135,
      "language_loss": 0.70843494,
      "learning_rate": 3.678003459253305e-07,
      "loss": 0.73036796,
      "num_input_tokens_seen": 145323070,
      "step": 6734,
      "time_per_iteration": 3.5998291969299316
    },
    {
      "auxiliary_loss_clip": 0.01173499,
      "auxiliary_loss_mlp": 0.01020778,
      "balance_loss_clip": 1.04181242,
      "balance_loss_mlp": 1.01316619,
      "epoch": 0.8098358684542777,
      "flos": 21799142098560.0,
      "grad_norm": 2.3059719028409758,
      "language_loss": 0.74345803,
      "learning_rate": 3.673502934895236e-07,
      "loss": 0.76540077,
      "num_input_tokens_seen": 145342575,
      "step": 6735,
      "time_per_iteration": 2.6994762420654297
    },
    {
      "auxiliary_loss_clip": 0.01062972,
      "auxiliary_loss_mlp": 0.01000064,
      "balance_loss_clip": 1.01020956,
      "balance_loss_mlp": 0.99930143,
      "epoch": 0.8099561113449167,
      "flos": 68809515966720.0,
      "grad_norm": 0.6907670056921709,
      "language_loss": 0.57821065,
      "learning_rate": 3.669004887253802e-07,
      "loss": 0.59884101,
      "num_input_tokens_seen": 145408865,
      "step": 6736,
      "time_per_iteration": 4.1583781242370605
    },
    {
      "auxiliary_loss_clip": 0.01176436,
      "auxiliary_loss_mlp": 0.0102456,
      "balance_loss_clip": 1.04673231,
      "balance_loss_mlp": 1.01756847,
      "epoch": 0.8100763542355558,
      "flos": 23586056916480.0,
      "grad_norm": 1.8521890609783238,
      "language_loss": 0.79140937,
      "learning_rate": 3.664509317011335e-07,
      "loss": 0.81341934,
      "num_input_tokens_seen": 145429200,
      "step": 6737,
      "time_per_iteration": 2.6714541912078857
    },
    {
      "auxiliary_loss_clip": 0.01170397,
      "auxiliary_loss_mlp": 0.01027125,
      "balance_loss_clip": 1.04892397,
      "balance_loss_mlp": 1.01909614,
      "epoch": 0.810196597126195,
      "flos": 31650408207360.0,
      "grad_norm": 2.555546006970509,
      "language_loss": 0.73843831,
      "learning_rate": 3.6600162248498134e-07,
      "loss": 0.76041359,
      "num_input_tokens_seen": 145452830,
      "step": 6738,
      "time_per_iteration": 2.759580373764038
    },
    {
      "auxiliary_loss_clip": 0.01162695,
      "auxiliary_loss_mlp": 0.01024088,
      "balance_loss_clip": 1.03618085,
      "balance_loss_mlp": 1.01733518,
      "epoch": 0.810316840016834,
      "flos": 24900459298560.0,
      "grad_norm": 2.080319091998337,
      "language_loss": 0.75999302,
      "learning_rate": 3.6555256114508426e-07,
      "loss": 0.78186083,
      "num_input_tokens_seen": 145472625,
      "step": 6739,
      "time_per_iteration": 2.7554967403411865
    },
    {
      "auxiliary_loss_clip": 0.01172493,
      "auxiliary_loss_mlp": 0.01024132,
      "balance_loss_clip": 1.04196382,
      "balance_loss_mlp": 1.01649094,
      "epoch": 0.8104370829074731,
      "flos": 27965003950080.0,
      "grad_norm": 2.1427259930281233,
      "language_loss": 0.73437405,
      "learning_rate": 3.651037477495642e-07,
      "loss": 0.75634027,
      "num_input_tokens_seen": 145494075,
      "step": 6740,
      "time_per_iteration": 2.7355475425720215
    },
    {
      "auxiliary_loss_clip": 0.0116776,
      "auxiliary_loss_mlp": 0.01025109,
      "balance_loss_clip": 1.04631162,
      "balance_loss_mlp": 1.01781392,
      "epoch": 0.8105573257981122,
      "flos": 24640752988800.0,
      "grad_norm": 2.567190706293706,
      "language_loss": 0.68131119,
      "learning_rate": 3.6465518236650584e-07,
      "loss": 0.70323992,
      "num_input_tokens_seen": 145514220,
      "step": 6741,
      "time_per_iteration": 3.6432785987854004
    },
    {
      "auxiliary_loss_clip": 0.01175072,
      "auxiliary_loss_mlp": 0.01026957,
      "balance_loss_clip": 1.0419662,
      "balance_loss_mlp": 1.01994801,
      "epoch": 0.8106775686887513,
      "flos": 26358935132160.0,
      "grad_norm": 1.805004641125907,
      "language_loss": 0.78629851,
      "learning_rate": 3.642068650639558e-07,
      "loss": 0.80831885,
      "num_input_tokens_seen": 145533965,
      "step": 6742,
      "time_per_iteration": 2.7875707149505615
    },
    {
      "auxiliary_loss_clip": 0.01165369,
      "auxiliary_loss_mlp": 0.01024986,
      "balance_loss_clip": 1.03954387,
      "balance_loss_mlp": 1.01772881,
      "epoch": 0.8107978115793903,
      "flos": 27271892136960.0,
      "grad_norm": 2.0662491295107417,
      "language_loss": 0.64431083,
      "learning_rate": 3.6375879590992334e-07,
      "loss": 0.66621441,
      "num_input_tokens_seen": 145554310,
      "step": 6743,
      "time_per_iteration": 2.662712812423706
    },
    {
      "auxiliary_loss_clip": 0.01169006,
      "auxiliary_loss_mlp": 0.01026632,
      "balance_loss_clip": 1.04356599,
      "balance_loss_mlp": 1.018538,
      "epoch": 0.8109180544700295,
      "flos": 24934322845440.0,
      "grad_norm": 1.9860104883402805,
      "language_loss": 0.81136578,
      "learning_rate": 3.6331097497238173e-07,
      "loss": 0.83332217,
      "num_input_tokens_seen": 145573755,
      "step": 6744,
      "time_per_iteration": 2.7122156620025635
    },
    {
      "auxiliary_loss_clip": 0.01173044,
      "auxiliary_loss_mlp": 0.01020901,
      "balance_loss_clip": 1.04311645,
      "balance_loss_mlp": 1.01379025,
      "epoch": 0.8110382973606686,
      "flos": 21105383840640.0,
      "grad_norm": 4.2112267069456895,
      "language_loss": 0.80343163,
      "learning_rate": 3.628634023192627e-07,
      "loss": 0.82537103,
      "num_input_tokens_seen": 145594000,
      "step": 6745,
      "time_per_iteration": 2.7191336154937744
    },
    {
      "auxiliary_loss_clip": 0.0117278,
      "auxiliary_loss_mlp": 0.01027972,
      "balance_loss_clip": 1.04666877,
      "balance_loss_mlp": 1.01956224,
      "epoch": 0.8111585402513076,
      "flos": 15414081500160.0,
      "grad_norm": 2.0163405250400186,
      "language_loss": 0.7510134,
      "learning_rate": 3.624160780184644e-07,
      "loss": 0.77302086,
      "num_input_tokens_seen": 145611215,
      "step": 6746,
      "time_per_iteration": 2.774146795272827
    },
    {
      "auxiliary_loss_clip": 0.01166434,
      "auxiliary_loss_mlp": 0.01023789,
      "balance_loss_clip": 1.04319,
      "balance_loss_mlp": 1.01611471,
      "epoch": 0.8112787831419467,
      "flos": 24095736950400.0,
      "grad_norm": 1.9004202962494106,
      "language_loss": 0.74538058,
      "learning_rate": 3.6196900213784496e-07,
      "loss": 0.76728284,
      "num_input_tokens_seen": 145630530,
      "step": 6747,
      "time_per_iteration": 2.8217403888702393
    },
    {
      "auxiliary_loss_clip": 0.01169231,
      "auxiliary_loss_mlp": 0.0102652,
      "balance_loss_clip": 1.04547262,
      "balance_loss_mlp": 1.01944518,
      "epoch": 0.8113990260325858,
      "flos": 20483374999680.0,
      "grad_norm": 2.0348985083899316,
      "language_loss": 0.86771864,
      "learning_rate": 3.6152217474522527e-07,
      "loss": 0.88967615,
      "num_input_tokens_seen": 145647345,
      "step": 6748,
      "time_per_iteration": 2.6151230335235596
    },
    {
      "auxiliary_loss_clip": 0.01169021,
      "auxiliary_loss_mlp": 0.01026567,
      "balance_loss_clip": 1.0477072,
      "balance_loss_mlp": 1.01959085,
      "epoch": 0.8115192689232249,
      "flos": 24901141656960.0,
      "grad_norm": 2.138308229876319,
      "language_loss": 0.72612512,
      "learning_rate": 3.6107559590838975e-07,
      "loss": 0.74808097,
      "num_input_tokens_seen": 145666330,
      "step": 6749,
      "time_per_iteration": 2.6955132484436035
    },
    {
      "auxiliary_loss_clip": 0.01172554,
      "auxiliary_loss_mlp": 0.01021722,
      "balance_loss_clip": 1.03876686,
      "balance_loss_mlp": 1.01422977,
      "epoch": 0.811639511813864,
      "flos": 24057204635520.0,
      "grad_norm": 2.2397497964004605,
      "language_loss": 0.66552365,
      "learning_rate": 3.606292656950822e-07,
      "loss": 0.68746638,
      "num_input_tokens_seen": 145684740,
      "step": 6750,
      "time_per_iteration": 2.7353482246398926
    },
    {
      "auxiliary_loss_clip": 0.01168905,
      "auxiliary_loss_mlp": 0.0102178,
      "balance_loss_clip": 1.0427922,
      "balance_loss_mlp": 1.01387966,
      "epoch": 0.8117597547045031,
      "flos": 23185150243200.0,
      "grad_norm": 2.5728856550204333,
      "language_loss": 0.86756587,
      "learning_rate": 3.601831841730121e-07,
      "loss": 0.88947266,
      "num_input_tokens_seen": 145702660,
      "step": 6751,
      "time_per_iteration": 2.6404409408569336
    },
    {
      "auxiliary_loss_clip": 0.01170603,
      "auxiliary_loss_mlp": 0.01022692,
      "balance_loss_clip": 1.04781651,
      "balance_loss_mlp": 1.01499104,
      "epoch": 0.8118799975951422,
      "flos": 23040250778880.0,
      "grad_norm": 1.6612779344949815,
      "language_loss": 0.72813916,
      "learning_rate": 3.5973735140984916e-07,
      "loss": 0.750072,
      "num_input_tokens_seen": 145722830,
      "step": 6752,
      "time_per_iteration": 2.612809181213379
    },
    {
      "auxiliary_loss_clip": 0.01173124,
      "auxiliary_loss_mlp": 0.01206807,
      "balance_loss_clip": 1.03940988,
      "balance_loss_mlp": 1.00058305,
      "epoch": 0.8120002404857812,
      "flos": 24639962889600.0,
      "grad_norm": 2.750725473038223,
      "language_loss": 0.793437,
      "learning_rate": 3.5929176747322607e-07,
      "loss": 0.8172363,
      "num_input_tokens_seen": 145741935,
      "step": 6753,
      "time_per_iteration": 2.7368757724761963
    },
    {
      "auxiliary_loss_clip": 0.01079789,
      "auxiliary_loss_mlp": 0.01000867,
      "balance_loss_clip": 1.01122999,
      "balance_loss_mlp": 1.00005674,
      "epoch": 0.8121204833764204,
      "flos": 57415742156160.0,
      "grad_norm": 0.8050496337509707,
      "language_loss": 0.56174135,
      "learning_rate": 3.588464324307372e-07,
      "loss": 0.58254796,
      "num_input_tokens_seen": 145805560,
      "step": 6754,
      "time_per_iteration": 4.1639604568481445
    },
    {
      "auxiliary_loss_clip": 0.01169736,
      "auxiliary_loss_mlp": 0.01023603,
      "balance_loss_clip": 1.04331768,
      "balance_loss_mlp": 1.01637268,
      "epoch": 0.8122407262670595,
      "flos": 19464589549440.0,
      "grad_norm": 1.8530916423315569,
      "language_loss": 0.75534296,
      "learning_rate": 3.584013463499391e-07,
      "loss": 0.77727634,
      "num_input_tokens_seen": 145824180,
      "step": 6755,
      "time_per_iteration": 2.6190385818481445
    },
    {
      "auxiliary_loss_clip": 0.0107701,
      "auxiliary_loss_mlp": 0.01000311,
      "balance_loss_clip": 1.01119852,
      "balance_loss_mlp": 0.9995122,
      "epoch": 0.8123609691576985,
      "flos": 56425325472000.0,
      "grad_norm": 0.7349202430738949,
      "language_loss": 0.64468288,
      "learning_rate": 3.579565092983521e-07,
      "loss": 0.66545606,
      "num_input_tokens_seen": 145885300,
      "step": 6756,
      "time_per_iteration": 3.0588839054107666
    },
    {
      "auxiliary_loss_clip": 0.01169255,
      "auxiliary_loss_mlp": 0.01031185,
      "balance_loss_clip": 1.04914808,
      "balance_loss_mlp": 1.02369297,
      "epoch": 0.8124812120483377,
      "flos": 20631973564800.0,
      "grad_norm": 2.056355292471918,
      "language_loss": 0.84048593,
      "learning_rate": 3.575119213434565e-07,
      "loss": 0.86249042,
      "num_input_tokens_seen": 145903815,
      "step": 6757,
      "time_per_iteration": 2.659235715866089
    },
    {
      "auxiliary_loss_clip": 0.01167009,
      "auxiliary_loss_mlp": 0.01021317,
      "balance_loss_clip": 1.04585409,
      "balance_loss_mlp": 1.01432538,
      "epoch": 0.8126014549389767,
      "flos": 22492397566080.0,
      "grad_norm": 3.59904829896989,
      "language_loss": 0.81870258,
      "learning_rate": 3.5706758255269765e-07,
      "loss": 0.84058589,
      "num_input_tokens_seen": 145922270,
      "step": 6758,
      "time_per_iteration": 2.635814905166626
    },
    {
      "auxiliary_loss_clip": 0.01174295,
      "auxiliary_loss_mlp": 0.01024536,
      "balance_loss_clip": 1.04435802,
      "balance_loss_mlp": 1.0169127,
      "epoch": 0.8127216978296158,
      "flos": 23287961946240.0,
      "grad_norm": 1.6996633344776622,
      "language_loss": 0.69986826,
      "learning_rate": 3.566234929934795e-07,
      "loss": 0.72185659,
      "num_input_tokens_seen": 145941470,
      "step": 6759,
      "time_per_iteration": 2.656095266342163
    },
    {
      "auxiliary_loss_clip": 0.01166208,
      "auxiliary_loss_mlp": 0.01026075,
      "balance_loss_clip": 1.04743195,
      "balance_loss_mlp": 1.01857996,
      "epoch": 0.812841940720255,
      "flos": 25154994049920.0,
      "grad_norm": 1.5627017217269625,
      "language_loss": 0.71761173,
      "learning_rate": 3.561796527331706e-07,
      "loss": 0.73953456,
      "num_input_tokens_seen": 145963145,
      "step": 6760,
      "time_per_iteration": 3.6412627696990967
    },
    {
      "auxiliary_loss_clip": 0.01179606,
      "auxiliary_loss_mlp": 0.01022938,
      "balance_loss_clip": 1.04417598,
      "balance_loss_mlp": 1.01532996,
      "epoch": 0.812962183610894,
      "flos": 26648446752000.0,
      "grad_norm": 2.0146970778762388,
      "language_loss": 0.77514875,
      "learning_rate": 3.5573606183910163e-07,
      "loss": 0.79717422,
      "num_input_tokens_seen": 145983150,
      "step": 6761,
      "time_per_iteration": 2.771045446395874
    },
    {
      "auxiliary_loss_clip": 0.01176078,
      "auxiliary_loss_mlp": 0.01024246,
      "balance_loss_clip": 1.04550469,
      "balance_loss_mlp": 1.0168221,
      "epoch": 0.8130824265015331,
      "flos": 24966965329920.0,
      "grad_norm": 1.7309293274194795,
      "language_loss": 0.78751713,
      "learning_rate": 3.5529272037856493e-07,
      "loss": 0.80952036,
      "num_input_tokens_seen": 146001365,
      "step": 6762,
      "time_per_iteration": 2.6563010215759277
    },
    {
      "auxiliary_loss_clip": 0.01100333,
      "auxiliary_loss_mlp": 0.01001417,
      "balance_loss_clip": 1.01056862,
      "balance_loss_mlp": 1.00064802,
      "epoch": 0.8132026693921722,
      "flos": 67622918175360.0,
      "grad_norm": 0.7051551454094197,
      "language_loss": 0.53769481,
      "learning_rate": 3.548496284188149e-07,
      "loss": 0.55871236,
      "num_input_tokens_seen": 146061570,
      "step": 6763,
      "time_per_iteration": 4.198409080505371
    },
    {
      "auxiliary_loss_clip": 0.01173003,
      "auxiliary_loss_mlp": 0.010221,
      "balance_loss_clip": 1.04440546,
      "balance_loss_mlp": 1.01462603,
      "epoch": 0.8133229122828113,
      "flos": 19495149045120.0,
      "grad_norm": 1.9494862875920707,
      "language_loss": 0.79420418,
      "learning_rate": 3.544067860270681e-07,
      "loss": 0.8161552,
      "num_input_tokens_seen": 146079145,
      "step": 6764,
      "time_per_iteration": 2.7812106609344482
    },
    {
      "auxiliary_loss_clip": 0.01175182,
      "auxiliary_loss_mlp": 0.01021748,
      "balance_loss_clip": 1.04211044,
      "balance_loss_mlp": 1.01406527,
      "epoch": 0.8134431551734503,
      "flos": 20668135582080.0,
      "grad_norm": 2.3638947154429824,
      "language_loss": 0.70902014,
      "learning_rate": 3.539641932705029e-07,
      "loss": 0.7309894,
      "num_input_tokens_seen": 146097625,
      "step": 6765,
      "time_per_iteration": 2.6676411628723145
    },
    {
      "auxiliary_loss_clip": 0.01170827,
      "auxiliary_loss_mlp": 0.01026011,
      "balance_loss_clip": 1.04761243,
      "balance_loss_mlp": 1.01782107,
      "epoch": 0.8135633980640895,
      "flos": 21507332008320.0,
      "grad_norm": 2.7145493249513026,
      "language_loss": 0.77273607,
      "learning_rate": 3.53521850216262e-07,
      "loss": 0.79470444,
      "num_input_tokens_seen": 146117195,
      "step": 6766,
      "time_per_iteration": 2.667816638946533
    },
    {
      "auxiliary_loss_clip": 0.01169903,
      "auxiliary_loss_mlp": 0.01025257,
      "balance_loss_clip": 1.04890597,
      "balance_loss_mlp": 1.01735342,
      "epoch": 0.8136836409547286,
      "flos": 20554442058240.0,
      "grad_norm": 1.7373379473760624,
      "language_loss": 0.7672621,
      "learning_rate": 3.530797569314461e-07,
      "loss": 0.78921366,
      "num_input_tokens_seen": 146136220,
      "step": 6767,
      "time_per_iteration": 3.5725317001342773
    },
    {
      "auxiliary_loss_clip": 0.01168173,
      "auxiliary_loss_mlp": 0.01020275,
      "balance_loss_clip": 1.04848385,
      "balance_loss_mlp": 1.01286006,
      "epoch": 0.8138038838453676,
      "flos": 20299045380480.0,
      "grad_norm": 2.0706540172250882,
      "language_loss": 0.77941948,
      "learning_rate": 3.5263791348312235e-07,
      "loss": 0.80130398,
      "num_input_tokens_seen": 146155415,
      "step": 6768,
      "time_per_iteration": 2.758465528488159
    },
    {
      "auxiliary_loss_clip": 0.01168948,
      "auxiliary_loss_mlp": 0.01023004,
      "balance_loss_clip": 1.04212511,
      "balance_loss_mlp": 1.015306,
      "epoch": 0.8139241267360068,
      "flos": 29789840551680.0,
      "grad_norm": 1.9039989153621661,
      "language_loss": 0.70992947,
      "learning_rate": 3.521963199383171e-07,
      "loss": 0.73184896,
      "num_input_tokens_seen": 146178370,
      "step": 6769,
      "time_per_iteration": 2.660184383392334
    },
    {
      "auxiliary_loss_clip": 0.01178345,
      "auxiliary_loss_mlp": 0.01021841,
      "balance_loss_clip": 1.0421865,
      "balance_loss_mlp": 1.01371408,
      "epoch": 0.8140443696266458,
      "flos": 19713270384000.0,
      "grad_norm": 2.1908230987409594,
      "language_loss": 0.76828337,
      "learning_rate": 3.517549763640197e-07,
      "loss": 0.79028523,
      "num_input_tokens_seen": 146196010,
      "step": 6770,
      "time_per_iteration": 2.7945919036865234
    },
    {
      "auxiliary_loss_clip": 0.01168448,
      "auxiliary_loss_mlp": 0.01207639,
      "balance_loss_clip": 1.04863584,
      "balance_loss_mlp": 1.00072908,
      "epoch": 0.8141646125172849,
      "flos": 27160568910720.0,
      "grad_norm": 2.018643201432914,
      "language_loss": 0.7150048,
      "learning_rate": 3.513138828271829e-07,
      "loss": 0.73876566,
      "num_input_tokens_seen": 146215880,
      "step": 6771,
      "time_per_iteration": 2.6513304710388184
    },
    {
      "auxiliary_loss_clip": 0.01169821,
      "auxiliary_loss_mlp": 0.01024909,
      "balance_loss_clip": 1.04266286,
      "balance_loss_mlp": 1.01732087,
      "epoch": 0.8142848554079241,
      "flos": 39673102700160.0,
      "grad_norm": 4.539478032241119,
      "language_loss": 0.70205295,
      "learning_rate": 3.508730393947179e-07,
      "loss": 0.72400028,
      "num_input_tokens_seen": 146239135,
      "step": 6772,
      "time_per_iteration": 2.8187296390533447
    },
    {
      "auxiliary_loss_clip": 0.0117517,
      "auxiliary_loss_mlp": 0.01022747,
      "balance_loss_clip": 1.04355824,
      "balance_loss_mlp": 1.01521885,
      "epoch": 0.8144050982985631,
      "flos": 22237288197120.0,
      "grad_norm": 1.75847486183599,
      "language_loss": 0.7201851,
      "learning_rate": 3.504324461335024e-07,
      "loss": 0.74216425,
      "num_input_tokens_seen": 146259245,
      "step": 6773,
      "time_per_iteration": 2.708259105682373
    },
    {
      "auxiliary_loss_clip": 0.01166422,
      "auxiliary_loss_mlp": 0.01025092,
      "balance_loss_clip": 1.03842723,
      "balance_loss_mlp": 1.01644933,
      "epoch": 0.8145253411892022,
      "flos": 23038239617280.0,
      "grad_norm": 1.8623537784661897,
      "language_loss": 0.8857702,
      "learning_rate": 3.499921031103732e-07,
      "loss": 0.90768534,
      "num_input_tokens_seen": 146280015,
      "step": 6774,
      "time_per_iteration": 2.687459945678711
    },
    {
      "auxiliary_loss_clip": 0.01184469,
      "auxiliary_loss_mlp": 0.01024104,
      "balance_loss_clip": 1.04251599,
      "balance_loss_mlp": 1.01654601,
      "epoch": 0.8146455840798413,
      "flos": 24827668387200.0,
      "grad_norm": 2.3237794265308196,
      "language_loss": 0.78538668,
      "learning_rate": 3.4955201039212987e-07,
      "loss": 0.80747241,
      "num_input_tokens_seen": 146300935,
      "step": 6775,
      "time_per_iteration": 2.715653896331787
    },
    {
      "auxiliary_loss_clip": 0.01176931,
      "auxiliary_loss_mlp": 0.0102371,
      "balance_loss_clip": 1.04806447,
      "balance_loss_mlp": 1.01624751,
      "epoch": 0.8147658269704804,
      "flos": 19974520978560.0,
      "grad_norm": 2.1440668315402394,
      "language_loss": 0.65847659,
      "learning_rate": 3.4911216804553465e-07,
      "loss": 0.68048298,
      "num_input_tokens_seen": 146319835,
      "step": 6776,
      "time_per_iteration": 2.633619546890259
    },
    {
      "auxiliary_loss_clip": 0.0117399,
      "auxiliary_loss_mlp": 0.01027246,
      "balance_loss_clip": 1.04562521,
      "balance_loss_mlp": 1.01886559,
      "epoch": 0.8148860698611194,
      "flos": 21178031097600.0,
      "grad_norm": 2.252089396289886,
      "language_loss": 0.70790708,
      "learning_rate": 3.4867257613731017e-07,
      "loss": 0.72991949,
      "num_input_tokens_seen": 146339030,
      "step": 6777,
      "time_per_iteration": 2.6441779136657715
    },
    {
      "auxiliary_loss_clip": 0.01173266,
      "auxiliary_loss_mlp": 0.01027637,
      "balance_loss_clip": 1.04437959,
      "balance_loss_mlp": 1.02042449,
      "epoch": 0.8150063127517585,
      "flos": 19606903234560.0,
      "grad_norm": 2.6322296436999766,
      "language_loss": 0.85580426,
      "learning_rate": 3.4823323473414343e-07,
      "loss": 0.87781328,
      "num_input_tokens_seen": 146358550,
      "step": 6778,
      "time_per_iteration": 2.5782196521759033
    },
    {
      "auxiliary_loss_clip": 0.01180877,
      "auxiliary_loss_mlp": 0.01025853,
      "balance_loss_clip": 1.04271698,
      "balance_loss_mlp": 1.0174545,
      "epoch": 0.8151265556423977,
      "flos": 22638374438400.0,
      "grad_norm": 2.780009631775973,
      "language_loss": 0.76446772,
      "learning_rate": 3.477941439026812e-07,
      "loss": 0.78653502,
      "num_input_tokens_seen": 146376770,
      "step": 6779,
      "time_per_iteration": 2.7064096927642822
    },
    {
      "auxiliary_loss_clip": 0.01172745,
      "auxiliary_loss_mlp": 0.01022484,
      "balance_loss_clip": 1.04522741,
      "balance_loss_mlp": 1.01553977,
      "epoch": 0.8152467985330367,
      "flos": 17968048277760.0,
      "grad_norm": 3.4467892556444593,
      "language_loss": 0.73079282,
      "learning_rate": 3.473553037095349e-07,
      "loss": 0.75274515,
      "num_input_tokens_seen": 146395795,
      "step": 6780,
      "time_per_iteration": 2.6021511554718018
    },
    {
      "auxiliary_loss_clip": 0.0116574,
      "auxiliary_loss_mlp": 0.01020299,
      "balance_loss_clip": 1.0436089,
      "balance_loss_mlp": 1.01332545,
      "epoch": 0.8153670414236758,
      "flos": 24969012405120.0,
      "grad_norm": 1.7021671343611755,
      "language_loss": 0.83203238,
      "learning_rate": 3.469167142212743e-07,
      "loss": 0.8538928,
      "num_input_tokens_seen": 146417640,
      "step": 6781,
      "time_per_iteration": 3.6661899089813232
    },
    {
      "auxiliary_loss_clip": 0.01171687,
      "auxiliary_loss_mlp": 0.01021872,
      "balance_loss_clip": 1.04728687,
      "balance_loss_mlp": 1.01384878,
      "epoch": 0.8154872843143149,
      "flos": 31066069754880.0,
      "grad_norm": 2.4001185311994613,
      "language_loss": 0.63176823,
      "learning_rate": 3.4647837550443337e-07,
      "loss": 0.65370381,
      "num_input_tokens_seen": 146436205,
      "step": 6782,
      "time_per_iteration": 2.6889591217041016
    },
    {
      "auxiliary_loss_clip": 0.01177399,
      "auxiliary_loss_mlp": 0.01027373,
      "balance_loss_clip": 1.04438615,
      "balance_loss_mlp": 1.02023792,
      "epoch": 0.815607527204954,
      "flos": 19391654983680.0,
      "grad_norm": 2.115407174302952,
      "language_loss": 0.74407017,
      "learning_rate": 3.460402876255086e-07,
      "loss": 0.76611781,
      "num_input_tokens_seen": 146453595,
      "step": 6783,
      "time_per_iteration": 2.590153217315674
    },
    {
      "auxiliary_loss_clip": 0.01170655,
      "auxiliary_loss_mlp": 0.01025166,
      "balance_loss_clip": 1.04502416,
      "balance_loss_mlp": 1.01759672,
      "epoch": 0.815727770095593,
      "flos": 26140418743680.0,
      "grad_norm": 2.2100119250796575,
      "language_loss": 0.71876585,
      "learning_rate": 3.456024506509574e-07,
      "loss": 0.74072409,
      "num_input_tokens_seen": 146474515,
      "step": 6784,
      "time_per_iteration": 2.657773494720459
    },
    {
      "auxiliary_loss_clip": 0.01169846,
      "auxiliary_loss_mlp": 0.01208017,
      "balance_loss_clip": 1.04804456,
      "balance_loss_mlp": 1.00058198,
      "epoch": 0.8158480129862322,
      "flos": 25337527989120.0,
      "grad_norm": 1.7211276539895566,
      "language_loss": 0.73829466,
      "learning_rate": 3.4516486464719873e-07,
      "loss": 0.76207334,
      "num_input_tokens_seen": 146493905,
      "step": 6785,
      "time_per_iteration": 2.6507678031921387
    },
    {
      "auxiliary_loss_clip": 0.01171728,
      "auxiliary_loss_mlp": 0.01026189,
      "balance_loss_clip": 1.0405637,
      "balance_loss_mlp": 1.0186069,
      "epoch": 0.8159682558768713,
      "flos": 34423645559040.0,
      "grad_norm": 1.713354695394073,
      "language_loss": 0.62209815,
      "learning_rate": 3.4472752968061445e-07,
      "loss": 0.6440773,
      "num_input_tokens_seen": 146518335,
      "step": 6786,
      "time_per_iteration": 2.7895894050598145
    },
    {
      "auxiliary_loss_clip": 0.01169835,
      "auxiliary_loss_mlp": 0.01024257,
      "balance_loss_clip": 1.04550302,
      "balance_loss_mlp": 1.01708937,
      "epoch": 0.8160884987675103,
      "flos": 18653223185280.0,
      "grad_norm": 1.7929011139722038,
      "language_loss": 0.73999947,
      "learning_rate": 3.442904458175475e-07,
      "loss": 0.76194036,
      "num_input_tokens_seen": 146535655,
      "step": 6787,
      "time_per_iteration": 3.4828827381134033
    },
    {
      "auxiliary_loss_clip": 0.01168241,
      "auxiliary_loss_mlp": 0.01022358,
      "balance_loss_clip": 1.04494023,
      "balance_loss_mlp": 1.01484489,
      "epoch": 0.8162087416581495,
      "flos": 31430527102080.0,
      "grad_norm": 2.1827482877313895,
      "language_loss": 0.7608611,
      "learning_rate": 3.438536131243044e-07,
      "loss": 0.78276706,
      "num_input_tokens_seen": 146556815,
      "step": 6788,
      "time_per_iteration": 2.7389116287231445
    },
    {
      "auxiliary_loss_clip": 0.01175612,
      "auxiliary_loss_mlp": 0.01025628,
      "balance_loss_clip": 1.04428411,
      "balance_loss_mlp": 1.01771879,
      "epoch": 0.8163289845487885,
      "flos": 37593910915200.0,
      "grad_norm": 2.2333438459705697,
      "language_loss": 0.6195541,
      "learning_rate": 3.434170316671503e-07,
      "loss": 0.64156651,
      "num_input_tokens_seen": 146581845,
      "step": 6789,
      "time_per_iteration": 2.856266736984253
    },
    {
      "auxiliary_loss_clip": 0.0117052,
      "auxiliary_loss_mlp": 0.01022632,
      "balance_loss_clip": 1.04668117,
      "balance_loss_mlp": 1.01532102,
      "epoch": 0.8164492274394276,
      "flos": 13953989554560.0,
      "grad_norm": 3.293915847034053,
      "language_loss": 0.89929211,
      "learning_rate": 3.4298070151231583e-07,
      "loss": 0.92122364,
      "num_input_tokens_seen": 146597245,
      "step": 6790,
      "time_per_iteration": 3.5718629360198975
    },
    {
      "auxiliary_loss_clip": 0.01175807,
      "auxiliary_loss_mlp": 0.01022953,
      "balance_loss_clip": 1.04439485,
      "balance_loss_mlp": 1.01566935,
      "epoch": 0.8165694703300668,
      "flos": 28986554747520.0,
      "grad_norm": 2.6083566259450537,
      "language_loss": 0.59843993,
      "learning_rate": 3.425446227259916e-07,
      "loss": 0.62042761,
      "num_input_tokens_seen": 146618210,
      "step": 6791,
      "time_per_iteration": 2.7649545669555664
    },
    {
      "auxiliary_loss_clip": 0.01172583,
      "auxiliary_loss_mlp": 0.01019757,
      "balance_loss_clip": 1.04403734,
      "balance_loss_mlp": 1.01333165,
      "epoch": 0.8166897132207058,
      "flos": 25118365155840.0,
      "grad_norm": 2.03964392661833,
      "language_loss": 0.82532203,
      "learning_rate": 3.421087953743296e-07,
      "loss": 0.84724545,
      "num_input_tokens_seen": 146637975,
      "step": 6792,
      "time_per_iteration": 2.785600423812866
    },
    {
      "auxiliary_loss_clip": 0.01170288,
      "auxiliary_loss_mlp": 0.01023425,
      "balance_loss_clip": 1.04454553,
      "balance_loss_mlp": 1.01536012,
      "epoch": 0.8168099561113449,
      "flos": 23148593176320.0,
      "grad_norm": 2.748741075178097,
      "language_loss": 0.80538779,
      "learning_rate": 3.416732195234464e-07,
      "loss": 0.82732487,
      "num_input_tokens_seen": 146658030,
      "step": 6793,
      "time_per_iteration": 3.6192610263824463
    },
    {
      "auxiliary_loss_clip": 0.01173387,
      "auxiliary_loss_mlp": 0.01019618,
      "balance_loss_clip": 1.04646254,
      "balance_loss_mlp": 1.01277804,
      "epoch": 0.816930199001984,
      "flos": 18407666833920.0,
      "grad_norm": 1.6025364144736172,
      "language_loss": 0.79801947,
      "learning_rate": 3.4123789523941613e-07,
      "loss": 0.81994957,
      "num_input_tokens_seen": 146677855,
      "step": 6794,
      "time_per_iteration": 2.6894447803497314
    },
    {
      "auxiliary_loss_clip": 0.01165071,
      "auxiliary_loss_mlp": 0.01023999,
      "balance_loss_clip": 1.04467714,
      "balance_loss_mlp": 1.01599431,
      "epoch": 0.8170504418926231,
      "flos": 21251324799360.0,
      "grad_norm": 1.7642923759546278,
      "language_loss": 0.63790679,
      "learning_rate": 3.4080282258827884e-07,
      "loss": 0.65979743,
      "num_input_tokens_seen": 146696230,
      "step": 6795,
      "time_per_iteration": 2.587517023086548
    },
    {
      "auxiliary_loss_clip": 0.01170469,
      "auxiliary_loss_mlp": 0.01023574,
      "balance_loss_clip": 1.04489172,
      "balance_loss_mlp": 1.01607549,
      "epoch": 0.8171706847832622,
      "flos": 19099234362240.0,
      "grad_norm": 2.289623430038986,
      "language_loss": 0.72507191,
      "learning_rate": 3.403680016360342e-07,
      "loss": 0.74701226,
      "num_input_tokens_seen": 146714835,
      "step": 6796,
      "time_per_iteration": 2.690547466278076
    },
    {
      "auxiliary_loss_clip": 0.0116446,
      "auxiliary_loss_mlp": 0.01031755,
      "balance_loss_clip": 1.04711115,
      "balance_loss_mlp": 1.02343428,
      "epoch": 0.8172909276739013,
      "flos": 21470128496640.0,
      "grad_norm": 1.5964624645970442,
      "language_loss": 0.6768629,
      "learning_rate": 3.3993343244864403e-07,
      "loss": 0.69882506,
      "num_input_tokens_seen": 146734425,
      "step": 6797,
      "time_per_iteration": 2.6271135807037354
    },
    {
      "auxiliary_loss_clip": 0.01167945,
      "auxiliary_loss_mlp": 0.01024669,
      "balance_loss_clip": 1.04557776,
      "balance_loss_mlp": 1.01750183,
      "epoch": 0.8174111705645404,
      "flos": 27599792417280.0,
      "grad_norm": 2.5175984517724084,
      "language_loss": 0.72686744,
      "learning_rate": 3.394991150920323e-07,
      "loss": 0.7487936,
      "num_input_tokens_seen": 146757545,
      "step": 6798,
      "time_per_iteration": 2.7137129306793213
    },
    {
      "auxiliary_loss_clip": 0.01182807,
      "auxiliary_loss_mlp": 0.01208625,
      "balance_loss_clip": 1.04399061,
      "balance_loss_mlp": 1.00075853,
      "epoch": 0.8175314134551794,
      "flos": 14064594508800.0,
      "grad_norm": 4.92030844749401,
      "language_loss": 0.74553061,
      "learning_rate": 3.3906504963208396e-07,
      "loss": 0.76944488,
      "num_input_tokens_seen": 146774240,
      "step": 6799,
      "time_per_iteration": 2.6376030445098877
    },
    {
      "auxiliary_loss_clip": 0.01174423,
      "auxiliary_loss_mlp": 0.0102282,
      "balance_loss_clip": 1.04393482,
      "balance_loss_mlp": 1.01553667,
      "epoch": 0.8176516563458186,
      "flos": 22708076780160.0,
      "grad_norm": 1.8594083074612169,
      "language_loss": 0.66706473,
      "learning_rate": 3.3863123613464774e-07,
      "loss": 0.6890372,
      "num_input_tokens_seen": 146793140,
      "step": 6800,
      "time_per_iteration": 2.750307559967041
    },
    {
      "auxiliary_loss_clip": 0.01172142,
      "auxiliary_loss_mlp": 0.01026458,
      "balance_loss_clip": 1.04022956,
      "balance_loss_mlp": 1.01907921,
      "epoch": 0.8177718992364577,
      "flos": 21945406279680.0,
      "grad_norm": 1.8692544170139775,
      "language_loss": 0.7509892,
      "learning_rate": 3.381976746655317e-07,
      "loss": 0.77297521,
      "num_input_tokens_seen": 146812895,
      "step": 6801,
      "time_per_iteration": 2.678812026977539
    },
    {
      "auxiliary_loss_clip": 0.01168242,
      "auxiliary_loss_mlp": 0.01025545,
      "balance_loss_clip": 1.04266453,
      "balance_loss_mlp": 1.01848817,
      "epoch": 0.8178921421270967,
      "flos": 22017443005440.0,
      "grad_norm": 2.1843757239921975,
      "language_loss": 0.67209446,
      "learning_rate": 3.3776436529050756e-07,
      "loss": 0.69403237,
      "num_input_tokens_seen": 146832445,
      "step": 6802,
      "time_per_iteration": 2.74851131439209
    },
    {
      "auxiliary_loss_clip": 0.01163259,
      "auxiliary_loss_mlp": 0.01024114,
      "balance_loss_clip": 1.04552865,
      "balance_loss_mlp": 1.01655006,
      "epoch": 0.8180123850177359,
      "flos": 33183111496320.0,
      "grad_norm": 2.105392807581491,
      "language_loss": 0.72826123,
      "learning_rate": 3.373313080753073e-07,
      "loss": 0.750135,
      "num_input_tokens_seen": 146856505,
      "step": 6803,
      "time_per_iteration": 2.7054805755615234
    },
    {
      "auxiliary_loss_clip": 0.0116354,
      "auxiliary_loss_mlp": 0.0102485,
      "balance_loss_clip": 1.04359663,
      "balance_loss_mlp": 1.01756036,
      "epoch": 0.8181326279083749,
      "flos": 22091167670400.0,
      "grad_norm": 1.6357364505439493,
      "language_loss": 0.77490628,
      "learning_rate": 3.3689850308562527e-07,
      "loss": 0.79679024,
      "num_input_tokens_seen": 146876950,
      "step": 6804,
      "time_per_iteration": 2.6420254707336426
    },
    {
      "auxiliary_loss_clip": 0.0116907,
      "auxiliary_loss_mlp": 0.01027445,
      "balance_loss_clip": 1.0431118,
      "balance_loss_mlp": 1.02006316,
      "epoch": 0.818252870799014,
      "flos": 15705747936000.0,
      "grad_norm": 1.788551512476819,
      "language_loss": 0.77405947,
      "learning_rate": 3.364659503871183e-07,
      "loss": 0.79602462,
      "num_input_tokens_seen": 146894885,
      "step": 6805,
      "time_per_iteration": 2.652836561203003
    },
    {
      "auxiliary_loss_clip": 0.01172275,
      "auxiliary_loss_mlp": 0.01025608,
      "balance_loss_clip": 1.04157281,
      "balance_loss_mlp": 1.01905107,
      "epoch": 0.8183731136896532,
      "flos": 18770687637120.0,
      "grad_norm": 2.0218993139979062,
      "language_loss": 0.83801007,
      "learning_rate": 3.3603365004540417e-07,
      "loss": 0.85998881,
      "num_input_tokens_seen": 146913180,
      "step": 6806,
      "time_per_iteration": 2.695363998413086
    },
    {
      "auxiliary_loss_clip": 0.01167173,
      "auxiliary_loss_mlp": 0.01025348,
      "balance_loss_clip": 1.04903316,
      "balance_loss_mlp": 1.01771021,
      "epoch": 0.8184933565802922,
      "flos": 26541792293760.0,
      "grad_norm": 1.832335766096588,
      "language_loss": 0.77179384,
      "learning_rate": 3.356016021260624e-07,
      "loss": 0.79371905,
      "num_input_tokens_seen": 146933510,
      "step": 6807,
      "time_per_iteration": 3.6100192070007324
    },
    {
      "auxiliary_loss_clip": 0.01171242,
      "auxiliary_loss_mlp": 0.01028686,
      "balance_loss_clip": 1.04644728,
      "balance_loss_mlp": 1.02107477,
      "epoch": 0.8186135994709313,
      "flos": 17530117660800.0,
      "grad_norm": 3.803358603908529,
      "language_loss": 0.65753281,
      "learning_rate": 3.35169806694634e-07,
      "loss": 0.67953211,
      "num_input_tokens_seen": 146951760,
      "step": 6808,
      "time_per_iteration": 2.5953216552734375
    },
    {
      "auxiliary_loss_clip": 0.01083037,
      "auxiliary_loss_mlp": 0.01001104,
      "balance_loss_clip": 1.01532137,
      "balance_loss_mlp": 1.00035846,
      "epoch": 0.8187338423615703,
      "flos": 63480300675840.0,
      "grad_norm": 0.7132250656052834,
      "language_loss": 0.6055035,
      "learning_rate": 3.3473826381662186e-07,
      "loss": 0.62634486,
      "num_input_tokens_seen": 147022900,
      "step": 6809,
      "time_per_iteration": 3.311962842941284
    },
    {
      "auxiliary_loss_clip": 0.01161905,
      "auxiliary_loss_mlp": 0.01023245,
      "balance_loss_clip": 1.04493141,
      "balance_loss_mlp": 1.01610398,
      "epoch": 0.8188540852522095,
      "flos": 17529974006400.0,
      "grad_norm": 1.8310244721538558,
      "language_loss": 0.8171103,
      "learning_rate": 3.3430697355749216e-07,
      "loss": 0.83896184,
      "num_input_tokens_seen": 147040590,
      "step": 6810,
      "time_per_iteration": 2.5552732944488525
    },
    {
      "auxiliary_loss_clip": 0.01174396,
      "auxiliary_loss_mlp": 0.01022212,
      "balance_loss_clip": 1.04104543,
      "balance_loss_mlp": 1.01428473,
      "epoch": 0.8189743281428485,
      "flos": 14392530702720.0,
      "grad_norm": 7.1740686365688315,
      "language_loss": 0.7570712,
      "learning_rate": 3.3387593598266907e-07,
      "loss": 0.77903724,
      "num_input_tokens_seen": 147057200,
      "step": 6811,
      "time_per_iteration": 2.6911473274230957
    },
    {
      "auxiliary_loss_clip": 0.01166373,
      "auxiliary_loss_mlp": 0.01024286,
      "balance_loss_clip": 1.0410831,
      "balance_loss_mlp": 1.01696086,
      "epoch": 0.8190945710334876,
      "flos": 25080479285760.0,
      "grad_norm": 1.7970583495275876,
      "language_loss": 0.78260016,
      "learning_rate": 3.3344515115754225e-07,
      "loss": 0.80450678,
      "num_input_tokens_seen": 147076180,
      "step": 6812,
      "time_per_iteration": 2.778174877166748
    },
    {
      "auxiliary_loss_clip": 0.01179536,
      "auxiliary_loss_mlp": 0.01021051,
      "balance_loss_clip": 1.04270577,
      "balance_loss_mlp": 1.01365137,
      "epoch": 0.8192148139241268,
      "flos": 21507152440320.0,
      "grad_norm": 2.467304095074109,
      "language_loss": 0.79862708,
      "learning_rate": 3.33014619147461e-07,
      "loss": 0.82063293,
      "num_input_tokens_seen": 147094205,
      "step": 6813,
      "time_per_iteration": 2.749818801879883
    },
    {
      "auxiliary_loss_clip": 0.01176551,
      "auxiliary_loss_mlp": 0.01026377,
      "balance_loss_clip": 1.04954267,
      "balance_loss_mlp": 1.01903343,
      "epoch": 0.8193350568147658,
      "flos": 23952166289280.0,
      "grad_norm": 2.7333380690069577,
      "language_loss": 0.71920002,
      "learning_rate": 3.325843400177362e-07,
      "loss": 0.7412293,
      "num_input_tokens_seen": 147115545,
      "step": 6814,
      "time_per_iteration": 3.6383519172668457
    },
    {
      "auxiliary_loss_clip": 0.01176202,
      "auxiliary_loss_mlp": 0.01207995,
      "balance_loss_clip": 1.0485642,
      "balance_loss_mlp": 1.00067306,
      "epoch": 0.8194552997054049,
      "flos": 20559469962240.0,
      "grad_norm": 1.7537737692117106,
      "language_loss": 0.73550415,
      "learning_rate": 3.32154313833642e-07,
      "loss": 0.75934613,
      "num_input_tokens_seen": 147135700,
      "step": 6815,
      "time_per_iteration": 2.604358196258545
    },
    {
      "auxiliary_loss_clip": 0.01168654,
      "auxiliary_loss_mlp": 0.01023931,
      "balance_loss_clip": 1.04699123,
      "balance_loss_mlp": 1.01600933,
      "epoch": 0.819575542596044,
      "flos": 26031753123840.0,
      "grad_norm": 2.119955106698867,
      "language_loss": 0.59578592,
      "learning_rate": 3.3172454066041164e-07,
      "loss": 0.61771172,
      "num_input_tokens_seen": 147155205,
      "step": 6816,
      "time_per_iteration": 3.625274658203125
    },
    {
      "auxiliary_loss_clip": 0.01179455,
      "auxiliary_loss_mlp": 0.01207341,
      "balance_loss_clip": 1.04253733,
      "balance_loss_mlp": 1.00074816,
      "epoch": 0.8196957854866831,
      "flos": 29096944220160.0,
      "grad_norm": 2.330260861517878,
      "language_loss": 0.7619369,
      "learning_rate": 3.3129502056324234e-07,
      "loss": 0.78580487,
      "num_input_tokens_seen": 147176570,
      "step": 6817,
      "time_per_iteration": 2.8006069660186768
    },
    {
      "auxiliary_loss_clip": 0.01104663,
      "auxiliary_loss_mlp": 0.0100228,
      "balance_loss_clip": 1.01420212,
      "balance_loss_mlp": 1.00141621,
      "epoch": 0.8198160283773221,
      "flos": 69033631898880.0,
      "grad_norm": 0.7917442845609493,
      "language_loss": 0.59742343,
      "learning_rate": 3.3086575360729165e-07,
      "loss": 0.61849296,
      "num_input_tokens_seen": 147234105,
      "step": 6818,
      "time_per_iteration": 3.0414998531341553
    },
    {
      "auxiliary_loss_clip": 0.0116993,
      "auxiliary_loss_mlp": 0.01027963,
      "balance_loss_clip": 1.04489195,
      "balance_loss_mlp": 1.02030945,
      "epoch": 0.8199362712679613,
      "flos": 16618058496000.0,
      "grad_norm": 2.4566712909022583,
      "language_loss": 0.7138043,
      "learning_rate": 3.3043673985767906e-07,
      "loss": 0.73578334,
      "num_input_tokens_seen": 147253170,
      "step": 6819,
      "time_per_iteration": 2.6457197666168213
    },
    {
      "auxiliary_loss_clip": 0.01165139,
      "auxiliary_loss_mlp": 0.01029117,
      "balance_loss_clip": 1.03977084,
      "balance_loss_mlp": 1.02108264,
      "epoch": 0.8200565141586004,
      "flos": 21757664868480.0,
      "grad_norm": 1.7396988110222669,
      "language_loss": 0.77455688,
      "learning_rate": 3.3000797937948564e-07,
      "loss": 0.79649949,
      "num_input_tokens_seen": 147271465,
      "step": 6820,
      "time_per_iteration": 3.531776189804077
    },
    {
      "auxiliary_loss_clip": 0.01083472,
      "auxiliary_loss_mlp": 0.01000839,
      "balance_loss_clip": 1.01074255,
      "balance_loss_mlp": 1.0000639,
      "epoch": 0.8201767570492394,
      "flos": 69807112392960.0,
      "grad_norm": 0.9474651574918097,
      "language_loss": 0.64938325,
      "learning_rate": 3.295794722377534e-07,
      "loss": 0.67022634,
      "num_input_tokens_seen": 147335070,
      "step": 6821,
      "time_per_iteration": 3.2719125747680664
    },
    {
      "auxiliary_loss_clip": 0.0116292,
      "auxiliary_loss_mlp": 0.01021329,
      "balance_loss_clip": 1.04429889,
      "balance_loss_mlp": 1.01455784,
      "epoch": 0.8202969999398786,
      "flos": 23111892455040.0,
      "grad_norm": 1.6013835302474355,
      "language_loss": 0.80043185,
      "learning_rate": 3.291512184974876e-07,
      "loss": 0.82227433,
      "num_input_tokens_seen": 147355460,
      "step": 6822,
      "time_per_iteration": 2.602734327316284
    },
    {
      "auxiliary_loss_clip": 0.0117011,
      "auxiliary_loss_mlp": 0.01022555,
      "balance_loss_clip": 1.04127526,
      "balance_loss_mlp": 1.0148536,
      "epoch": 0.8204172428305176,
      "flos": 28220616109440.0,
      "grad_norm": 2.466241872406661,
      "language_loss": 0.66766649,
      "learning_rate": 3.2872321822365346e-07,
      "loss": 0.68959308,
      "num_input_tokens_seen": 147375675,
      "step": 6823,
      "time_per_iteration": 2.6598715782165527
    },
    {
      "auxiliary_loss_clip": 0.01168981,
      "auxiliary_loss_mlp": 0.01020813,
      "balance_loss_clip": 1.04623199,
      "balance_loss_mlp": 1.01314485,
      "epoch": 0.8205374857211567,
      "flos": 20887011106560.0,
      "grad_norm": 1.9312415731436405,
      "language_loss": 0.73671091,
      "learning_rate": 3.282954714811783e-07,
      "loss": 0.75860882,
      "num_input_tokens_seen": 147394580,
      "step": 6824,
      "time_per_iteration": 2.7324817180633545
    },
    {
      "auxiliary_loss_clip": 0.01159329,
      "auxiliary_loss_mlp": 0.01024463,
      "balance_loss_clip": 1.04040587,
      "balance_loss_mlp": 1.01647651,
      "epoch": 0.8206577286117959,
      "flos": 13152140294400.0,
      "grad_norm": 2.404213226106812,
      "language_loss": 0.71061099,
      "learning_rate": 3.2786797833495093e-07,
      "loss": 0.732449,
      "num_input_tokens_seen": 147409935,
      "step": 6825,
      "time_per_iteration": 2.6081297397613525
    },
    {
      "auxiliary_loss_clip": 0.01165558,
      "auxiliary_loss_mlp": 0.01024278,
      "balance_loss_clip": 1.04753077,
      "balance_loss_mlp": 1.0175333,
      "epoch": 0.8207779715024349,
      "flos": 25265634917760.0,
      "grad_norm": 1.8442792112756308,
      "language_loss": 0.72706616,
      "learning_rate": 3.274407388498213e-07,
      "loss": 0.74896455,
      "num_input_tokens_seen": 147428065,
      "step": 6826,
      "time_per_iteration": 2.646162748336792
    },
    {
      "auxiliary_loss_clip": 0.01169456,
      "auxiliary_loss_mlp": 0.01025779,
      "balance_loss_clip": 1.04158688,
      "balance_loss_mlp": 1.01839674,
      "epoch": 0.820898214393074,
      "flos": 19610243199360.0,
      "grad_norm": 1.990368875825207,
      "language_loss": 0.74144864,
      "learning_rate": 3.270137530906021e-07,
      "loss": 0.76340103,
      "num_input_tokens_seen": 147447300,
      "step": 6827,
      "time_per_iteration": 2.659693956375122
    },
    {
      "auxiliary_loss_clip": 0.0116637,
      "auxiliary_loss_mlp": 0.01024414,
      "balance_loss_clip": 1.04199159,
      "balance_loss_mlp": 1.01735365,
      "epoch": 0.8210184572837131,
      "flos": 15596615439360.0,
      "grad_norm": 2.1107340814159237,
      "language_loss": 0.83608848,
      "learning_rate": 3.265870211220665e-07,
      "loss": 0.85799634,
      "num_input_tokens_seen": 147465135,
      "step": 6828,
      "time_per_iteration": 2.693033218383789
    },
    {
      "auxiliary_loss_clip": 0.01168053,
      "auxiliary_loss_mlp": 0.01031181,
      "balance_loss_clip": 1.04115462,
      "balance_loss_mlp": 1.02330756,
      "epoch": 0.8211387001743522,
      "flos": 20813932886400.0,
      "grad_norm": 1.8434766068763433,
      "language_loss": 0.8178792,
      "learning_rate": 3.2616054300894934e-07,
      "loss": 0.83987159,
      "num_input_tokens_seen": 147484585,
      "step": 6829,
      "time_per_iteration": 2.6621081829071045
    },
    {
      "auxiliary_loss_clip": 0.01160092,
      "auxiliary_loss_mlp": 0.01023341,
      "balance_loss_clip": 1.0432471,
      "balance_loss_mlp": 1.01597381,
      "epoch": 0.8212589430649913,
      "flos": 27704579368320.0,
      "grad_norm": 9.012948468724266,
      "language_loss": 0.84552073,
      "learning_rate": 3.2573431881594693e-07,
      "loss": 0.86735505,
      "num_input_tokens_seen": 147504130,
      "step": 6830,
      "time_per_iteration": 2.6781301498413086
    },
    {
      "auxiliary_loss_clip": 0.01179151,
      "auxiliary_loss_mlp": 0.01023796,
      "balance_loss_clip": 1.03866768,
      "balance_loss_mlp": 1.0162617,
      "epoch": 0.8213791859556304,
      "flos": 22455625017600.0,
      "grad_norm": 2.124009647497,
      "language_loss": 0.66366112,
      "learning_rate": 3.2530834860771663e-07,
      "loss": 0.68569058,
      "num_input_tokens_seen": 147523510,
      "step": 6831,
      "time_per_iteration": 2.839514970779419
    },
    {
      "auxiliary_loss_clip": 0.01169144,
      "auxiliary_loss_mlp": 0.01026004,
      "balance_loss_clip": 1.04457617,
      "balance_loss_mlp": 1.01784396,
      "epoch": 0.8214994288462695,
      "flos": 16654471908480.0,
      "grad_norm": 2.0065177658943316,
      "language_loss": 0.74247396,
      "learning_rate": 3.248826324488794e-07,
      "loss": 0.76442546,
      "num_input_tokens_seen": 147540805,
      "step": 6832,
      "time_per_iteration": 2.664440631866455
    },
    {
      "auxiliary_loss_clip": 0.01168395,
      "auxiliary_loss_mlp": 0.0102813,
      "balance_loss_clip": 1.0511322,
      "balance_loss_mlp": 1.02057505,
      "epoch": 0.8216196717369085,
      "flos": 25221787390080.0,
      "grad_norm": 1.7682734414946608,
      "language_loss": 0.87676227,
      "learning_rate": 3.244571704040138e-07,
      "loss": 0.89872748,
      "num_input_tokens_seen": 147560965,
      "step": 6833,
      "time_per_iteration": 3.602130651473999
    },
    {
      "auxiliary_loss_clip": 0.01166328,
      "auxiliary_loss_mlp": 0.01027398,
      "balance_loss_clip": 1.04325199,
      "balance_loss_mlp": 1.01876163,
      "epoch": 0.8217399146275477,
      "flos": 25371930240000.0,
      "grad_norm": 2.0126469011382127,
      "language_loss": 0.73721296,
      "learning_rate": 3.2403196253766374e-07,
      "loss": 0.75915027,
      "num_input_tokens_seen": 147580045,
      "step": 6834,
      "time_per_iteration": 2.6735739707946777
    },
    {
      "auxiliary_loss_clip": 0.01166917,
      "auxiliary_loss_mlp": 0.01025441,
      "balance_loss_clip": 1.04540479,
      "balance_loss_mlp": 1.01723957,
      "epoch": 0.8218601575181868,
      "flos": 25629625388160.0,
      "grad_norm": 3.870288515444234,
      "language_loss": 0.79460233,
      "learning_rate": 3.2360700891433254e-07,
      "loss": 0.81652588,
      "num_input_tokens_seen": 147599070,
      "step": 6835,
      "time_per_iteration": 2.6729025840759277
    },
    {
      "auxiliary_loss_clip": 0.01091482,
      "auxiliary_loss_mlp": 0.01001631,
      "balance_loss_clip": 1.01389444,
      "balance_loss_mlp": 1.00085568,
      "epoch": 0.8219804004088258,
      "flos": 67660229427840.0,
      "grad_norm": 0.7902618454229888,
      "language_loss": 0.57253212,
      "learning_rate": 3.231823095984847e-07,
      "loss": 0.59346324,
      "num_input_tokens_seen": 147653710,
      "step": 6836,
      "time_per_iteration": 3.1348817348480225
    },
    {
      "auxiliary_loss_clip": 0.01170825,
      "auxiliary_loss_mlp": 0.01022099,
      "balance_loss_clip": 1.04458356,
      "balance_loss_mlp": 1.01494908,
      "epoch": 0.822100643299465,
      "flos": 19464266327040.0,
      "grad_norm": 2.493756221097206,
      "language_loss": 0.76170993,
      "learning_rate": 3.2275786465454814e-07,
      "loss": 0.78363913,
      "num_input_tokens_seen": 147670360,
      "step": 6837,
      "time_per_iteration": 2.666085720062256
    },
    {
      "auxiliary_loss_clip": 0.01171674,
      "auxiliary_loss_mlp": 0.01022209,
      "balance_loss_clip": 1.04267359,
      "balance_loss_mlp": 1.01513147,
      "epoch": 0.822220886190104,
      "flos": 24681368292480.0,
      "grad_norm": 1.669793480436262,
      "language_loss": 0.75639141,
      "learning_rate": 3.2233367414690917e-07,
      "loss": 0.77833021,
      "num_input_tokens_seen": 147692550,
      "step": 6838,
      "time_per_iteration": 2.8032844066619873
    },
    {
      "auxiliary_loss_clip": 0.01169782,
      "auxiliary_loss_mlp": 0.01022971,
      "balance_loss_clip": 1.04049718,
      "balance_loss_mlp": 1.01596475,
      "epoch": 0.8223411290807431,
      "flos": 27819062991360.0,
      "grad_norm": 2.4383432313240254,
      "language_loss": 0.84865034,
      "learning_rate": 3.219097381399183e-07,
      "loss": 0.87057781,
      "num_input_tokens_seen": 147709725,
      "step": 6839,
      "time_per_iteration": 2.735483407974243
    },
    {
      "auxiliary_loss_clip": 0.01178687,
      "auxiliary_loss_mlp": 0.01023072,
      "balance_loss_clip": 1.0455761,
      "balance_loss_mlp": 1.01633382,
      "epoch": 0.8224613719713821,
      "flos": 23218546913280.0,
      "grad_norm": 3.3067881534416856,
      "language_loss": 0.8093977,
      "learning_rate": 3.2148605669788584e-07,
      "loss": 0.8314153,
      "num_input_tokens_seen": 147729615,
      "step": 6840,
      "time_per_iteration": 3.5530481338500977
    },
    {
      "auxiliary_loss_clip": 0.01174063,
      "auxiliary_loss_mlp": 0.01023655,
      "balance_loss_clip": 1.04630911,
      "balance_loss_mlp": 1.01629412,
      "epoch": 0.8225816148620213,
      "flos": 15706250726400.0,
      "grad_norm": 3.1128771677119915,
      "language_loss": 0.77498186,
      "learning_rate": 3.2106262988508405e-07,
      "loss": 0.79695904,
      "num_input_tokens_seen": 147747665,
      "step": 6841,
      "time_per_iteration": 2.609637498855591
    },
    {
      "auxiliary_loss_clip": 0.01174155,
      "auxiliary_loss_mlp": 0.01024113,
      "balance_loss_clip": 1.04477477,
      "balance_loss_mlp": 1.01660252,
      "epoch": 0.8227018577526604,
      "flos": 18515111391360.0,
      "grad_norm": 11.621989196575377,
      "language_loss": 0.74441195,
      "learning_rate": 3.206394577657465e-07,
      "loss": 0.76639462,
      "num_input_tokens_seen": 147765445,
      "step": 6842,
      "time_per_iteration": 2.6310250759124756
    },
    {
      "auxiliary_loss_clip": 0.01174838,
      "auxiliary_loss_mlp": 0.0102718,
      "balance_loss_clip": 1.04822767,
      "balance_loss_mlp": 1.01938319,
      "epoch": 0.8228221006432994,
      "flos": 22236785406720.0,
      "grad_norm": 2.3721227285327346,
      "language_loss": 0.7279774,
      "learning_rate": 3.202165404040675e-07,
      "loss": 0.7499975,
      "num_input_tokens_seen": 147783365,
      "step": 6843,
      "time_per_iteration": 3.5321972370147705
    },
    {
      "auxiliary_loss_clip": 0.01179948,
      "auxiliary_loss_mlp": 0.01024662,
      "balance_loss_clip": 1.0417155,
      "balance_loss_mlp": 1.01693094,
      "epoch": 0.8229423435339386,
      "flos": 24097532630400.0,
      "grad_norm": 2.2771302786338756,
      "language_loss": 0.74788851,
      "learning_rate": 3.1979387786420396e-07,
      "loss": 0.76993465,
      "num_input_tokens_seen": 147803605,
      "step": 6844,
      "time_per_iteration": 2.71639347076416
    },
    {
      "auxiliary_loss_clip": 0.011741,
      "auxiliary_loss_mlp": 0.0102107,
      "balance_loss_clip": 1.04276848,
      "balance_loss_mlp": 1.01404572,
      "epoch": 0.8230625864245776,
      "flos": 23878549365120.0,
      "grad_norm": 2.03105809200877,
      "language_loss": 0.82470274,
      "learning_rate": 3.1937147021027346e-07,
      "loss": 0.84665442,
      "num_input_tokens_seen": 147822060,
      "step": 6845,
      "time_per_iteration": 2.779191017150879
    },
    {
      "auxiliary_loss_clip": 0.01167218,
      "auxiliary_loss_mlp": 0.01021915,
      "balance_loss_clip": 1.0454303,
      "balance_loss_mlp": 1.01506305,
      "epoch": 0.8231828293152167,
      "flos": 16581106379520.0,
      "grad_norm": 5.424920106521118,
      "language_loss": 0.76664603,
      "learning_rate": 3.189493175063547e-07,
      "loss": 0.78853738,
      "num_input_tokens_seen": 147839295,
      "step": 6846,
      "time_per_iteration": 3.6055662631988525
    },
    {
      "auxiliary_loss_clip": 0.01172669,
      "auxiliary_loss_mlp": 0.01025188,
      "balance_loss_clip": 1.04533243,
      "balance_loss_mlp": 1.01739168,
      "epoch": 0.8233030722058559,
      "flos": 18880071528960.0,
      "grad_norm": 2.4740355154656757,
      "language_loss": 0.67241806,
      "learning_rate": 3.1852741981648776e-07,
      "loss": 0.69439662,
      "num_input_tokens_seen": 147857945,
      "step": 6847,
      "time_per_iteration": 2.69881272315979
    },
    {
      "auxiliary_loss_clip": 0.01160909,
      "auxiliary_loss_mlp": 0.01026769,
      "balance_loss_clip": 1.04134083,
      "balance_loss_mlp": 1.01924717,
      "epoch": 0.8234233150964949,
      "flos": 28439024757120.0,
      "grad_norm": 2.089958387875075,
      "language_loss": 0.70366961,
      "learning_rate": 3.1810577720467404e-07,
      "loss": 0.72554636,
      "num_input_tokens_seen": 147879675,
      "step": 6848,
      "time_per_iteration": 2.705634593963623
    },
    {
      "auxiliary_loss_clip": 0.01175259,
      "auxiliary_loss_mlp": 0.01022793,
      "balance_loss_clip": 1.04611802,
      "balance_loss_mlp": 1.01519632,
      "epoch": 0.823543557987134,
      "flos": 33765941577600.0,
      "grad_norm": 1.459437681136115,
      "language_loss": 0.56714821,
      "learning_rate": 3.176843897348769e-07,
      "loss": 0.58912867,
      "num_input_tokens_seen": 147902870,
      "step": 6849,
      "time_per_iteration": 2.835833787918091
    },
    {
      "auxiliary_loss_clip": 0.01169025,
      "auxiliary_loss_mlp": 0.01024224,
      "balance_loss_clip": 1.0447464,
      "balance_loss_mlp": 1.01640368,
      "epoch": 0.8236638008777731,
      "flos": 17092366611840.0,
      "grad_norm": 2.668542494249972,
      "language_loss": 0.75818616,
      "learning_rate": 3.1726325747102034e-07,
      "loss": 0.78011858,
      "num_input_tokens_seen": 147921245,
      "step": 6850,
      "time_per_iteration": 2.6349852085113525
    },
    {
      "auxiliary_loss_clip": 0.01169696,
      "auxiliary_loss_mlp": 0.01022954,
      "balance_loss_clip": 1.03701425,
      "balance_loss_mlp": 1.01527071,
      "epoch": 0.8237840437684122,
      "flos": 61639982334720.0,
      "grad_norm": 1.512601929023528,
      "language_loss": 0.64119959,
      "learning_rate": 3.1684238047698974e-07,
      "loss": 0.66312611,
      "num_input_tokens_seen": 147949515,
      "step": 6851,
      "time_per_iteration": 3.0905754566192627
    },
    {
      "auxiliary_loss_clip": 0.0117613,
      "auxiliary_loss_mlp": 0.01026381,
      "balance_loss_clip": 1.04614305,
      "balance_loss_mlp": 1.01872146,
      "epoch": 0.8239042866590512,
      "flos": 27309023821440.0,
      "grad_norm": 2.1356239010109475,
      "language_loss": 0.53039128,
      "learning_rate": 3.1642175881663155e-07,
      "loss": 0.55241638,
      "num_input_tokens_seen": 147969245,
      "step": 6852,
      "time_per_iteration": 2.7499642372131348
    },
    {
      "auxiliary_loss_clip": 0.01165182,
      "auxiliary_loss_mlp": 0.01020944,
      "balance_loss_clip": 1.04695606,
      "balance_loss_mlp": 1.01404452,
      "epoch": 0.8240245295496904,
      "flos": 21726351187200.0,
      "grad_norm": 2.164776079672991,
      "language_loss": 0.83741951,
      "learning_rate": 3.160013925537537e-07,
      "loss": 0.85928077,
      "num_input_tokens_seen": 147990080,
      "step": 6853,
      "time_per_iteration": 2.604522943496704
    },
    {
      "auxiliary_loss_clip": 0.01177976,
      "auxiliary_loss_mlp": 0.01021531,
      "balance_loss_clip": 1.04273558,
      "balance_loss_mlp": 1.01381874,
      "epoch": 0.8241447724403295,
      "flos": 20009318279040.0,
      "grad_norm": 3.395747259811684,
      "language_loss": 0.75613129,
      "learning_rate": 3.155812817521266e-07,
      "loss": 0.77812636,
      "num_input_tokens_seen": 148010455,
      "step": 6854,
      "time_per_iteration": 2.811920642852783
    },
    {
      "auxiliary_loss_clip": 0.01175739,
      "auxiliary_loss_mlp": 0.01024533,
      "balance_loss_clip": 1.04689145,
      "balance_loss_mlp": 1.01712465,
      "epoch": 0.8242650153309685,
      "flos": 22272983337600.0,
      "grad_norm": 2.2253676311295654,
      "language_loss": 0.77915162,
      "learning_rate": 3.151614264754787e-07,
      "loss": 0.80115438,
      "num_input_tokens_seen": 148028400,
      "step": 6855,
      "time_per_iteration": 2.6655633449554443
    },
    {
      "auxiliary_loss_clip": 0.01166676,
      "auxiliary_loss_mlp": 0.0102365,
      "balance_loss_clip": 1.04578567,
      "balance_loss_mlp": 1.0162766,
      "epoch": 0.8243852582216077,
      "flos": 22309971367680.0,
      "grad_norm": 2.3685341084766924,
      "language_loss": 0.79674435,
      "learning_rate": 3.147418267875035e-07,
      "loss": 0.81864762,
      "num_input_tokens_seen": 148046530,
      "step": 6856,
      "time_per_iteration": 2.6574745178222656
    },
    {
      "auxiliary_loss_clip": 0.01178832,
      "auxiliary_loss_mlp": 0.01207649,
      "balance_loss_clip": 1.03997827,
      "balance_loss_mlp": 1.0007385,
      "epoch": 0.8245055011122467,
      "flos": 24645421756800.0,
      "grad_norm": 5.390059127246155,
      "language_loss": 0.66131806,
      "learning_rate": 3.1432248275185315e-07,
      "loss": 0.68518287,
      "num_input_tokens_seen": 148067040,
      "step": 6857,
      "time_per_iteration": 2.793978452682495
    },
    {
      "auxiliary_loss_clip": 0.01167052,
      "auxiliary_loss_mlp": 0.01023667,
      "balance_loss_clip": 1.04670537,
      "balance_loss_mlp": 1.01628757,
      "epoch": 0.8246257440028858,
      "flos": 17487275713920.0,
      "grad_norm": 4.3324159542876455,
      "language_loss": 0.77391541,
      "learning_rate": 3.139033944321412e-07,
      "loss": 0.79582262,
      "num_input_tokens_seen": 148084400,
      "step": 6858,
      "time_per_iteration": 2.7720324993133545
    },
    {
      "auxiliary_loss_clip": 0.01171575,
      "auxiliary_loss_mlp": 0.01020778,
      "balance_loss_clip": 1.04550052,
      "balance_loss_mlp": 1.01365209,
      "epoch": 0.824745986893525,
      "flos": 25010130499200.0,
      "grad_norm": 1.7074659897367868,
      "language_loss": 0.79138088,
      "learning_rate": 3.1348456189194507e-07,
      "loss": 0.81330442,
      "num_input_tokens_seen": 148104860,
      "step": 6859,
      "time_per_iteration": 2.744678258895874
    },
    {
      "auxiliary_loss_clip": 0.01164891,
      "auxiliary_loss_mlp": 0.01021763,
      "balance_loss_clip": 1.04022264,
      "balance_loss_mlp": 1.0141933,
      "epoch": 0.824866229784164,
      "flos": 18772698798720.0,
      "grad_norm": 1.8528520715705539,
      "language_loss": 0.83032286,
      "learning_rate": 3.1306598519479876e-07,
      "loss": 0.85218936,
      "num_input_tokens_seen": 148124680,
      "step": 6860,
      "time_per_iteration": 3.860123872756958
    },
    {
      "auxiliary_loss_clip": 0.01166216,
      "auxiliary_loss_mlp": 0.01020734,
      "balance_loss_clip": 1.04263353,
      "balance_loss_mlp": 1.01378727,
      "epoch": 0.8249864726748031,
      "flos": 23842171866240.0,
      "grad_norm": 1.641994858322648,
      "language_loss": 0.78207779,
      "learning_rate": 3.1264766440420177e-07,
      "loss": 0.80394739,
      "num_input_tokens_seen": 148147150,
      "step": 6861,
      "time_per_iteration": 2.727728843688965
    },
    {
      "auxiliary_loss_clip": 0.01167375,
      "auxiliary_loss_mlp": 0.01021118,
      "balance_loss_clip": 1.0468235,
      "balance_loss_mlp": 1.0138402,
      "epoch": 0.8251067155654422,
      "flos": 20303103617280.0,
      "grad_norm": 2.1216032298413188,
      "language_loss": 0.69337285,
      "learning_rate": 3.122295995836124e-07,
      "loss": 0.71525776,
      "num_input_tokens_seen": 148167020,
      "step": 6862,
      "time_per_iteration": 2.6643221378326416
    },
    {
      "auxiliary_loss_clip": 0.01172698,
      "auxiliary_loss_mlp": 0.01023146,
      "balance_loss_clip": 1.04339015,
      "balance_loss_mlp": 1.01498008,
      "epoch": 0.8252269584560813,
      "flos": 25009699536000.0,
      "grad_norm": 2.132963684568733,
      "language_loss": 0.7796644,
      "learning_rate": 3.118117907964508e-07,
      "loss": 0.80162287,
      "num_input_tokens_seen": 148188965,
      "step": 6863,
      "time_per_iteration": 2.7351531982421875
    },
    {
      "auxiliary_loss_clip": 0.01181372,
      "auxiliary_loss_mlp": 0.01024408,
      "balance_loss_clip": 1.04352605,
      "balance_loss_mlp": 1.01721716,
      "epoch": 0.8253472013467203,
      "flos": 17128564542720.0,
      "grad_norm": 1.896812610610129,
      "language_loss": 0.8017509,
      "learning_rate": 3.1139423810609856e-07,
      "loss": 0.82380867,
      "num_input_tokens_seen": 148205660,
      "step": 6864,
      "time_per_iteration": 2.655909776687622
    },
    {
      "auxiliary_loss_clip": 0.01165406,
      "auxiliary_loss_mlp": 0.01024949,
      "balance_loss_clip": 1.04467964,
      "balance_loss_mlp": 1.0170331,
      "epoch": 0.8254674442373595,
      "flos": 22414794232320.0,
      "grad_norm": 2.0298141627533495,
      "language_loss": 0.7562865,
      "learning_rate": 3.1097694157589714e-07,
      "loss": 0.77819002,
      "num_input_tokens_seen": 148225545,
      "step": 6865,
      "time_per_iteration": 2.651899576187134
    },
    {
      "auxiliary_loss_clip": 0.01167235,
      "auxiliary_loss_mlp": 0.01027145,
      "balance_loss_clip": 1.04669356,
      "balance_loss_mlp": 1.01943243,
      "epoch": 0.8255876871279986,
      "flos": 24786765774720.0,
      "grad_norm": 2.837415153050985,
      "language_loss": 0.76467609,
      "learning_rate": 3.105599012691511e-07,
      "loss": 0.7866199,
      "num_input_tokens_seen": 148243975,
      "step": 6866,
      "time_per_iteration": 2.7283074855804443
    },
    {
      "auxiliary_loss_clip": 0.01166372,
      "auxiliary_loss_mlp": 0.01023764,
      "balance_loss_clip": 1.04552007,
      "balance_loss_mlp": 1.01675439,
      "epoch": 0.8257079300186376,
      "flos": 27455431656960.0,
      "grad_norm": 1.5586516935327395,
      "language_loss": 0.82446504,
      "learning_rate": 3.101431172491249e-07,
      "loss": 0.84636635,
      "num_input_tokens_seen": 148265520,
      "step": 6867,
      "time_per_iteration": 3.5761218070983887
    },
    {
      "auxiliary_loss_clip": 0.01177949,
      "auxiliary_loss_mlp": 0.01207834,
      "balance_loss_clip": 1.04153061,
      "balance_loss_mlp": 1.00064218,
      "epoch": 0.8258281729092768,
      "flos": 16471866142080.0,
      "grad_norm": 2.1225968994680238,
      "language_loss": 0.7185024,
      "learning_rate": 3.097265895790444e-07,
      "loss": 0.74236023,
      "num_input_tokens_seen": 148283730,
      "step": 6868,
      "time_per_iteration": 2.803138017654419
    },
    {
      "auxiliary_loss_clip": 0.01176139,
      "auxiliary_loss_mlp": 0.01021714,
      "balance_loss_clip": 1.04260874,
      "balance_loss_mlp": 1.01474619,
      "epoch": 0.8259484157999158,
      "flos": 21433822824960.0,
      "grad_norm": 2.6900631177048426,
      "language_loss": 0.83199334,
      "learning_rate": 3.093103183220962e-07,
      "loss": 0.8539719,
      "num_input_tokens_seen": 148303775,
      "step": 6869,
      "time_per_iteration": 2.698693037033081
    },
    {
      "auxiliary_loss_clip": 0.01070689,
      "auxiliary_loss_mlp": 0.01002761,
      "balance_loss_clip": 1.01133204,
      "balance_loss_mlp": 1.00200653,
      "epoch": 0.8260686586905549,
      "flos": 58322342453760.0,
      "grad_norm": 0.8429501104276744,
      "language_loss": 0.5930227,
      "learning_rate": 3.0889430354142796e-07,
      "loss": 0.61375725,
      "num_input_tokens_seen": 148365285,
      "step": 6870,
      "time_per_iteration": 4.023713111877441
    },
    {
      "auxiliary_loss_clip": 0.01179375,
      "auxiliary_loss_mlp": 0.01025213,
      "balance_loss_clip": 1.04303503,
      "balance_loss_mlp": 1.01752353,
      "epoch": 0.826188901581194,
      "flos": 27527288814720.0,
      "grad_norm": 2.0837492687662076,
      "language_loss": 0.70588458,
      "learning_rate": 3.084785453001497e-07,
      "loss": 0.72793055,
      "num_input_tokens_seen": 148386200,
      "step": 6871,
      "time_per_iteration": 2.760721206665039
    },
    {
      "auxiliary_loss_clip": 0.0117072,
      "auxiliary_loss_mlp": 0.0120777,
      "balance_loss_clip": 1.04654837,
      "balance_loss_mlp": 1.0007621,
      "epoch": 0.8263091444718331,
      "flos": 23696051339520.0,
      "grad_norm": 2.4907197578378226,
      "language_loss": 0.81887734,
      "learning_rate": 3.080630436613314e-07,
      "loss": 0.84266227,
      "num_input_tokens_seen": 148403970,
      "step": 6872,
      "time_per_iteration": 2.622114419937134
    },
    {
      "auxiliary_loss_clip": 0.01161756,
      "auxiliary_loss_mlp": 0.01028843,
      "balance_loss_clip": 1.04449975,
      "balance_loss_mlp": 1.02125549,
      "epoch": 0.8264293873624722,
      "flos": 17165157523200.0,
      "grad_norm": 2.269873362797605,
      "language_loss": 0.86046433,
      "learning_rate": 3.076477986880039e-07,
      "loss": 0.88237035,
      "num_input_tokens_seen": 148421765,
      "step": 6873,
      "time_per_iteration": 2.6580650806427
    },
    {
      "auxiliary_loss_clip": 0.01170827,
      "auxiliary_loss_mlp": 0.01023068,
      "balance_loss_clip": 1.04523683,
      "balance_loss_mlp": 1.01564455,
      "epoch": 0.8265496302531112,
      "flos": 24098645952000.0,
      "grad_norm": 2.195108202855107,
      "language_loss": 0.69948,
      "learning_rate": 3.0723281044315986e-07,
      "loss": 0.72141898,
      "num_input_tokens_seen": 148443720,
      "step": 6874,
      "time_per_iteration": 3.527837038040161
    },
    {
      "auxiliary_loss_clip": 0.01161287,
      "auxiliary_loss_mlp": 0.01024549,
      "balance_loss_clip": 1.04452825,
      "balance_loss_mlp": 1.01763439,
      "epoch": 0.8266698731437504,
      "flos": 14099894599680.0,
      "grad_norm": 2.414112506099974,
      "language_loss": 0.7613709,
      "learning_rate": 3.068180789897521e-07,
      "loss": 0.78322935,
      "num_input_tokens_seen": 148462130,
      "step": 6875,
      "time_per_iteration": 2.600313186645508
    },
    {
      "auxiliary_loss_clip": 0.01173558,
      "auxiliary_loss_mlp": 0.0102413,
      "balance_loss_clip": 1.04562473,
      "balance_loss_mlp": 1.0163846,
      "epoch": 0.8267901160343895,
      "flos": 30777563715840.0,
      "grad_norm": 1.4838869226491171,
      "language_loss": 0.81829232,
      "learning_rate": 3.064036043906966e-07,
      "loss": 0.84026921,
      "num_input_tokens_seen": 148485570,
      "step": 6876,
      "time_per_iteration": 2.7328972816467285
    },
    {
      "auxiliary_loss_clip": 0.01182466,
      "auxiliary_loss_mlp": 0.0102682,
      "balance_loss_clip": 1.04358006,
      "balance_loss_mlp": 1.01885676,
      "epoch": 0.8269103589250285,
      "flos": 40624915242240.0,
      "grad_norm": 2.3136228684834523,
      "language_loss": 0.68365228,
      "learning_rate": 3.059893867088668e-07,
      "loss": 0.7057451,
      "num_input_tokens_seen": 148509715,
      "step": 6877,
      "time_per_iteration": 2.854830265045166
    },
    {
      "auxiliary_loss_clip": 0.01166117,
      "auxiliary_loss_mlp": 0.01026335,
      "balance_loss_clip": 1.04565108,
      "balance_loss_mlp": 1.01931024,
      "epoch": 0.8270306018156677,
      "flos": 30263645877120.0,
      "grad_norm": 2.495358004635591,
      "language_loss": 0.67252827,
      "learning_rate": 3.055754260071004e-07,
      "loss": 0.69445288,
      "num_input_tokens_seen": 148532010,
      "step": 6878,
      "time_per_iteration": 2.745008707046509
    },
    {
      "auxiliary_loss_clip": 0.01169234,
      "auxiliary_loss_mlp": 0.01024592,
      "balance_loss_clip": 1.04652095,
      "balance_loss_mlp": 1.01774001,
      "epoch": 0.8271508447063067,
      "flos": 25226599812480.0,
      "grad_norm": 1.8915482226659508,
      "language_loss": 0.73982167,
      "learning_rate": 3.051617223481948e-07,
      "loss": 0.76175988,
      "num_input_tokens_seen": 148553330,
      "step": 6879,
      "time_per_iteration": 2.723937749862671
    },
    {
      "auxiliary_loss_clip": 0.01187082,
      "auxiliary_loss_mlp": 0.01028673,
      "balance_loss_clip": 1.04436243,
      "balance_loss_mlp": 1.02075446,
      "epoch": 0.8272710875969458,
      "flos": 17566602900480.0,
      "grad_norm": 1.9363511956746071,
      "language_loss": 0.75541866,
      "learning_rate": 3.047482757949078e-07,
      "loss": 0.77757621,
      "num_input_tokens_seen": 148570960,
      "step": 6880,
      "time_per_iteration": 2.776512384414673
    },
    {
      "auxiliary_loss_clip": 0.011694,
      "auxiliary_loss_mlp": 0.0120709,
      "balance_loss_clip": 1.04102111,
      "balance_loss_mlp": 1.0007472,
      "epoch": 0.827391330487585,
      "flos": 19755465886080.0,
      "grad_norm": 3.570797147426798,
      "language_loss": 0.86071789,
      "learning_rate": 3.043350864099605e-07,
      "loss": 0.88448274,
      "num_input_tokens_seen": 148589520,
      "step": 6881,
      "time_per_iteration": 2.6769187450408936
    },
    {
      "auxiliary_loss_clip": 0.01169712,
      "auxiliary_loss_mlp": 0.01022028,
      "balance_loss_clip": 1.04353094,
      "balance_loss_mlp": 1.01459825,
      "epoch": 0.827511573378224,
      "flos": 16835174254080.0,
      "grad_norm": 2.2577157590928842,
      "language_loss": 0.80965132,
      "learning_rate": 3.039221542560315e-07,
      "loss": 0.83156872,
      "num_input_tokens_seen": 148606085,
      "step": 6882,
      "time_per_iteration": 2.652085304260254
    },
    {
      "auxiliary_loss_clip": 0.01169355,
      "auxiliary_loss_mlp": 0.01020543,
      "balance_loss_clip": 1.04731226,
      "balance_loss_mlp": 1.01306009,
      "epoch": 0.8276318162688631,
      "flos": 18369242259840.0,
      "grad_norm": 2.0258348083866786,
      "language_loss": 0.73854387,
      "learning_rate": 3.0350947939576356e-07,
      "loss": 0.76044285,
      "num_input_tokens_seen": 148625240,
      "step": 6883,
      "time_per_iteration": 2.6283390522003174
    },
    {
      "auxiliary_loss_clip": 0.01176826,
      "auxiliary_loss_mlp": 0.01027967,
      "balance_loss_clip": 1.0480442,
      "balance_loss_mlp": 1.0198338,
      "epoch": 0.8277520591595022,
      "flos": 19352691705600.0,
      "grad_norm": 1.8676797862257843,
      "language_loss": 0.72508824,
      "learning_rate": 3.0309706189175876e-07,
      "loss": 0.74713618,
      "num_input_tokens_seen": 148645075,
      "step": 6884,
      "time_per_iteration": 2.713585138320923
    },
    {
      "auxiliary_loss_clip": 0.01078631,
      "auxiliary_loss_mlp": 0.01002782,
      "balance_loss_clip": 1.01189542,
      "balance_loss_mlp": 1.0019778,
      "epoch": 0.8278723020501413,
      "flos": 67918858329600.0,
      "grad_norm": 0.7579332699876977,
      "language_loss": 0.57323354,
      "learning_rate": 3.0268490180658045e-07,
      "loss": 0.59404767,
      "num_input_tokens_seen": 148707855,
      "step": 6885,
      "time_per_iteration": 3.2034683227539062
    },
    {
      "auxiliary_loss_clip": 0.01171313,
      "auxiliary_loss_mlp": 0.01022852,
      "balance_loss_clip": 1.05027246,
      "balance_loss_mlp": 1.01540101,
      "epoch": 0.8279925449407803,
      "flos": 18185738653440.0,
      "grad_norm": 2.306230483614143,
      "language_loss": 0.79936516,
      "learning_rate": 3.0227299920275305e-07,
      "loss": 0.82130682,
      "num_input_tokens_seen": 148724170,
      "step": 6886,
      "time_per_iteration": 2.5949227809906006
    },
    {
      "auxiliary_loss_clip": 0.01180576,
      "auxiliary_loss_mlp": 0.01027515,
      "balance_loss_clip": 1.04568458,
      "balance_loss_mlp": 1.01926565,
      "epoch": 0.8281127878314195,
      "flos": 20631434860800.0,
      "grad_norm": 3.0145078081803267,
      "language_loss": 0.85757375,
      "learning_rate": 3.018613541427613e-07,
      "loss": 0.87965465,
      "num_input_tokens_seen": 148743690,
      "step": 6887,
      "time_per_iteration": 3.718418836593628
    },
    {
      "auxiliary_loss_clip": 0.0116583,
      "auxiliary_loss_mlp": 0.01024684,
      "balance_loss_clip": 1.04596913,
      "balance_loss_mlp": 1.0173409,
      "epoch": 0.8282330307220586,
      "flos": 18004282122240.0,
      "grad_norm": 1.642568483318732,
      "language_loss": 0.73931098,
      "learning_rate": 3.0144996668905243e-07,
      "loss": 0.76121604,
      "num_input_tokens_seen": 148761070,
      "step": 6888,
      "time_per_iteration": 2.616854429244995
    },
    {
      "auxiliary_loss_clip": 0.0118205,
      "auxiliary_loss_mlp": 0.0120753,
      "balance_loss_clip": 1.0377171,
      "balance_loss_mlp": 1.00074673,
      "epoch": 0.8283532736126976,
      "flos": 20084120352000.0,
      "grad_norm": 2.0023038595283627,
      "language_loss": 0.82322794,
      "learning_rate": 3.010388369040331e-07,
      "loss": 0.84712374,
      "num_input_tokens_seen": 148779730,
      "step": 6889,
      "time_per_iteration": 2.797945261001587
    },
    {
      "auxiliary_loss_clip": 0.01170898,
      "auxiliary_loss_mlp": 0.01024364,
      "balance_loss_clip": 1.04686284,
      "balance_loss_mlp": 1.01706576,
      "epoch": 0.8284735165033368,
      "flos": 31868421805440.0,
      "grad_norm": 1.6073675516147565,
      "language_loss": 0.83168149,
      "learning_rate": 3.0062796485007156e-07,
      "loss": 0.85363406,
      "num_input_tokens_seen": 148800670,
      "step": 6890,
      "time_per_iteration": 2.745530605316162
    },
    {
      "auxiliary_loss_clip": 0.01167765,
      "auxiliary_loss_mlp": 0.01208146,
      "balance_loss_clip": 1.04740357,
      "balance_loss_mlp": 1.00076818,
      "epoch": 0.8285937593939758,
      "flos": 26651319840000.0,
      "grad_norm": 5.068932671710271,
      "language_loss": 0.65533698,
      "learning_rate": 3.002173505894965e-07,
      "loss": 0.6790961,
      "num_input_tokens_seen": 148819820,
      "step": 6891,
      "time_per_iteration": 2.635438919067383
    },
    {
      "auxiliary_loss_clip": 0.01174262,
      "auxiliary_loss_mlp": 0.01026065,
      "balance_loss_clip": 1.04478574,
      "balance_loss_mlp": 1.01807821,
      "epoch": 0.8287140022846149,
      "flos": 20193683811840.0,
      "grad_norm": 2.5412721320898,
      "language_loss": 0.62772548,
      "learning_rate": 2.998069941845973e-07,
      "loss": 0.64972878,
      "num_input_tokens_seen": 148838890,
      "step": 6892,
      "time_per_iteration": 2.6154427528381348
    },
    {
      "auxiliary_loss_clip": 0.01063227,
      "auxiliary_loss_mlp": 0.01001377,
      "balance_loss_clip": 1.0105983,
      "balance_loss_mlp": 1.00060833,
      "epoch": 0.8288342451752541,
      "flos": 70755980019840.0,
      "grad_norm": 1.2389315702455532,
      "language_loss": 0.57469082,
      "learning_rate": 2.993968956976258e-07,
      "loss": 0.59533691,
      "num_input_tokens_seen": 148906635,
      "step": 6893,
      "time_per_iteration": 3.2569427490234375
    },
    {
      "auxiliary_loss_clip": 0.0117373,
      "auxiliary_loss_mlp": 0.01022913,
      "balance_loss_clip": 1.04904985,
      "balance_loss_mlp": 1.01426399,
      "epoch": 0.8289544880658931,
      "flos": 24572235795840.0,
      "grad_norm": 4.320106316270634,
      "language_loss": 0.7030617,
      "learning_rate": 2.9898705519079313e-07,
      "loss": 0.72502816,
      "num_input_tokens_seen": 148925740,
      "step": 6894,
      "time_per_iteration": 3.595019817352295
    },
    {
      "auxiliary_loss_clip": 0.01165223,
      "auxiliary_loss_mlp": 0.01022388,
      "balance_loss_clip": 1.0423075,
      "balance_loss_mlp": 1.01512504,
      "epoch": 0.8290747309565322,
      "flos": 22273378387200.0,
      "grad_norm": 1.8432800668806304,
      "language_loss": 0.75096709,
      "learning_rate": 2.985774727262715e-07,
      "loss": 0.77284324,
      "num_input_tokens_seen": 148944585,
      "step": 6895,
      "time_per_iteration": 3.6540517807006836
    },
    {
      "auxiliary_loss_clip": 0.01164686,
      "auxiliary_loss_mlp": 0.01022801,
      "balance_loss_clip": 1.04640806,
      "balance_loss_mlp": 1.01577377,
      "epoch": 0.8291949738471713,
      "flos": 23255570856960.0,
      "grad_norm": 1.7680484748815672,
      "language_loss": 0.82025766,
      "learning_rate": 2.981681483661949e-07,
      "loss": 0.84213257,
      "num_input_tokens_seen": 148964170,
      "step": 6896,
      "time_per_iteration": 2.6245458126068115
    },
    {
      "auxiliary_loss_clip": 0.01171152,
      "auxiliary_loss_mlp": 0.01026903,
      "balance_loss_clip": 1.04878342,
      "balance_loss_mlp": 1.01979554,
      "epoch": 0.8293152167378104,
      "flos": 52555768185600.0,
      "grad_norm": 2.0830633720712632,
      "language_loss": 0.70924956,
      "learning_rate": 2.9775908217265633e-07,
      "loss": 0.73123014,
      "num_input_tokens_seen": 148989405,
      "step": 6897,
      "time_per_iteration": 2.992060422897339
    },
    {
      "auxiliary_loss_clip": 0.01098347,
      "auxiliary_loss_mlp": 0.01001921,
      "balance_loss_clip": 1.01420128,
      "balance_loss_mlp": 1.00110435,
      "epoch": 0.8294354596284494,
      "flos": 63356156294400.0,
      "grad_norm": 0.8286676790372957,
      "language_loss": 0.50340688,
      "learning_rate": 2.9735027420771253e-07,
      "loss": 0.52440953,
      "num_input_tokens_seen": 149049740,
      "step": 6898,
      "time_per_iteration": 3.165802001953125
    },
    {
      "auxiliary_loss_clip": 0.01164624,
      "auxiliary_loss_mlp": 0.01027096,
      "balance_loss_clip": 1.04614949,
      "balance_loss_mlp": 1.02035213,
      "epoch": 0.8295557025190886,
      "flos": 24827021942400.0,
      "grad_norm": 6.022991277522802,
      "language_loss": 0.71387064,
      "learning_rate": 2.969417245333774e-07,
      "loss": 0.73578781,
      "num_input_tokens_seen": 149069120,
      "step": 6899,
      "time_per_iteration": 2.651221752166748
    },
    {
      "auxiliary_loss_clip": 0.01171211,
      "auxiliary_loss_mlp": 0.01025385,
      "balance_loss_clip": 1.04420829,
      "balance_loss_mlp": 1.01795816,
      "epoch": 0.8296759454097277,
      "flos": 25118580637440.0,
      "grad_norm": 56.302594727506445,
      "language_loss": 0.77896667,
      "learning_rate": 2.9653343321162915e-07,
      "loss": 0.80093265,
      "num_input_tokens_seen": 149088630,
      "step": 6900,
      "time_per_iteration": 3.708144187927246
    },
    {
      "auxiliary_loss_clip": 0.01174865,
      "auxiliary_loss_mlp": 0.01021583,
      "balance_loss_clip": 1.04516315,
      "balance_loss_mlp": 1.01341128,
      "epoch": 0.8297961883003667,
      "flos": 24132581326080.0,
      "grad_norm": 4.2550591303738985,
      "language_loss": 0.65129346,
      "learning_rate": 2.9612540030440446e-07,
      "loss": 0.67325795,
      "num_input_tokens_seen": 149109175,
      "step": 6901,
      "time_per_iteration": 2.778226137161255
    },
    {
      "auxiliary_loss_clip": 0.01077156,
      "auxiliary_loss_mlp": 0.01001097,
      "balance_loss_clip": 1.01108432,
      "balance_loss_mlp": 1.00030982,
      "epoch": 0.8299164311910058,
      "flos": 67446561375360.0,
      "grad_norm": 0.8433996637891659,
      "language_loss": 0.64007199,
      "learning_rate": 2.9571762587360206e-07,
      "loss": 0.66085452,
      "num_input_tokens_seen": 149165560,
      "step": 6902,
      "time_per_iteration": 3.1605064868927
    },
    {
      "auxiliary_loss_clip": 0.01173557,
      "auxiliary_loss_mlp": 0.01024807,
      "balance_loss_clip": 1.03663623,
      "balance_loss_mlp": 1.017941,
      "epoch": 0.8300366740816449,
      "flos": 25228682801280.0,
      "grad_norm": 1.7755743419693544,
      "language_loss": 0.74004996,
      "learning_rate": 2.953101099810806e-07,
      "loss": 0.76203358,
      "num_input_tokens_seen": 149185165,
      "step": 6903,
      "time_per_iteration": 2.740947723388672
    },
    {
      "auxiliary_loss_clip": 0.01163919,
      "auxiliary_loss_mlp": 0.01027012,
      "balance_loss_clip": 1.04641795,
      "balance_loss_mlp": 1.01955593,
      "epoch": 0.830156916972284,
      "flos": 18041018757120.0,
      "grad_norm": 2.101700458786845,
      "language_loss": 0.82670712,
      "learning_rate": 2.9490285268865965e-07,
      "loss": 0.84861648,
      "num_input_tokens_seen": 149202655,
      "step": 6904,
      "time_per_iteration": 2.630763530731201
    },
    {
      "auxiliary_loss_clip": 0.01176153,
      "auxiliary_loss_mlp": 0.01019665,
      "balance_loss_clip": 1.04969549,
      "balance_loss_mlp": 1.01215196,
      "epoch": 0.830277159862923,
      "flos": 26322485806080.0,
      "grad_norm": 2.504435659093384,
      "language_loss": 0.7995801,
      "learning_rate": 2.9449585405812085e-07,
      "loss": 0.82153833,
      "num_input_tokens_seen": 149220035,
      "step": 6905,
      "time_per_iteration": 2.571657419204712
    },
    {
      "auxiliary_loss_clip": 0.01180097,
      "auxiliary_loss_mlp": 0.01022221,
      "balance_loss_clip": 1.04447854,
      "balance_loss_mlp": 1.01496959,
      "epoch": 0.8303974027535622,
      "flos": 19938861751680.0,
      "grad_norm": 2.763853804323069,
      "language_loss": 0.74290556,
      "learning_rate": 2.940891141512043e-07,
      "loss": 0.7649287,
      "num_input_tokens_seen": 149238055,
      "step": 6906,
      "time_per_iteration": 2.6902341842651367
    },
    {
      "auxiliary_loss_clip": 0.01170553,
      "auxiliary_loss_mlp": 0.01025464,
      "balance_loss_clip": 1.0431391,
      "balance_loss_mlp": 1.01757526,
      "epoch": 0.8305176456442013,
      "flos": 17165552572800.0,
      "grad_norm": 2.349772245303175,
      "language_loss": 0.719648,
      "learning_rate": 2.9368263302961385e-07,
      "loss": 0.74160826,
      "num_input_tokens_seen": 149256755,
      "step": 6907,
      "time_per_iteration": 2.622711181640625
    },
    {
      "auxiliary_loss_clip": 0.01180297,
      "auxiliary_loss_mlp": 0.01021245,
      "balance_loss_clip": 1.03762722,
      "balance_loss_mlp": 1.01371145,
      "epoch": 0.8306378885348403,
      "flos": 25627614226560.0,
      "grad_norm": 1.7787002840390425,
      "language_loss": 0.79818618,
      "learning_rate": 2.9327641075501075e-07,
      "loss": 0.82020164,
      "num_input_tokens_seen": 149275745,
      "step": 6908,
      "time_per_iteration": 2.7431564331054688
    },
    {
      "auxiliary_loss_clip": 0.01164752,
      "auxiliary_loss_mlp": 0.01030739,
      "balance_loss_clip": 1.04028547,
      "balance_loss_mlp": 1.02251399,
      "epoch": 0.8307581314254795,
      "flos": 33947864985600.0,
      "grad_norm": 2.504918047171727,
      "language_loss": 0.66800559,
      "learning_rate": 2.9287044738901866e-07,
      "loss": 0.68996048,
      "num_input_tokens_seen": 149293730,
      "step": 6909,
      "time_per_iteration": 2.800718307495117
    },
    {
      "auxiliary_loss_clip": 0.01170635,
      "auxiliary_loss_mlp": 0.01207582,
      "balance_loss_clip": 1.04557121,
      "balance_loss_mlp": 1.00073028,
      "epoch": 0.8308783743161186,
      "flos": 17562724231680.0,
      "grad_norm": 2.145027096504121,
      "language_loss": 0.90760839,
      "learning_rate": 2.9246474299322274e-07,
      "loss": 0.93139058,
      "num_input_tokens_seen": 149309290,
      "step": 6910,
      "time_per_iteration": 2.6056859493255615
    },
    {
      "auxiliary_loss_clip": 0.01083382,
      "auxiliary_loss_mlp": 0.01001816,
      "balance_loss_clip": 1.01263714,
      "balance_loss_mlp": 1.00103533,
      "epoch": 0.8309986172067576,
      "flos": 69412885649280.0,
      "grad_norm": 0.893030839219376,
      "language_loss": 0.63119429,
      "learning_rate": 2.920592976291678e-07,
      "loss": 0.65204626,
      "num_input_tokens_seen": 149366620,
      "step": 6911,
      "time_per_iteration": 3.142883062362671
    },
    {
      "auxiliary_loss_clip": 0.01168574,
      "auxiliary_loss_mlp": 0.0102659,
      "balance_loss_clip": 1.0450747,
      "balance_loss_mlp": 1.01885653,
      "epoch": 0.8311188600973968,
      "flos": 22309755886080.0,
      "grad_norm": 1.8690171000365985,
      "language_loss": 0.80894363,
      "learning_rate": 2.916541113583595e-07,
      "loss": 0.83089525,
      "num_input_tokens_seen": 149385120,
      "step": 6912,
      "time_per_iteration": 2.6672301292419434
    },
    {
      "auxiliary_loss_clip": 0.01181232,
      "auxiliary_loss_mlp": 0.01024624,
      "balance_loss_clip": 1.04644275,
      "balance_loss_mlp": 1.01680982,
      "epoch": 0.8312391029880358,
      "flos": 18770077105920.0,
      "grad_norm": 2.4846056482010064,
      "language_loss": 0.66703236,
      "learning_rate": 2.912491842422642e-07,
      "loss": 0.68909091,
      "num_input_tokens_seen": 149402825,
      "step": 6913,
      "time_per_iteration": 3.6193387508392334
    },
    {
      "auxiliary_loss_clip": 0.01172146,
      "auxiliary_loss_mlp": 0.01020836,
      "balance_loss_clip": 1.04763389,
      "balance_loss_mlp": 1.01327252,
      "epoch": 0.8313593458786749,
      "flos": 20376648714240.0,
      "grad_norm": 1.7159313230288042,
      "language_loss": 0.71139014,
      "learning_rate": 2.9084451634230857e-07,
      "loss": 0.73332,
      "num_input_tokens_seen": 149422125,
      "step": 6914,
      "time_per_iteration": 2.6640610694885254
    },
    {
      "auxiliary_loss_clip": 0.01172507,
      "auxiliary_loss_mlp": 0.01022952,
      "balance_loss_clip": 1.04058158,
      "balance_loss_mlp": 1.01526308,
      "epoch": 0.831479588769314,
      "flos": 32124069878400.0,
      "grad_norm": 1.9860849001910537,
      "language_loss": 0.71109635,
      "learning_rate": 2.9044010771988125e-07,
      "loss": 0.73305094,
      "num_input_tokens_seen": 149441940,
      "step": 6915,
      "time_per_iteration": 2.7340078353881836
    },
    {
      "auxiliary_loss_clip": 0.01165478,
      "auxiliary_loss_mlp": 0.01024577,
      "balance_loss_clip": 1.04398203,
      "balance_loss_mlp": 1.01728773,
      "epoch": 0.8315998316599531,
      "flos": 45185929338240.0,
      "grad_norm": 2.659230405235303,
      "language_loss": 0.7217105,
      "learning_rate": 2.900359584363303e-07,
      "loss": 0.7436111,
      "num_input_tokens_seen": 149465045,
      "step": 6916,
      "time_per_iteration": 2.890779972076416
    },
    {
      "auxiliary_loss_clip": 0.01171664,
      "auxiliary_loss_mlp": 0.0103172,
      "balance_loss_clip": 1.04227114,
      "balance_loss_mlp": 1.02336335,
      "epoch": 0.8317200745505922,
      "flos": 18363747479040.0,
      "grad_norm": 2.1648839987209314,
      "language_loss": 0.84721845,
      "learning_rate": 2.8963206855296494e-07,
      "loss": 0.86925232,
      "num_input_tokens_seen": 149481285,
      "step": 6917,
      "time_per_iteration": 2.683915615081787
    },
    {
      "auxiliary_loss_clip": 0.0117098,
      "auxiliary_loss_mlp": 0.01027839,
      "balance_loss_clip": 1.04588842,
      "balance_loss_mlp": 1.02032864,
      "epoch": 0.8318403174412313,
      "flos": 24206557386240.0,
      "grad_norm": 1.7783263241231964,
      "language_loss": 0.77423346,
      "learning_rate": 2.892284381310548e-07,
      "loss": 0.79622161,
      "num_input_tokens_seen": 149502700,
      "step": 6918,
      "time_per_iteration": 2.712219715118408
    },
    {
      "auxiliary_loss_clip": 0.01166273,
      "auxiliary_loss_mlp": 0.0102423,
      "balance_loss_clip": 1.04234588,
      "balance_loss_mlp": 1.01619506,
      "epoch": 0.8319605603318704,
      "flos": 22418780641920.0,
      "grad_norm": 3.4764648059052754,
      "language_loss": 0.72782266,
      "learning_rate": 2.888250672318302e-07,
      "loss": 0.74972773,
      "num_input_tokens_seen": 149520100,
      "step": 6919,
      "time_per_iteration": 2.6275222301483154
    },
    {
      "auxiliary_loss_clip": 0.01172489,
      "auxiliary_loss_mlp": 0.0102945,
      "balance_loss_clip": 1.05138946,
      "balance_loss_mlp": 1.02197528,
      "epoch": 0.8320808032225094,
      "flos": 37414501459200.0,
      "grad_norm": 1.689361699842272,
      "language_loss": 0.6849038,
      "learning_rate": 2.884219559164831e-07,
      "loss": 0.70692325,
      "num_input_tokens_seen": 149543245,
      "step": 6920,
      "time_per_iteration": 3.629643201828003
    },
    {
      "auxiliary_loss_clip": 0.01168983,
      "auxiliary_loss_mlp": 0.01024512,
      "balance_loss_clip": 1.04612517,
      "balance_loss_mlp": 1.01723409,
      "epoch": 0.8322010461131486,
      "flos": 12787395638400.0,
      "grad_norm": 2.440138782347127,
      "language_loss": 0.81856006,
      "learning_rate": 2.880191042461635e-07,
      "loss": 0.84049499,
      "num_input_tokens_seen": 149559185,
      "step": 6921,
      "time_per_iteration": 2.6528844833374023
    },
    {
      "auxiliary_loss_clip": 0.01181865,
      "auxiliary_loss_mlp": 0.01026229,
      "balance_loss_clip": 1.04141855,
      "balance_loss_mlp": 1.01944947,
      "epoch": 0.8323212890037877,
      "flos": 15815455050240.0,
      "grad_norm": 1.8509071045172352,
      "language_loss": 0.80296779,
      "learning_rate": 2.876165122819849e-07,
      "loss": 0.82504874,
      "num_input_tokens_seen": 149577165,
      "step": 6922,
      "time_per_iteration": 3.607708215713501
    },
    {
      "auxiliary_loss_clip": 0.0116434,
      "auxiliary_loss_mlp": 0.01020589,
      "balance_loss_clip": 1.0457449,
      "balance_loss_mlp": 1.01335847,
      "epoch": 0.8324415318944267,
      "flos": 21719276208000.0,
      "grad_norm": 1.8538715617875818,
      "language_loss": 0.79363751,
      "learning_rate": 2.872141800850201e-07,
      "loss": 0.81548679,
      "num_input_tokens_seen": 149594340,
      "step": 6923,
      "time_per_iteration": 2.757263422012329
    },
    {
      "auxiliary_loss_clip": 0.01165623,
      "auxiliary_loss_mlp": 0.01023987,
      "balance_loss_clip": 1.046826,
      "balance_loss_mlp": 1.01726079,
      "epoch": 0.8325617747850659,
      "flos": 34198700636160.0,
      "grad_norm": 9.450421874550102,
      "language_loss": 0.73422587,
      "learning_rate": 2.868121077163024e-07,
      "loss": 0.75612193,
      "num_input_tokens_seen": 149613895,
      "step": 6924,
      "time_per_iteration": 2.694591522216797
    },
    {
      "auxiliary_loss_clip": 0.01174094,
      "auxiliary_loss_mlp": 0.01028727,
      "balance_loss_clip": 1.04627872,
      "balance_loss_mlp": 1.02147925,
      "epoch": 0.8326820176757049,
      "flos": 18369457741440.0,
      "grad_norm": 1.875422035514894,
      "language_loss": 0.72608018,
      "learning_rate": 2.864102952368257e-07,
      "loss": 0.74810839,
      "num_input_tokens_seen": 149631820,
      "step": 6925,
      "time_per_iteration": 2.6685049533843994
    },
    {
      "auxiliary_loss_clip": 0.01164931,
      "auxiliary_loss_mlp": 0.01023524,
      "balance_loss_clip": 1.03624344,
      "balance_loss_mlp": 1.01540554,
      "epoch": 0.832802260566344,
      "flos": 35991325716480.0,
      "grad_norm": 1.4218957139975772,
      "language_loss": 0.59471834,
      "learning_rate": 2.860087427075444e-07,
      "loss": 0.61660284,
      "num_input_tokens_seen": 149656070,
      "step": 6926,
      "time_per_iteration": 3.8084518909454346
    },
    {
      "auxiliary_loss_clip": 0.0116608,
      "auxiliary_loss_mlp": 0.01027695,
      "balance_loss_clip": 1.04374671,
      "balance_loss_mlp": 1.02023292,
      "epoch": 0.8329225034569832,
      "flos": 14244434928000.0,
      "grad_norm": 2.343156566219566,
      "language_loss": 0.85894024,
      "learning_rate": 2.856074501893744e-07,
      "loss": 0.88087797,
      "num_input_tokens_seen": 149671270,
      "step": 6927,
      "time_per_iteration": 2.5433971881866455
    },
    {
      "auxiliary_loss_clip": 0.01174041,
      "auxiliary_loss_mlp": 0.01024314,
      "balance_loss_clip": 1.04932535,
      "balance_loss_mlp": 1.01725078,
      "epoch": 0.8330427463476222,
      "flos": 18077468083200.0,
      "grad_norm": 1.7592502877968794,
      "language_loss": 0.81727004,
      "learning_rate": 2.8520641774319054e-07,
      "loss": 0.83925354,
      "num_input_tokens_seen": 149689360,
      "step": 6928,
      "time_per_iteration": 2.620643138885498
    },
    {
      "auxiliary_loss_clip": 0.01173636,
      "auxiliary_loss_mlp": 0.01025458,
      "balance_loss_clip": 1.04082298,
      "balance_loss_mlp": 1.01748276,
      "epoch": 0.8331629892382613,
      "flos": 18040839189120.0,
      "grad_norm": 4.564870988588691,
      "language_loss": 0.75515825,
      "learning_rate": 2.848056454298309e-07,
      "loss": 0.7771492,
      "num_input_tokens_seen": 149706685,
      "step": 6929,
      "time_per_iteration": 2.6874825954437256
    },
    {
      "auxiliary_loss_clip": 0.01170129,
      "auxiliary_loss_mlp": 0.01020679,
      "balance_loss_clip": 1.04538751,
      "balance_loss_mlp": 1.01314795,
      "epoch": 0.8332832321289004,
      "flos": 17457398576640.0,
      "grad_norm": 3.9191278406597623,
      "language_loss": 0.65085065,
      "learning_rate": 2.844051333100905e-07,
      "loss": 0.67275876,
      "num_input_tokens_seen": 149724230,
      "step": 6930,
      "time_per_iteration": 2.6394729614257812
    },
    {
      "auxiliary_loss_clip": 0.01173217,
      "auxiliary_loss_mlp": 0.0102073,
      "balance_loss_clip": 1.04700589,
      "balance_loss_mlp": 1.01431966,
      "epoch": 0.8334034750195395,
      "flos": 15084852416640.0,
      "grad_norm": 2.0356849861236506,
      "language_loss": 0.8392998,
      "learning_rate": 2.840048814447269e-07,
      "loss": 0.86123919,
      "num_input_tokens_seen": 149742395,
      "step": 6931,
      "time_per_iteration": 2.555072784423828
    },
    {
      "auxiliary_loss_clip": 0.011679,
      "auxiliary_loss_mlp": 0.01022318,
      "balance_loss_clip": 1.04443216,
      "balance_loss_mlp": 1.01493001,
      "epoch": 0.8335237179101785,
      "flos": 19427170556160.0,
      "grad_norm": 3.776462734301338,
      "language_loss": 0.74670041,
      "learning_rate": 2.836048898944587e-07,
      "loss": 0.76860261,
      "num_input_tokens_seen": 149760820,
      "step": 6932,
      "time_per_iteration": 2.6403021812438965
    },
    {
      "auxiliary_loss_clip": 0.01170724,
      "auxiliary_loss_mlp": 0.01022334,
      "balance_loss_clip": 1.04343462,
      "balance_loss_mlp": 1.01554847,
      "epoch": 0.8336439608008177,
      "flos": 21762046327680.0,
      "grad_norm": 2.880908846256439,
      "language_loss": 0.73118818,
      "learning_rate": 2.832051587199642e-07,
      "loss": 0.75311875,
      "num_input_tokens_seen": 149778075,
      "step": 6933,
      "time_per_iteration": 2.6512458324432373
    },
    {
      "auxiliary_loss_clip": 0.01072058,
      "auxiliary_loss_mlp": 0.01001291,
      "balance_loss_clip": 1.01189053,
      "balance_loss_mlp": 1.0005163,
      "epoch": 0.8337642036914568,
      "flos": 59702783990400.0,
      "grad_norm": 0.8026383057636705,
      "language_loss": 0.57656574,
      "learning_rate": 2.828056879818821e-07,
      "loss": 0.59729922,
      "num_input_tokens_seen": 149837150,
      "step": 6934,
      "time_per_iteration": 3.1243529319763184
    },
    {
      "auxiliary_loss_clip": 0.01173206,
      "auxiliary_loss_mlp": 0.01023884,
      "balance_loss_clip": 1.04003978,
      "balance_loss_mlp": 1.01745856,
      "epoch": 0.8338844465820958,
      "flos": 27162185022720.0,
      "grad_norm": 1.9099964202035604,
      "language_loss": 0.83000708,
      "learning_rate": 2.824064777408117e-07,
      "loss": 0.85197794,
      "num_input_tokens_seen": 149856940,
      "step": 6935,
      "time_per_iteration": 2.8266024589538574
    },
    {
      "auxiliary_loss_clip": 0.01167159,
      "auxiliary_loss_mlp": 0.01022905,
      "balance_loss_clip": 1.04642057,
      "balance_loss_mlp": 1.01546669,
      "epoch": 0.8340046894727349,
      "flos": 30481264425600.0,
      "grad_norm": 1.6469207131404229,
      "language_loss": 0.76035178,
      "learning_rate": 2.8200752805731263e-07,
      "loss": 0.78225243,
      "num_input_tokens_seen": 149879930,
      "step": 6936,
      "time_per_iteration": 2.7512834072113037
    },
    {
      "auxiliary_loss_clip": 0.01170258,
      "auxiliary_loss_mlp": 0.01023741,
      "balance_loss_clip": 1.04742336,
      "balance_loss_mlp": 1.01638556,
      "epoch": 0.834124932363374,
      "flos": 27126166659840.0,
      "grad_norm": 1.595293409971093,
      "language_loss": 0.80992091,
      "learning_rate": 2.8160883899190625e-07,
      "loss": 0.8318609,
      "num_input_tokens_seen": 149903200,
      "step": 6937,
      "time_per_iteration": 2.7530529499053955
    },
    {
      "auxiliary_loss_clip": 0.01165657,
      "auxiliary_loss_mlp": 0.01026373,
      "balance_loss_clip": 1.04381776,
      "balance_loss_mlp": 1.01898193,
      "epoch": 0.8342451752540131,
      "flos": 24569865498240.0,
      "grad_norm": 2.101691129120781,
      "language_loss": 0.73105693,
      "learning_rate": 2.8121041060507234e-07,
      "loss": 0.75297725,
      "num_input_tokens_seen": 149922230,
      "step": 6938,
      "time_per_iteration": 2.74841570854187
    },
    {
      "auxiliary_loss_clip": 0.01174737,
      "auxiliary_loss_mlp": 0.01022263,
      "balance_loss_clip": 1.04635978,
      "balance_loss_mlp": 1.01493454,
      "epoch": 0.8343654181446521,
      "flos": 26615085995520.0,
      "grad_norm": 1.6419862525267155,
      "language_loss": 0.71511555,
      "learning_rate": 2.808122429572528e-07,
      "loss": 0.73708558,
      "num_input_tokens_seen": 149942435,
      "step": 6939,
      "time_per_iteration": 2.690253734588623
    },
    {
      "auxiliary_loss_clip": 0.01181884,
      "auxiliary_loss_mlp": 0.01024121,
      "balance_loss_clip": 1.04298377,
      "balance_loss_mlp": 1.01652741,
      "epoch": 0.8344856610352913,
      "flos": 20777268078720.0,
      "grad_norm": 3.4797814668840505,
      "language_loss": 0.75894046,
      "learning_rate": 2.804143361088489e-07,
      "loss": 0.78100049,
      "num_input_tokens_seen": 149961615,
      "step": 6940,
      "time_per_iteration": 3.5603346824645996
    },
    {
      "auxiliary_loss_clip": 0.0116643,
      "auxiliary_loss_mlp": 0.01024707,
      "balance_loss_clip": 1.04376221,
      "balance_loss_mlp": 1.01684546,
      "epoch": 0.8346059039259304,
      "flos": 26095960684800.0,
      "grad_norm": 2.845963404522298,
      "language_loss": 0.77964497,
      "learning_rate": 2.8001669012022277e-07,
      "loss": 0.80155635,
      "num_input_tokens_seen": 149979585,
      "step": 6941,
      "time_per_iteration": 2.6519105434417725
    },
    {
      "auxiliary_loss_clip": 0.01170394,
      "auxiliary_loss_mlp": 0.0102684,
      "balance_loss_clip": 1.04944837,
      "balance_loss_mlp": 1.01904964,
      "epoch": 0.8347261468165694,
      "flos": 29027708755200.0,
      "grad_norm": 2.0324346109006024,
      "language_loss": 0.69361424,
      "learning_rate": 2.7961930505169795e-07,
      "loss": 0.7155866,
      "num_input_tokens_seen": 150003830,
      "step": 6942,
      "time_per_iteration": 2.674227237701416
    },
    {
      "auxiliary_loss_clip": 0.01173336,
      "auxiliary_loss_mlp": 0.01207829,
      "balance_loss_clip": 1.04875195,
      "balance_loss_mlp": 1.00084603,
      "epoch": 0.8348463897072086,
      "flos": 26396461866240.0,
      "grad_norm": 1.965799469173582,
      "language_loss": 0.76709324,
      "learning_rate": 2.792221809635558e-07,
      "loss": 0.79090488,
      "num_input_tokens_seen": 150024460,
      "step": 6943,
      "time_per_iteration": 2.680677652359009
    },
    {
      "auxiliary_loss_clip": 0.01184995,
      "auxiliary_loss_mlp": 0.01024343,
      "balance_loss_clip": 1.03997636,
      "balance_loss_mlp": 1.01704097,
      "epoch": 0.8349666325978476,
      "flos": 23367720096000.0,
      "grad_norm": 2.6635428824825227,
      "language_loss": 0.75025308,
      "learning_rate": 2.788253179160411e-07,
      "loss": 0.77234638,
      "num_input_tokens_seen": 150045620,
      "step": 6944,
      "time_per_iteration": 2.788248300552368
    },
    {
      "auxiliary_loss_clip": 0.01171802,
      "auxiliary_loss_mlp": 0.010281,
      "balance_loss_clip": 1.04479218,
      "balance_loss_mlp": 1.02101922,
      "epoch": 0.8350868754884867,
      "flos": 12896528135040.0,
      "grad_norm": 2.074319109989375,
      "language_loss": 0.6548717,
      "learning_rate": 2.7842871596935725e-07,
      "loss": 0.67687076,
      "num_input_tokens_seen": 150064135,
      "step": 6945,
      "time_per_iteration": 2.6883764266967773
    },
    {
      "auxiliary_loss_clip": 0.01174473,
      "auxiliary_loss_mlp": 0.01019316,
      "balance_loss_clip": 1.04598522,
      "balance_loss_mlp": 1.01185668,
      "epoch": 0.8352071183791259,
      "flos": 26505522535680.0,
      "grad_norm": 1.8332796885135687,
      "language_loss": 0.69326985,
      "learning_rate": 2.780323751836682e-07,
      "loss": 0.71520776,
      "num_input_tokens_seen": 150085350,
      "step": 6946,
      "time_per_iteration": 3.582214832305908
    },
    {
      "auxiliary_loss_clip": 0.01170453,
      "auxiliary_loss_mlp": 0.01207217,
      "balance_loss_clip": 1.04144931,
      "balance_loss_mlp": 1.00080085,
      "epoch": 0.8353273612697649,
      "flos": 20668063754880.0,
      "grad_norm": 1.4683907821438864,
      "language_loss": 0.78651857,
      "learning_rate": 2.7763629561909876e-07,
      "loss": 0.81029534,
      "num_input_tokens_seen": 150106180,
      "step": 6947,
      "time_per_iteration": 2.6853604316711426
    },
    {
      "auxiliary_loss_clip": 0.01162921,
      "auxiliary_loss_mlp": 0.01021396,
      "balance_loss_clip": 1.04513955,
      "balance_loss_mlp": 1.01360893,
      "epoch": 0.835447604160404,
      "flos": 19754137082880.0,
      "grad_norm": 2.0427664212728787,
      "language_loss": 0.77086473,
      "learning_rate": 2.772404773357335e-07,
      "loss": 0.79270792,
      "num_input_tokens_seen": 150125585,
      "step": 6948,
      "time_per_iteration": 3.473496913909912
    },
    {
      "auxiliary_loss_clip": 0.0116897,
      "auxiliary_loss_mlp": 0.01027316,
      "balance_loss_clip": 1.04314125,
      "balance_loss_mlp": 1.01944208,
      "epoch": 0.8355678470510431,
      "flos": 23435842239360.0,
      "grad_norm": 2.378750723939499,
      "language_loss": 0.78573442,
      "learning_rate": 2.7684492039361853e-07,
      "loss": 0.8076973,
      "num_input_tokens_seen": 150144810,
      "step": 6949,
      "time_per_iteration": 2.7112269401550293
    },
    {
      "auxiliary_loss_clip": 0.0116984,
      "auxiliary_loss_mlp": 0.01024138,
      "balance_loss_clip": 1.04968905,
      "balance_loss_mlp": 1.01684797,
      "epoch": 0.8356880899416822,
      "flos": 21214588164480.0,
      "grad_norm": 1.7019532368765444,
      "language_loss": 0.83765382,
      "learning_rate": 2.764496248527586e-07,
      "loss": 0.85959351,
      "num_input_tokens_seen": 150163785,
      "step": 6950,
      "time_per_iteration": 2.5957577228546143
    },
    {
      "auxiliary_loss_clip": 0.01185315,
      "auxiliary_loss_mlp": 0.01022723,
      "balance_loss_clip": 1.04435515,
      "balance_loss_mlp": 1.01540899,
      "epoch": 0.8358083328323213,
      "flos": 28037543466240.0,
      "grad_norm": 2.1185080465826744,
      "language_loss": 0.7885102,
      "learning_rate": 2.760545907731211e-07,
      "loss": 0.81059051,
      "num_input_tokens_seen": 150184360,
      "step": 6951,
      "time_per_iteration": 2.856400489807129
    },
    {
      "auxiliary_loss_clip": 0.01171142,
      "auxiliary_loss_mlp": 0.01022134,
      "balance_loss_clip": 1.04476213,
      "balance_loss_mlp": 1.01440966,
      "epoch": 0.8359285757229604,
      "flos": 27783655159680.0,
      "grad_norm": 1.8246538844314162,
      "language_loss": 0.67566979,
      "learning_rate": 2.75659818214631e-07,
      "loss": 0.69760257,
      "num_input_tokens_seen": 150205465,
      "step": 6952,
      "time_per_iteration": 2.6590113639831543
    },
    {
      "auxiliary_loss_clip": 0.01173295,
      "auxiliary_loss_mlp": 0.01020608,
      "balance_loss_clip": 1.04378414,
      "balance_loss_mlp": 1.01339245,
      "epoch": 0.8360488186135995,
      "flos": 21435115714560.0,
      "grad_norm": 3.81050833393234,
      "language_loss": 0.78014886,
      "learning_rate": 2.752653072371749e-07,
      "loss": 0.80208784,
      "num_input_tokens_seen": 150224900,
      "step": 6953,
      "time_per_iteration": 3.6957640647888184
    },
    {
      "auxiliary_loss_clip": 0.01171001,
      "auxiliary_loss_mlp": 0.01026739,
      "balance_loss_clip": 1.04435647,
      "balance_loss_mlp": 1.01996827,
      "epoch": 0.8361690615042385,
      "flos": 27632327160960.0,
      "grad_norm": 1.8531923981044336,
      "language_loss": 0.74881351,
      "learning_rate": 2.7487105790060105e-07,
      "loss": 0.77079093,
      "num_input_tokens_seen": 150244310,
      "step": 6954,
      "time_per_iteration": 2.6531083583831787
    },
    {
      "auxiliary_loss_clip": 0.01171247,
      "auxiliary_loss_mlp": 0.01024296,
      "balance_loss_clip": 1.0457865,
      "balance_loss_mlp": 1.01721227,
      "epoch": 0.8362893043948777,
      "flos": 39202529598720.0,
      "grad_norm": 1.9361077948435437,
      "language_loss": 0.69013649,
      "learning_rate": 2.7447707026471587e-07,
      "loss": 0.71209192,
      "num_input_tokens_seen": 150267285,
      "step": 6955,
      "time_per_iteration": 2.824571371078491
    },
    {
      "auxiliary_loss_clip": 0.01175946,
      "auxiliary_loss_mlp": 0.01021081,
      "balance_loss_clip": 1.04262054,
      "balance_loss_mlp": 1.01405048,
      "epoch": 0.8364095472855168,
      "flos": 24785329230720.0,
      "grad_norm": 2.1849357041030038,
      "language_loss": 0.80107594,
      "learning_rate": 2.740833443892874e-07,
      "loss": 0.82304615,
      "num_input_tokens_seen": 150285455,
      "step": 6956,
      "time_per_iteration": 2.7653346061706543
    },
    {
      "auxiliary_loss_clip": 0.01171565,
      "auxiliary_loss_mlp": 0.01020294,
      "balance_loss_clip": 1.04421282,
      "balance_loss_mlp": 1.01308155,
      "epoch": 0.8365297901761558,
      "flos": 22743412784640.0,
      "grad_norm": 1.817146273271877,
      "language_loss": 0.79987776,
      "learning_rate": 2.7368988033404327e-07,
      "loss": 0.8217963,
      "num_input_tokens_seen": 150302970,
      "step": 6957,
      "time_per_iteration": 2.6481802463531494
    },
    {
      "auxiliary_loss_clip": 0.01178046,
      "auxiliary_loss_mlp": 0.01023788,
      "balance_loss_clip": 1.04384995,
      "balance_loss_mlp": 1.01716638,
      "epoch": 0.836650033066795,
      "flos": 28396003242240.0,
      "grad_norm": 1.4937638877570472,
      "language_loss": 0.84646434,
      "learning_rate": 2.732966781586712e-07,
      "loss": 0.86848265,
      "num_input_tokens_seen": 150322715,
      "step": 6958,
      "time_per_iteration": 2.753801107406616
    },
    {
      "auxiliary_loss_clip": 0.01163355,
      "auxiliary_loss_mlp": 0.01019742,
      "balance_loss_clip": 1.04317331,
      "balance_loss_mlp": 1.01286924,
      "epoch": 0.836770275957434,
      "flos": 22236857233920.0,
      "grad_norm": 1.93541423425391,
      "language_loss": 0.66752815,
      "learning_rate": 2.729037379228205e-07,
      "loss": 0.68935907,
      "num_input_tokens_seen": 150342900,
      "step": 6959,
      "time_per_iteration": 2.611593008041382
    },
    {
      "auxiliary_loss_clip": 0.01171453,
      "auxiliary_loss_mlp": 0.01023317,
      "balance_loss_clip": 1.04643178,
      "balance_loss_mlp": 1.01587224,
      "epoch": 0.8368905188480731,
      "flos": 22491930689280.0,
      "grad_norm": 1.8097949826896353,
      "language_loss": 0.80411971,
      "learning_rate": 2.725110596860998e-07,
      "loss": 0.82606733,
      "num_input_tokens_seen": 150363580,
      "step": 6960,
      "time_per_iteration": 2.6505162715911865
    },
    {
      "auxiliary_loss_clip": 0.01173954,
      "auxiliary_loss_mlp": 0.01023418,
      "balance_loss_clip": 1.04289699,
      "balance_loss_mlp": 1.01627159,
      "epoch": 0.8370107617387123,
      "flos": 13370405287680.0,
      "grad_norm": 1.8510527346293397,
      "language_loss": 0.70413315,
      "learning_rate": 2.7211864350807776e-07,
      "loss": 0.72610688,
      "num_input_tokens_seen": 150381780,
      "step": 6961,
      "time_per_iteration": 2.648607015609741
    },
    {
      "auxiliary_loss_clip": 0.01166459,
      "auxiliary_loss_mlp": 0.01027466,
      "balance_loss_clip": 1.04694545,
      "balance_loss_mlp": 1.01961052,
      "epoch": 0.8371310046293513,
      "flos": 25261289372160.0,
      "grad_norm": 1.6399030110260666,
      "language_loss": 0.73871958,
      "learning_rate": 2.717264894482836e-07,
      "loss": 0.7606588,
      "num_input_tokens_seen": 150402120,
      "step": 6962,
      "time_per_iteration": 2.6671841144561768
    },
    {
      "auxiliary_loss_clip": 0.01173276,
      "auxiliary_loss_mlp": 0.0102344,
      "balance_loss_clip": 1.04837751,
      "balance_loss_mlp": 1.01549494,
      "epoch": 0.8372512475199904,
      "flos": 19792705311360.0,
      "grad_norm": 2.7826603122523434,
      "language_loss": 0.80975688,
      "learning_rate": 2.7133459756620646e-07,
      "loss": 0.83172393,
      "num_input_tokens_seen": 150419315,
      "step": 6963,
      "time_per_iteration": 2.6578195095062256
    },
    {
      "auxiliary_loss_clip": 0.01164599,
      "auxiliary_loss_mlp": 0.01028897,
      "balance_loss_clip": 1.04500806,
      "balance_loss_mlp": 1.02142298,
      "epoch": 0.8373714904106295,
      "flos": 19391224020480.0,
      "grad_norm": 1.6988727945467448,
      "language_loss": 0.74172539,
      "learning_rate": 2.7094296792129733e-07,
      "loss": 0.76366037,
      "num_input_tokens_seen": 150438915,
      "step": 6964,
      "time_per_iteration": 2.7343251705169678
    },
    {
      "auxiliary_loss_clip": 0.01168427,
      "auxiliary_loss_mlp": 0.01021476,
      "balance_loss_clip": 1.04532456,
      "balance_loss_mlp": 1.01417494,
      "epoch": 0.8374917333012686,
      "flos": 14975935401600.0,
      "grad_norm": 1.9535951535922578,
      "language_loss": 0.75502723,
      "learning_rate": 2.7055160057296424e-07,
      "loss": 0.77692628,
      "num_input_tokens_seen": 150456155,
      "step": 6965,
      "time_per_iteration": 2.545140504837036
    },
    {
      "auxiliary_loss_clip": 0.01173205,
      "auxiliary_loss_mlp": 0.01024511,
      "balance_loss_clip": 1.0425117,
      "balance_loss_mlp": 1.01686668,
      "epoch": 0.8376119761919076,
      "flos": 30331839847680.0,
      "grad_norm": 1.7248646722616694,
      "language_loss": 0.72644997,
      "learning_rate": 2.7016049558057896e-07,
      "loss": 0.74842715,
      "num_input_tokens_seen": 150478115,
      "step": 6966,
      "time_per_iteration": 3.7127199172973633
    },
    {
      "auxiliary_loss_clip": 0.01168902,
      "auxiliary_loss_mlp": 0.01026096,
      "balance_loss_clip": 1.04657865,
      "balance_loss_mlp": 1.01816845,
      "epoch": 0.8377322190825467,
      "flos": 29423336129280.0,
      "grad_norm": 1.8455778843593735,
      "language_loss": 0.70924318,
      "learning_rate": 2.6976965300347074e-07,
      "loss": 0.73119318,
      "num_input_tokens_seen": 150500725,
      "step": 6967,
      "time_per_iteration": 2.7995288372039795
    },
    {
      "auxiliary_loss_clip": 0.01168372,
      "auxiliary_loss_mlp": 0.0102058,
      "balance_loss_clip": 1.04276752,
      "balance_loss_mlp": 1.01290929,
      "epoch": 0.8378524619731859,
      "flos": 26687086807680.0,
      "grad_norm": 2.4668428341213944,
      "language_loss": 0.69026452,
      "learning_rate": 2.693790729009309e-07,
      "loss": 0.71215403,
      "num_input_tokens_seen": 150522335,
      "step": 6968,
      "time_per_iteration": 2.704291582107544
    },
    {
      "auxiliary_loss_clip": 0.01172272,
      "auxiliary_loss_mlp": 0.01022008,
      "balance_loss_clip": 1.04462647,
      "balance_loss_mlp": 1.01484895,
      "epoch": 0.8379727048638249,
      "flos": 20703866636160.0,
      "grad_norm": 2.0913619132358017,
      "language_loss": 0.88891339,
      "learning_rate": 2.6898875533220946e-07,
      "loss": 0.91085619,
      "num_input_tokens_seen": 150541640,
      "step": 6969,
      "time_per_iteration": 2.6520164012908936
    },
    {
      "auxiliary_loss_clip": 0.01162644,
      "auxiliary_loss_mlp": 0.01020754,
      "balance_loss_clip": 1.04742122,
      "balance_loss_mlp": 1.01425433,
      "epoch": 0.838092947754464,
      "flos": 20084084438400.0,
      "grad_norm": 1.6981260028241993,
      "language_loss": 0.81950855,
      "learning_rate": 2.685987003565171e-07,
      "loss": 0.84134257,
      "num_input_tokens_seen": 150559680,
      "step": 6970,
      "time_per_iteration": 2.5881550312042236
    },
    {
      "auxiliary_loss_clip": 0.01169102,
      "auxiliary_loss_mlp": 0.01025063,
      "balance_loss_clip": 1.04641116,
      "balance_loss_mlp": 1.01723075,
      "epoch": 0.8382131906451031,
      "flos": 18113270964480.0,
      "grad_norm": 2.3593879192602216,
      "language_loss": 0.75162983,
      "learning_rate": 2.6820890803302566e-07,
      "loss": 0.77357143,
      "num_input_tokens_seen": 150575205,
      "step": 6971,
      "time_per_iteration": 2.6632890701293945
    },
    {
      "auxiliary_loss_clip": 0.01168824,
      "auxiliary_loss_mlp": 0.01022679,
      "balance_loss_clip": 1.04502797,
      "balance_loss_mlp": 1.01552081,
      "epoch": 0.8383334335357422,
      "flos": 17092653920640.0,
      "grad_norm": 2.31596025276731,
      "language_loss": 0.82234234,
      "learning_rate": 2.6781937842086557e-07,
      "loss": 0.84425735,
      "num_input_tokens_seen": 150593995,
      "step": 6972,
      "time_per_iteration": 2.6071925163269043
    },
    {
      "auxiliary_loss_clip": 0.01170452,
      "auxiliary_loss_mlp": 0.01028004,
      "balance_loss_clip": 1.04535401,
      "balance_loss_mlp": 1.02057147,
      "epoch": 0.8384536764263812,
      "flos": 20704728562560.0,
      "grad_norm": 2.067942550576771,
      "language_loss": 0.67491388,
      "learning_rate": 2.6743011157912933e-07,
      "loss": 0.69689846,
      "num_input_tokens_seen": 150613715,
      "step": 6973,
      "time_per_iteration": 3.5630156993865967
    },
    {
      "auxiliary_loss_clip": 0.01176021,
      "auxiliary_loss_mlp": 0.01023481,
      "balance_loss_clip": 1.03841066,
      "balance_loss_mlp": 1.01525569,
      "epoch": 0.8385739193170204,
      "flos": 28986842056320.0,
      "grad_norm": 4.241525529937181,
      "language_loss": 0.65326667,
      "learning_rate": 2.6704110756686725e-07,
      "loss": 0.67526168,
      "num_input_tokens_seen": 150634540,
      "step": 6974,
      "time_per_iteration": 3.4842631816864014
    },
    {
      "auxiliary_loss_clip": 0.01168207,
      "auxiliary_loss_mlp": 0.01207429,
      "balance_loss_clip": 1.04221964,
      "balance_loss_mlp": 1.00075173,
      "epoch": 0.8386941622076595,
      "flos": 23438068882560.0,
      "grad_norm": 3.1845584052567317,
      "language_loss": 0.84022784,
      "learning_rate": 2.6665236644309085e-07,
      "loss": 0.86398429,
      "num_input_tokens_seen": 150654850,
      "step": 6975,
      "time_per_iteration": 2.771084785461426
    },
    {
      "auxiliary_loss_clip": 0.01169022,
      "auxiliary_loss_mlp": 0.0102254,
      "balance_loss_clip": 1.04437053,
      "balance_loss_mlp": 1.01565003,
      "epoch": 0.8388144050982985,
      "flos": 23002724044800.0,
      "grad_norm": 2.4894077349227333,
      "language_loss": 0.7984522,
      "learning_rate": 2.662638882667727e-07,
      "loss": 0.82036781,
      "num_input_tokens_seen": 150673790,
      "step": 6976,
      "time_per_iteration": 2.7895712852478027
    },
    {
      "auxiliary_loss_clip": 0.01169583,
      "auxiliary_loss_mlp": 0.01026827,
      "balance_loss_clip": 1.04759502,
      "balance_loss_mlp": 1.01888728,
      "epoch": 0.8389346479889377,
      "flos": 24280353878400.0,
      "grad_norm": 1.895146795904957,
      "language_loss": 0.73416489,
      "learning_rate": 2.658756730968443e-07,
      "loss": 0.75612903,
      "num_input_tokens_seen": 150692255,
      "step": 6977,
      "time_per_iteration": 2.7024049758911133
    },
    {
      "auxiliary_loss_clip": 0.01176885,
      "auxiliary_loss_mlp": 0.01023144,
      "balance_loss_clip": 1.04831672,
      "balance_loss_mlp": 1.01582468,
      "epoch": 0.8390548908795767,
      "flos": 21215019127680.0,
      "grad_norm": 3.0706546133212482,
      "language_loss": 0.88264525,
      "learning_rate": 2.654877209921975e-07,
      "loss": 0.90464556,
      "num_input_tokens_seen": 150709790,
      "step": 6978,
      "time_per_iteration": 2.706303119659424
    },
    {
      "auxiliary_loss_clip": 0.01184201,
      "auxiliary_loss_mlp": 0.01029888,
      "balance_loss_clip": 1.04080176,
      "balance_loss_mlp": 1.02107227,
      "epoch": 0.8391751337702158,
      "flos": 35627299332480.0,
      "grad_norm": 2.507474696081912,
      "language_loss": 0.62849033,
      "learning_rate": 2.651000320116843e-07,
      "loss": 0.65063119,
      "num_input_tokens_seen": 150730675,
      "step": 6979,
      "time_per_iteration": 3.7971255779266357
    },
    {
      "auxiliary_loss_clip": 0.01173375,
      "auxiliary_loss_mlp": 0.01208033,
      "balance_loss_clip": 1.043648,
      "balance_loss_mlp": 1.00072336,
      "epoch": 0.839295376660855,
      "flos": 21325229032320.0,
      "grad_norm": 2.5265861014967843,
      "language_loss": 0.76686788,
      "learning_rate": 2.647126062141163e-07,
      "loss": 0.79068196,
      "num_input_tokens_seen": 150749750,
      "step": 6980,
      "time_per_iteration": 2.6870462894439697
    },
    {
      "auxiliary_loss_clip": 0.01175152,
      "auxiliary_loss_mlp": 0.01023179,
      "balance_loss_clip": 1.04205692,
      "balance_loss_mlp": 1.01594853,
      "epoch": 0.839415619551494,
      "flos": 18442535961600.0,
      "grad_norm": 3.1706963744270293,
      "language_loss": 0.8429516,
      "learning_rate": 2.643254436582669e-07,
      "loss": 0.86493492,
      "num_input_tokens_seen": 150769240,
      "step": 6981,
      "time_per_iteration": 2.75103497505188
    },
    {
      "auxiliary_loss_clip": 0.01177998,
      "auxiliary_loss_mlp": 0.01023574,
      "balance_loss_clip": 1.04183996,
      "balance_loss_mlp": 1.01595914,
      "epoch": 0.8395358624421331,
      "flos": 23221958705280.0,
      "grad_norm": 1.875600777016739,
      "language_loss": 0.82331836,
      "learning_rate": 2.6393854440286743e-07,
      "loss": 0.84533405,
      "num_input_tokens_seen": 150788410,
      "step": 6982,
      "time_per_iteration": 2.6734619140625
    },
    {
      "auxiliary_loss_clip": 0.01167989,
      "auxiliary_loss_mlp": 0.01025883,
      "balance_loss_clip": 1.04933786,
      "balance_loss_mlp": 1.01846874,
      "epoch": 0.8396561053327722,
      "flos": 24381657210240.0,
      "grad_norm": 2.0415477231641184,
      "language_loss": 0.70918882,
      "learning_rate": 2.6355190850661045e-07,
      "loss": 0.7311275,
      "num_input_tokens_seen": 150805245,
      "step": 6983,
      "time_per_iteration": 2.648805856704712
    },
    {
      "auxiliary_loss_clip": 0.01171934,
      "auxiliary_loss_mlp": 0.01023598,
      "balance_loss_clip": 1.04542112,
      "balance_loss_mlp": 1.01588559,
      "epoch": 0.8397763482234113,
      "flos": 22237755073920.0,
      "grad_norm": 1.5756961140114814,
      "language_loss": 0.8666265,
      "learning_rate": 2.631655360281486e-07,
      "loss": 0.88858181,
      "num_input_tokens_seen": 150824920,
      "step": 6984,
      "time_per_iteration": 2.629054546356201
    },
    {
      "auxiliary_loss_clip": 0.01174472,
      "auxiliary_loss_mlp": 0.01207759,
      "balance_loss_clip": 1.04540658,
      "balance_loss_mlp": 1.00079048,
      "epoch": 0.8398965911140504,
      "flos": 22163743100160.0,
      "grad_norm": 1.9002990576181724,
      "language_loss": 0.6604079,
      "learning_rate": 2.6277942702609323e-07,
      "loss": 0.68423021,
      "num_input_tokens_seen": 150844400,
      "step": 6985,
      "time_per_iteration": 2.659217119216919
    },
    {
      "auxiliary_loss_clip": 0.01176923,
      "auxiliary_loss_mlp": 0.01025408,
      "balance_loss_clip": 1.04433894,
      "balance_loss_mlp": 1.01805234,
      "epoch": 0.8400168340046895,
      "flos": 21542775753600.0,
      "grad_norm": 9.866660743380175,
      "language_loss": 0.87904346,
      "learning_rate": 2.623935815590186e-07,
      "loss": 0.90106678,
      "num_input_tokens_seen": 150862780,
      "step": 6986,
      "time_per_iteration": 2.6386878490448
    },
    {
      "auxiliary_loss_clip": 0.0117241,
      "auxiliary_loss_mlp": 0.01021402,
      "balance_loss_clip": 1.04724658,
      "balance_loss_mlp": 1.01410639,
      "epoch": 0.8401370768953286,
      "flos": 22491966602880.0,
      "grad_norm": 1.7859790000983344,
      "language_loss": 0.81103063,
      "learning_rate": 2.6200799968545516e-07,
      "loss": 0.83296871,
      "num_input_tokens_seen": 150883075,
      "step": 6987,
      "time_per_iteration": 2.717200756072998
    },
    {
      "auxiliary_loss_clip": 0.01074257,
      "auxiliary_loss_mlp": 0.00999313,
      "balance_loss_clip": 1.01500201,
      "balance_loss_mlp": 0.99864256,
      "epoch": 0.8402573197859676,
      "flos": 59238890818560.0,
      "grad_norm": 0.7880751987056268,
      "language_loss": 0.56422281,
      "learning_rate": 2.616226814638969e-07,
      "loss": 0.58495849,
      "num_input_tokens_seen": 150948180,
      "step": 6988,
      "time_per_iteration": 3.2256290912628174
    },
    {
      "auxiliary_loss_clip": 0.01174741,
      "auxiliary_loss_mlp": 0.01024271,
      "balance_loss_clip": 1.04629612,
      "balance_loss_mlp": 1.0172677,
      "epoch": 0.8403775626766068,
      "flos": 22674608282880.0,
      "grad_norm": 2.0579652364183914,
      "language_loss": 0.77953959,
      "learning_rate": 2.612376269527954e-07,
      "loss": 0.80152977,
      "num_input_tokens_seen": 150967885,
      "step": 6989,
      "time_per_iteration": 2.6532154083251953
    },
    {
      "auxiliary_loss_clip": 0.01168951,
      "auxiliary_loss_mlp": 0.01025486,
      "balance_loss_clip": 1.04477167,
      "balance_loss_mlp": 1.01852751,
      "epoch": 0.8404978055672458,
      "flos": 19609704495360.0,
      "grad_norm": 6.626523689487392,
      "language_loss": 0.67752135,
      "learning_rate": 2.608528362105635e-07,
      "loss": 0.69946575,
      "num_input_tokens_seen": 150987255,
      "step": 6990,
      "time_per_iteration": 2.7444093227386475
    },
    {
      "auxiliary_loss_clip": 0.0117454,
      "auxiliary_loss_mlp": 0.01023857,
      "balance_loss_clip": 1.0397315,
      "balance_loss_mlp": 1.0164721,
      "epoch": 0.8406180484578849,
      "flos": 27526929678720.0,
      "grad_norm": 2.147091471035175,
      "language_loss": 0.73776186,
      "learning_rate": 2.6046830929557374e-07,
      "loss": 0.75974584,
      "num_input_tokens_seen": 151006905,
      "step": 6991,
      "time_per_iteration": 2.711765766143799
    },
    {
      "auxiliary_loss_clip": 0.01171468,
      "auxiliary_loss_mlp": 0.01024651,
      "balance_loss_clip": 1.04403567,
      "balance_loss_mlp": 1.01745069,
      "epoch": 0.8407382913485241,
      "flos": 22127473342080.0,
      "grad_norm": 2.1136178153612373,
      "language_loss": 0.84935534,
      "learning_rate": 2.6008404626615776e-07,
      "loss": 0.87131649,
      "num_input_tokens_seen": 151025405,
      "step": 6992,
      "time_per_iteration": 2.7023911476135254
    },
    {
      "auxiliary_loss_clip": 0.01177048,
      "auxiliary_loss_mlp": 0.01024374,
      "balance_loss_clip": 1.04919314,
      "balance_loss_mlp": 1.01722431,
      "epoch": 0.8408585342391631,
      "flos": 13918473982080.0,
      "grad_norm": 2.88743064487771,
      "language_loss": 0.74069333,
      "learning_rate": 2.597000471806092e-07,
      "loss": 0.76270759,
      "num_input_tokens_seen": 151041970,
      "step": 6993,
      "time_per_iteration": 3.5564749240875244
    },
    {
      "auxiliary_loss_clip": 0.01167901,
      "auxiliary_loss_mlp": 0.01025929,
      "balance_loss_clip": 1.04766321,
      "balance_loss_mlp": 1.01754916,
      "epoch": 0.8409787771298022,
      "flos": 20187865808640.0,
      "grad_norm": 3.347273456498881,
      "language_loss": 0.73245293,
      "learning_rate": 2.593163120971793e-07,
      "loss": 0.75439119,
      "num_input_tokens_seen": 151060835,
      "step": 6994,
      "time_per_iteration": 2.64990496635437
    },
    {
      "auxiliary_loss_clip": 0.0116607,
      "auxiliary_loss_mlp": 0.0102348,
      "balance_loss_clip": 1.03673863,
      "balance_loss_mlp": 1.01619375,
      "epoch": 0.8410990200204413,
      "flos": 23142523777920.0,
      "grad_norm": 2.035726463003443,
      "language_loss": 0.68877375,
      "learning_rate": 2.5893284107408165e-07,
      "loss": 0.71066928,
      "num_input_tokens_seen": 151078205,
      "step": 6995,
      "time_per_iteration": 2.7452895641326904
    },
    {
      "auxiliary_loss_clip": 0.01178219,
      "auxiliary_loss_mlp": 0.01027289,
      "balance_loss_clip": 1.04376578,
      "balance_loss_mlp": 1.01948118,
      "epoch": 0.8412192629110804,
      "flos": 24027219757440.0,
      "grad_norm": 1.983028381776354,
      "language_loss": 0.77992076,
      "learning_rate": 2.5854963416948726e-07,
      "loss": 0.80197585,
      "num_input_tokens_seen": 151100470,
      "step": 6996,
      "time_per_iteration": 2.700516939163208
    },
    {
      "auxiliary_loss_clip": 0.01174856,
      "auxiliary_loss_mlp": 0.01025358,
      "balance_loss_clip": 1.03586733,
      "balance_loss_mlp": 1.01781189,
      "epoch": 0.8413395058017195,
      "flos": 25591703604480.0,
      "grad_norm": 2.124541514697946,
      "language_loss": 0.69580859,
      "learning_rate": 2.5816669144152816e-07,
      "loss": 0.71781075,
      "num_input_tokens_seen": 151121650,
      "step": 6997,
      "time_per_iteration": 2.752993583679199
    },
    {
      "auxiliary_loss_clip": 0.01064073,
      "auxiliary_loss_mlp": 0.0100142,
      "balance_loss_clip": 1.01125515,
      "balance_loss_mlp": 1.000705,
      "epoch": 0.8414597486923585,
      "flos": 63635396624640.0,
      "grad_norm": 0.8518011540337777,
      "language_loss": 0.66304529,
      "learning_rate": 2.5778401294829777e-07,
      "loss": 0.6837002,
      "num_input_tokens_seen": 151180390,
      "step": 6998,
      "time_per_iteration": 3.2034506797790527
    },
    {
      "auxiliary_loss_clip": 0.01166373,
      "auxiliary_loss_mlp": 0.01207537,
      "balance_loss_clip": 1.04615152,
      "balance_loss_mlp": 1.00071836,
      "epoch": 0.8415799915829977,
      "flos": 19098731571840.0,
      "grad_norm": 1.8323239136848553,
      "language_loss": 0.65179014,
      "learning_rate": 2.574015987478473e-07,
      "loss": 0.67552924,
      "num_input_tokens_seen": 151198520,
      "step": 6999,
      "time_per_iteration": 2.660203218460083
    },
    {
      "auxiliary_loss_clip": 0.01177472,
      "auxiliary_loss_mlp": 0.01023618,
      "balance_loss_clip": 1.04558337,
      "balance_loss_mlp": 1.01628637,
      "epoch": 0.8417002344736367,
      "flos": 19821612781440.0,
      "grad_norm": 2.4939660852220586,
      "language_loss": 0.87123919,
      "learning_rate": 2.570194488981887e-07,
      "loss": 0.89325011,
      "num_input_tokens_seen": 151215065,
      "step": 7000,
      "time_per_iteration": 3.6630935668945312
    },
    {
      "auxiliary_loss_clip": 0.01064176,
      "auxiliary_loss_mlp": 0.01001115,
      "balance_loss_clip": 1.01135719,
      "balance_loss_mlp": 1.00030482,
      "epoch": 0.8418204773642758,
      "flos": 62161516834560.0,
      "grad_norm": 1.7736758657180591,
      "language_loss": 0.60281193,
      "learning_rate": 2.566375634572939e-07,
      "loss": 0.62346482,
      "num_input_tokens_seen": 151275705,
      "step": 7001,
      "time_per_iteration": 3.964751958847046
    },
    {
      "auxiliary_loss_clip": 0.01178178,
      "auxiliary_loss_mlp": 0.01027889,
      "balance_loss_clip": 1.04109395,
      "balance_loss_mlp": 1.02093911,
      "epoch": 0.841940720254915,
      "flos": 17092905315840.0,
      "grad_norm": 1.8263059093536556,
      "language_loss": 0.76555514,
      "learning_rate": 2.562559424830943e-07,
      "loss": 0.78761578,
      "num_input_tokens_seen": 151293665,
      "step": 7002,
      "time_per_iteration": 2.8598732948303223
    },
    {
      "auxiliary_loss_clip": 0.01169112,
      "auxiliary_loss_mlp": 0.01021638,
      "balance_loss_clip": 1.04273772,
      "balance_loss_mlp": 1.01405072,
      "epoch": 0.842060963145554,
      "flos": 16283586026880.0,
      "grad_norm": 2.2841214393215052,
      "language_loss": 0.7066766,
      "learning_rate": 2.5587458603348256e-07,
      "loss": 0.72858417,
      "num_input_tokens_seen": 151310955,
      "step": 7003,
      "time_per_iteration": 2.702651023864746
    },
    {
      "auxiliary_loss_clip": 0.01167923,
      "auxiliary_loss_mlp": 0.01025937,
      "balance_loss_clip": 1.04216599,
      "balance_loss_mlp": 1.01856351,
      "epoch": 0.8421812060361931,
      "flos": 21908238681600.0,
      "grad_norm": 1.990083883011531,
      "language_loss": 0.8395257,
      "learning_rate": 2.554934941663085e-07,
      "loss": 0.86146432,
      "num_input_tokens_seen": 151328490,
      "step": 7004,
      "time_per_iteration": 2.813652515411377
    },
    {
      "auxiliary_loss_clip": 0.01173858,
      "auxiliary_loss_mlp": 0.01023998,
      "balance_loss_clip": 1.04240966,
      "balance_loss_mlp": 1.01580262,
      "epoch": 0.8423014489268322,
      "flos": 27777693502080.0,
      "grad_norm": 1.9908615385216535,
      "language_loss": 0.73562026,
      "learning_rate": 2.5511266693938484e-07,
      "loss": 0.75759888,
      "num_input_tokens_seen": 151346950,
      "step": 7005,
      "time_per_iteration": 2.8465588092803955
    },
    {
      "auxiliary_loss_clip": 0.0116992,
      "auxiliary_loss_mlp": 0.01022201,
      "balance_loss_clip": 1.04557967,
      "balance_loss_mlp": 1.01433611,
      "epoch": 0.8424216918174713,
      "flos": 25117610970240.0,
      "grad_norm": 1.664750061421613,
      "language_loss": 0.77728385,
      "learning_rate": 2.547321044104822e-07,
      "loss": 0.79920506,
      "num_input_tokens_seen": 151368445,
      "step": 7006,
      "time_per_iteration": 3.6213152408599854
    },
    {
      "auxiliary_loss_clip": 0.01169242,
      "auxiliary_loss_mlp": 0.01022867,
      "balance_loss_clip": 1.04828954,
      "balance_loss_mlp": 1.01493418,
      "epoch": 0.8425419347081103,
      "flos": 24748448941440.0,
      "grad_norm": 3.670215260809123,
      "language_loss": 0.76859558,
      "learning_rate": 2.5435180663733113e-07,
      "loss": 0.79051673,
      "num_input_tokens_seen": 151388745,
      "step": 7007,
      "time_per_iteration": 2.812932014465332
    },
    {
      "auxiliary_loss_clip": 0.01186178,
      "auxiliary_loss_mlp": 0.01028519,
      "balance_loss_clip": 1.04139185,
      "balance_loss_mlp": 1.0211066,
      "epoch": 0.8426621775987495,
      "flos": 24820916630400.0,
      "grad_norm": 2.2801319976458485,
      "language_loss": 0.71955383,
      "learning_rate": 2.539717736776241e-07,
      "loss": 0.74170077,
      "num_input_tokens_seen": 151404970,
      "step": 7008,
      "time_per_iteration": 2.7370691299438477
    },
    {
      "auxiliary_loss_clip": 0.01165209,
      "auxiliary_loss_mlp": 0.01020764,
      "balance_loss_clip": 1.04556394,
      "balance_loss_mlp": 1.0134027,
      "epoch": 0.8427824204893886,
      "flos": 23550074467200.0,
      "grad_norm": 1.6140013015758152,
      "language_loss": 0.76402378,
      "learning_rate": 2.535920055890097e-07,
      "loss": 0.78588349,
      "num_input_tokens_seen": 151426265,
      "step": 7009,
      "time_per_iteration": 2.7002766132354736
    },
    {
      "auxiliary_loss_clip": 0.01173685,
      "auxiliary_loss_mlp": 0.01023859,
      "balance_loss_clip": 1.04156351,
      "balance_loss_mlp": 1.01596713,
      "epoch": 0.8429026633800276,
      "flos": 16143858120960.0,
      "grad_norm": 4.017318068305195,
      "language_loss": 0.64408422,
      "learning_rate": 2.5321250242910006e-07,
      "loss": 0.66605973,
      "num_input_tokens_seen": 151444180,
      "step": 7010,
      "time_per_iteration": 2.704080104827881
    },
    {
      "auxiliary_loss_clip": 0.01167818,
      "auxiliary_loss_mlp": 0.01019433,
      "balance_loss_clip": 1.04905331,
      "balance_loss_mlp": 1.01229572,
      "epoch": 0.8430229062706668,
      "flos": 22198540400640.0,
      "grad_norm": 1.6573963301944488,
      "language_loss": 0.86381429,
      "learning_rate": 2.5283326425546493e-07,
      "loss": 0.88568676,
      "num_input_tokens_seen": 151463290,
      "step": 7011,
      "time_per_iteration": 2.6643168926239014
    },
    {
      "auxiliary_loss_clip": 0.01166975,
      "auxiliary_loss_mlp": 0.01021121,
      "balance_loss_clip": 1.04404092,
      "balance_loss_mlp": 1.0141114,
      "epoch": 0.8431431491613058,
      "flos": 35330317683840.0,
      "grad_norm": 1.864365179744693,
      "language_loss": 0.6965549,
      "learning_rate": 2.5245429112563443e-07,
      "loss": 0.71843588,
      "num_input_tokens_seen": 151483965,
      "step": 7012,
      "time_per_iteration": 2.811692714691162
    },
    {
      "auxiliary_loss_clip": 0.01169818,
      "auxiliary_loss_mlp": 0.01026846,
      "balance_loss_clip": 1.04722321,
      "balance_loss_mlp": 1.01972318,
      "epoch": 0.8432633920519449,
      "flos": 25812374808960.0,
      "grad_norm": 2.8348016438314168,
      "language_loss": 0.82452792,
      "learning_rate": 2.5207558309709865e-07,
      "loss": 0.84649456,
      "num_input_tokens_seen": 151503700,
      "step": 7013,
      "time_per_iteration": 2.643099308013916
    },
    {
      "auxiliary_loss_clip": 0.01087331,
      "auxiliary_loss_mlp": 0.01198785,
      "balance_loss_clip": 1.01089096,
      "balance_loss_mlp": 0.99984837,
      "epoch": 0.8433836349425841,
      "flos": 64959531592320.0,
      "grad_norm": 0.6593106713076297,
      "language_loss": 0.56213808,
      "learning_rate": 2.516971402273065e-07,
      "loss": 0.58499926,
      "num_input_tokens_seen": 151569765,
      "step": 7014,
      "time_per_iteration": 3.201387405395508
    },
    {
      "auxiliary_loss_clip": 0.01174188,
      "auxiliary_loss_mlp": 0.01020946,
      "balance_loss_clip": 1.04432726,
      "balance_loss_mlp": 1.0136925,
      "epoch": 0.8435038778332231,
      "flos": 20229989483520.0,
      "grad_norm": 2.2213704747407617,
      "language_loss": 0.67916143,
      "learning_rate": 2.513189625736687e-07,
      "loss": 0.70111281,
      "num_input_tokens_seen": 151586660,
      "step": 7015,
      "time_per_iteration": 2.6372828483581543
    },
    {
      "auxiliary_loss_clip": 0.01180039,
      "auxiliary_loss_mlp": 0.01030337,
      "balance_loss_clip": 1.04526401,
      "balance_loss_mlp": 1.0224576,
      "epoch": 0.8436241207238622,
      "flos": 20992229020800.0,
      "grad_norm": 2.4508606920760445,
      "language_loss": 0.71361935,
      "learning_rate": 2.509410501935534e-07,
      "loss": 0.73572314,
      "num_input_tokens_seen": 151602295,
      "step": 7016,
      "time_per_iteration": 2.637521982192993
    },
    {
      "auxiliary_loss_clip": 0.01175239,
      "auxiliary_loss_mlp": 0.01023564,
      "balance_loss_clip": 1.04514027,
      "balance_loss_mlp": 1.01575553,
      "epoch": 0.8437443636145013,
      "flos": 14682257804160.0,
      "grad_norm": 2.3259254796956954,
      "language_loss": 0.75698042,
      "learning_rate": 2.5056340314429116e-07,
      "loss": 0.77896845,
      "num_input_tokens_seen": 151619760,
      "step": 7017,
      "time_per_iteration": 2.576733350753784
    },
    {
      "auxiliary_loss_clip": 0.01181096,
      "auxiliary_loss_mlp": 0.01028007,
      "balance_loss_clip": 1.04093003,
      "balance_loss_mlp": 1.01981187,
      "epoch": 0.8438646065051404,
      "flos": 21608814908160.0,
      "grad_norm": 3.0777637623428675,
      "language_loss": 0.80522478,
      "learning_rate": 2.5018602148316904e-07,
      "loss": 0.82731581,
      "num_input_tokens_seen": 151635795,
      "step": 7018,
      "time_per_iteration": 2.6594018936157227
    },
    {
      "auxiliary_loss_clip": 0.01166277,
      "auxiliary_loss_mlp": 0.01024396,
      "balance_loss_clip": 1.04908037,
      "balance_loss_mlp": 1.01728535,
      "epoch": 0.8439848493957794,
      "flos": 23289937194240.0,
      "grad_norm": 6.3393423700737985,
      "language_loss": 0.80553049,
      "learning_rate": 2.498089052674359e-07,
      "loss": 0.82743722,
      "num_input_tokens_seen": 151653770,
      "step": 7019,
      "time_per_iteration": 3.5130577087402344
    },
    {
      "auxiliary_loss_clip": 0.01170161,
      "auxiliary_loss_mlp": 0.0103328,
      "balance_loss_clip": 1.04765141,
      "balance_loss_mlp": 1.02557898,
      "epoch": 0.8441050922864186,
      "flos": 19719339782400.0,
      "grad_norm": 2.456401492548943,
      "language_loss": 0.7553857,
      "learning_rate": 2.494320545543007e-07,
      "loss": 0.77742016,
      "num_input_tokens_seen": 151673340,
      "step": 7020,
      "time_per_iteration": 2.571737051010132
    },
    {
      "auxiliary_loss_clip": 0.01171007,
      "auxiliary_loss_mlp": 0.01027221,
      "balance_loss_clip": 1.04841459,
      "balance_loss_mlp": 1.01873899,
      "epoch": 0.8442253351770577,
      "flos": 21835268202240.0,
      "grad_norm": 2.0434880712836607,
      "language_loss": 0.66502076,
      "learning_rate": 2.490554694009308e-07,
      "loss": 0.68700302,
      "num_input_tokens_seen": 151694205,
      "step": 7021,
      "time_per_iteration": 2.620946168899536
    },
    {
      "auxiliary_loss_clip": 0.01171571,
      "auxiliary_loss_mlp": 0.01027403,
      "balance_loss_clip": 1.04445815,
      "balance_loss_mlp": 1.02039361,
      "epoch": 0.8443455780676967,
      "flos": 34346365447680.0,
      "grad_norm": 1.589837935884254,
      "language_loss": 0.78470397,
      "learning_rate": 2.4867914986445426e-07,
      "loss": 0.80669367,
      "num_input_tokens_seen": 151716595,
      "step": 7022,
      "time_per_iteration": 2.7303502559661865
    },
    {
      "auxiliary_loss_clip": 0.01174808,
      "auxiliary_loss_mlp": 0.01025434,
      "balance_loss_clip": 1.04331779,
      "balance_loss_mlp": 1.01847851,
      "epoch": 0.8444658209583359,
      "flos": 48214599281280.0,
      "grad_norm": 2.6406935277823558,
      "language_loss": 0.71119606,
      "learning_rate": 2.483030960019581e-07,
      "loss": 0.73319852,
      "num_input_tokens_seen": 151740525,
      "step": 7023,
      "time_per_iteration": 2.903935194015503
    },
    {
      "auxiliary_loss_clip": 0.01088738,
      "auxiliary_loss_mlp": 0.01002088,
      "balance_loss_clip": 1.01257706,
      "balance_loss_mlp": 1.0012356,
      "epoch": 0.8445860638489749,
      "flos": 68484773105280.0,
      "grad_norm": 0.746807094591477,
      "language_loss": 0.55408496,
      "learning_rate": 2.479273078704891e-07,
      "loss": 0.57499325,
      "num_input_tokens_seen": 151793890,
      "step": 7024,
      "time_per_iteration": 3.0009312629699707
    },
    {
      "auxiliary_loss_clip": 0.01083206,
      "auxiliary_loss_mlp": 0.01002282,
      "balance_loss_clip": 1.01396692,
      "balance_loss_mlp": 1.00153148,
      "epoch": 0.844706306739614,
      "flos": 62833331882880.0,
      "grad_norm": 0.8208447817996573,
      "language_loss": 0.64702451,
      "learning_rate": 2.475517855270552e-07,
      "loss": 0.66787946,
      "num_input_tokens_seen": 151853970,
      "step": 7025,
      "time_per_iteration": 3.1454970836639404
    },
    {
      "auxiliary_loss_clip": 0.01166227,
      "auxiliary_loss_mlp": 0.01023778,
      "balance_loss_clip": 1.04836309,
      "balance_loss_mlp": 1.01687574,
      "epoch": 0.8448265496302532,
      "flos": 14976114969600.0,
      "grad_norm": 2.354524363526939,
      "language_loss": 0.72696173,
      "learning_rate": 2.4717652902862143e-07,
      "loss": 0.74886179,
      "num_input_tokens_seen": 151872945,
      "step": 7026,
      "time_per_iteration": 3.518850564956665
    },
    {
      "auxiliary_loss_clip": 0.01176419,
      "auxiliary_loss_mlp": 0.01023914,
      "balance_loss_clip": 1.04544365,
      "balance_loss_mlp": 1.01676142,
      "epoch": 0.8449467925208922,
      "flos": 23441265192960.0,
      "grad_norm": 1.8754580865054122,
      "language_loss": 0.81405199,
      "learning_rate": 2.4680153843211495e-07,
      "loss": 0.83605534,
      "num_input_tokens_seen": 151892875,
      "step": 7027,
      "time_per_iteration": 2.5850682258605957
    },
    {
      "auxiliary_loss_clip": 0.01172857,
      "auxiliary_loss_mlp": 0.01028396,
      "balance_loss_clip": 1.04692161,
      "balance_loss_mlp": 1.02062964,
      "epoch": 0.8450670354115313,
      "flos": 22748045639040.0,
      "grad_norm": 1.8414749840549334,
      "language_loss": 0.72507322,
      "learning_rate": 2.464268137944212e-07,
      "loss": 0.74708581,
      "num_input_tokens_seen": 151914170,
      "step": 7028,
      "time_per_iteration": 3.556330680847168
    },
    {
      "auxiliary_loss_clip": 0.01165702,
      "auxiliary_loss_mlp": 0.01027345,
      "balance_loss_clip": 1.03873062,
      "balance_loss_mlp": 1.01889896,
      "epoch": 0.8451872783021703,
      "flos": 29825571605760.0,
      "grad_norm": 2.163321059407809,
      "language_loss": 0.7839855,
      "learning_rate": 2.46052355172385e-07,
      "loss": 0.80591601,
      "num_input_tokens_seen": 151932210,
      "step": 7029,
      "time_per_iteration": 2.724594831466675
    },
    {
      "auxiliary_loss_clip": 0.01167222,
      "auxiliary_loss_mlp": 0.01027985,
      "balance_loss_clip": 1.04717445,
      "balance_loss_mlp": 1.01962864,
      "epoch": 0.8453075211928095,
      "flos": 21870029589120.0,
      "grad_norm": 2.631833565518842,
      "language_loss": 0.74693465,
      "learning_rate": 2.456781626228128e-07,
      "loss": 0.7688868,
      "num_input_tokens_seen": 151951715,
      "step": 7030,
      "time_per_iteration": 2.5825631618499756
    },
    {
      "auxiliary_loss_clip": 0.01092611,
      "auxiliary_loss_mlp": 0.0119867,
      "balance_loss_clip": 1.01195788,
      "balance_loss_mlp": 0.99985182,
      "epoch": 0.8454277640834486,
      "flos": 58751869288320.0,
      "grad_norm": 0.9114508019603413,
      "language_loss": 0.66229153,
      "learning_rate": 2.453042362024675e-07,
      "loss": 0.68520433,
      "num_input_tokens_seen": 152004960,
      "step": 7031,
      "time_per_iteration": 3.219367742538452
    },
    {
      "auxiliary_loss_clip": 0.01165404,
      "auxiliary_loss_mlp": 0.01024691,
      "balance_loss_clip": 1.04671156,
      "balance_loss_mlp": 1.01750278,
      "epoch": 0.8455480069740876,
      "flos": 27090076469760.0,
      "grad_norm": 1.5665508162775057,
      "language_loss": 0.73191798,
      "learning_rate": 2.449305759680751e-07,
      "loss": 0.75381893,
      "num_input_tokens_seen": 152026285,
      "step": 7032,
      "time_per_iteration": 3.5529873371124268
    },
    {
      "auxiliary_loss_clip": 0.01174415,
      "auxiliary_loss_mlp": 0.01026591,
      "balance_loss_clip": 1.04466844,
      "balance_loss_mlp": 1.01928997,
      "epoch": 0.8456682498647268,
      "flos": 27198670262400.0,
      "grad_norm": 1.4189438689416385,
      "language_loss": 0.75087237,
      "learning_rate": 2.445571819763188e-07,
      "loss": 0.77288246,
      "num_input_tokens_seen": 152048585,
      "step": 7033,
      "time_per_iteration": 2.7174277305603027
    },
    {
      "auxiliary_loss_clip": 0.01168618,
      "auxiliary_loss_mlp": 0.01029086,
      "balance_loss_clip": 1.04965472,
      "balance_loss_mlp": 1.02129912,
      "epoch": 0.8457884927553658,
      "flos": 20631901737600.0,
      "grad_norm": 2.3466238618628874,
      "language_loss": 0.58809847,
      "learning_rate": 2.4418405428384227e-07,
      "loss": 0.61007547,
      "num_input_tokens_seen": 152068795,
      "step": 7034,
      "time_per_iteration": 2.600834846496582
    },
    {
      "auxiliary_loss_clip": 0.01168484,
      "auxiliary_loss_mlp": 0.01207892,
      "balance_loss_clip": 1.04840624,
      "balance_loss_mlp": 1.00077891,
      "epoch": 0.8459087356460049,
      "flos": 15299023259520.0,
      "grad_norm": 11.066277480941022,
      "language_loss": 0.72000134,
      "learning_rate": 2.4381119294724864e-07,
      "loss": 0.74376506,
      "num_input_tokens_seen": 152086240,
      "step": 7035,
      "time_per_iteration": 2.6011838912963867
    },
    {
      "auxiliary_loss_clip": 0.01167009,
      "auxiliary_loss_mlp": 0.01022143,
      "balance_loss_clip": 1.04712367,
      "balance_loss_mlp": 1.0148716,
      "epoch": 0.846028978536644,
      "flos": 18843155326080.0,
      "grad_norm": 2.759201363903635,
      "language_loss": 0.53574014,
      "learning_rate": 2.434385980231004e-07,
      "loss": 0.55763167,
      "num_input_tokens_seen": 152105080,
      "step": 7036,
      "time_per_iteration": 2.5560038089752197
    },
    {
      "auxiliary_loss_clip": 0.01168261,
      "auxiliary_loss_mlp": 0.01027261,
      "balance_loss_clip": 1.04597485,
      "balance_loss_mlp": 1.01972747,
      "epoch": 0.8461492214272831,
      "flos": 52661740285440.0,
      "grad_norm": 2.6130350320309423,
      "language_loss": 0.65894747,
      "learning_rate": 2.4306626956792043e-07,
      "loss": 0.68090272,
      "num_input_tokens_seen": 152130025,
      "step": 7037,
      "time_per_iteration": 2.9157555103302
    },
    {
      "auxiliary_loss_clip": 0.01167252,
      "auxiliary_loss_mlp": 0.01024198,
      "balance_loss_clip": 1.04378557,
      "balance_loss_mlp": 1.01707768,
      "epoch": 0.8462694643179222,
      "flos": 18588405093120.0,
      "grad_norm": 1.7237163016962314,
      "language_loss": 0.75791001,
      "learning_rate": 2.4269420763819017e-07,
      "loss": 0.7798245,
      "num_input_tokens_seen": 152148070,
      "step": 7038,
      "time_per_iteration": 2.648099184036255
    },
    {
      "auxiliary_loss_clip": 0.01164899,
      "auxiliary_loss_mlp": 0.01025346,
      "balance_loss_clip": 1.04465973,
      "balance_loss_mlp": 1.01865208,
      "epoch": 0.8463897072085613,
      "flos": 24387080163840.0,
      "grad_norm": 2.92249899141236,
      "language_loss": 0.83426845,
      "learning_rate": 2.4232241229035223e-07,
      "loss": 0.85617089,
      "num_input_tokens_seen": 152165825,
      "step": 7039,
      "time_per_iteration": 2.671191453933716
    },
    {
      "auxiliary_loss_clip": 0.01071821,
      "auxiliary_loss_mlp": 0.01001133,
      "balance_loss_clip": 1.01159787,
      "balance_loss_mlp": 1.00035214,
      "epoch": 0.8465099500992004,
      "flos": 68702140258560.0,
      "grad_norm": 0.7489228019637337,
      "language_loss": 0.56720376,
      "learning_rate": 2.419508835808064e-07,
      "loss": 0.5879333,
      "num_input_tokens_seen": 152222380,
      "step": 7040,
      "time_per_iteration": 3.21728253364563
    },
    {
      "auxiliary_loss_clip": 0.01172796,
      "auxiliary_loss_mlp": 0.01022904,
      "balance_loss_clip": 1.04639864,
      "balance_loss_mlp": 1.01538479,
      "epoch": 0.8466301929898394,
      "flos": 13735724561280.0,
      "grad_norm": 4.6293083517199225,
      "language_loss": 0.63184816,
      "learning_rate": 2.415796215659134e-07,
      "loss": 0.65380514,
      "num_input_tokens_seen": 152239085,
      "step": 7041,
      "time_per_iteration": 2.6728780269622803
    },
    {
      "auxiliary_loss_clip": 0.01177585,
      "auxiliary_loss_mlp": 0.01028886,
      "balance_loss_clip": 1.04077613,
      "balance_loss_mlp": 1.0214746,
      "epoch": 0.8467504358804786,
      "flos": 19241260738560.0,
      "grad_norm": 2.1405383002227176,
      "language_loss": 0.7717638,
      "learning_rate": 2.412086263019939e-07,
      "loss": 0.79382849,
      "num_input_tokens_seen": 152257110,
      "step": 7042,
      "time_per_iteration": 2.665539503097534
    },
    {
      "auxiliary_loss_clip": 0.01162338,
      "auxiliary_loss_mlp": 0.01022619,
      "balance_loss_clip": 1.04731429,
      "balance_loss_mlp": 1.01567507,
      "epoch": 0.8468706787711177,
      "flos": 21324115710720.0,
      "grad_norm": 2.104051270499623,
      "language_loss": 0.80149889,
      "learning_rate": 2.408378978453276e-07,
      "loss": 0.82334846,
      "num_input_tokens_seen": 152277230,
      "step": 7043,
      "time_per_iteration": 2.5627012252807617
    },
    {
      "auxiliary_loss_clip": 0.01070607,
      "auxiliary_loss_mlp": 0.01001315,
      "balance_loss_clip": 1.01082325,
      "balance_loss_mlp": 1.00056398,
      "epoch": 0.8469909216617567,
      "flos": 64877439058560.0,
      "grad_norm": 0.8205361742374072,
      "language_loss": 0.63886297,
      "learning_rate": 2.404674362521533e-07,
      "loss": 0.6595822,
      "num_input_tokens_seen": 152335725,
      "step": 7044,
      "time_per_iteration": 3.145047426223755
    },
    {
      "auxiliary_loss_clip": 0.01169686,
      "auxiliary_loss_mlp": 0.0102092,
      "balance_loss_clip": 1.04694295,
      "balance_loss_mlp": 1.01429212,
      "epoch": 0.8471111645523959,
      "flos": 19280583152640.0,
      "grad_norm": 2.454336171915012,
      "language_loss": 0.74865562,
      "learning_rate": 2.4009724157866997e-07,
      "loss": 0.7705617,
      "num_input_tokens_seen": 152352785,
      "step": 7045,
      "time_per_iteration": 2.736233949661255
    },
    {
      "auxiliary_loss_clip": 0.01164478,
      "auxiliary_loss_mlp": 0.01020978,
      "balance_loss_clip": 1.04680669,
      "balance_loss_mlp": 1.01420403,
      "epoch": 0.8472314074430349,
      "flos": 22015826893440.0,
      "grad_norm": 2.254752312169764,
      "language_loss": 0.76710546,
      "learning_rate": 2.3972731388103564e-07,
      "loss": 0.78895998,
      "num_input_tokens_seen": 152371265,
      "step": 7046,
      "time_per_iteration": 3.5891501903533936
    },
    {
      "auxiliary_loss_clip": 0.01091973,
      "auxiliary_loss_mlp": 0.01001059,
      "balance_loss_clip": 1.011374,
      "balance_loss_mlp": 1.00031984,
      "epoch": 0.847351650333674,
      "flos": 57882580243200.0,
      "grad_norm": 0.8117769560398954,
      "language_loss": 0.62337488,
      "learning_rate": 2.393576532153687e-07,
      "loss": 0.64430523,
      "num_input_tokens_seen": 152435050,
      "step": 7047,
      "time_per_iteration": 3.3054709434509277
    },
    {
      "auxiliary_loss_clip": 0.01068473,
      "auxiliary_loss_mlp": 0.01000123,
      "balance_loss_clip": 1.01069069,
      "balance_loss_mlp": 0.99940223,
      "epoch": 0.8474718932243132,
      "flos": 41284238313600.0,
      "grad_norm": 0.9270951168809975,
      "language_loss": 0.57760388,
      "learning_rate": 2.389882596377453e-07,
      "loss": 0.59828985,
      "num_input_tokens_seen": 152489315,
      "step": 7048,
      "time_per_iteration": 3.1129586696624756
    },
    {
      "auxiliary_loss_clip": 0.01164813,
      "auxiliary_loss_mlp": 0.01022021,
      "balance_loss_clip": 1.04610002,
      "balance_loss_mlp": 1.01467133,
      "epoch": 0.8475921361149522,
      "flos": 38180906974080.0,
      "grad_norm": 1.9578904652215194,
      "language_loss": 0.76683676,
      "learning_rate": 2.386191332042031e-07,
      "loss": 0.78870511,
      "num_input_tokens_seen": 152511210,
      "step": 7049,
      "time_per_iteration": 2.7542054653167725
    },
    {
      "auxiliary_loss_clip": 0.01173361,
      "auxiliary_loss_mlp": 0.01029275,
      "balance_loss_clip": 1.04956543,
      "balance_loss_mlp": 1.02186632,
      "epoch": 0.8477123790055913,
      "flos": 25375054723200.0,
      "grad_norm": 1.7822526012621365,
      "language_loss": 0.73149192,
      "learning_rate": 2.3825027397073794e-07,
      "loss": 0.75351828,
      "num_input_tokens_seen": 152531685,
      "step": 7050,
      "time_per_iteration": 2.6357319355010986
    },
    {
      "auxiliary_loss_clip": 0.01165531,
      "auxiliary_loss_mlp": 0.01022221,
      "balance_loss_clip": 1.04737973,
      "balance_loss_mlp": 1.01483047,
      "epoch": 0.8478326218962304,
      "flos": 30225185389440.0,
      "grad_norm": 14.641663622391691,
      "language_loss": 0.66525835,
      "learning_rate": 2.3788168199330515e-07,
      "loss": 0.68713582,
      "num_input_tokens_seen": 152553245,
      "step": 7051,
      "time_per_iteration": 2.639010429382324
    },
    {
      "auxiliary_loss_clip": 0.01155826,
      "auxiliary_loss_mlp": 0.01023605,
      "balance_loss_clip": 1.03897178,
      "balance_loss_mlp": 1.01575518,
      "epoch": 0.8479528647868695,
      "flos": 38213800853760.0,
      "grad_norm": 1.7717970298994279,
      "language_loss": 0.72772336,
      "learning_rate": 2.3751335732782074e-07,
      "loss": 0.74951762,
      "num_input_tokens_seen": 152574505,
      "step": 7052,
      "time_per_iteration": 2.8625779151916504
    },
    {
      "auxiliary_loss_clip": 0.01167223,
      "auxiliary_loss_mlp": 0.01021872,
      "balance_loss_clip": 1.04722095,
      "balance_loss_mlp": 1.0148567,
      "epoch": 0.8480731076775085,
      "flos": 20957790856320.0,
      "grad_norm": 2.190721886748428,
      "language_loss": 0.79483891,
      "learning_rate": 2.371453000301582e-07,
      "loss": 0.8167299,
      "num_input_tokens_seen": 152593190,
      "step": 7053,
      "time_per_iteration": 3.505032539367676
    },
    {
      "auxiliary_loss_clip": 0.0116858,
      "auxiliary_loss_mlp": 0.01022184,
      "balance_loss_clip": 1.04222214,
      "balance_loss_mlp": 1.01522517,
      "epoch": 0.8481933505681477,
      "flos": 32596510487040.0,
      "grad_norm": 4.693075128526909,
      "language_loss": 0.74381375,
      "learning_rate": 2.3677751015615222e-07,
      "loss": 0.76572144,
      "num_input_tokens_seen": 152615265,
      "step": 7054,
      "time_per_iteration": 2.7425079345703125
    },
    {
      "auxiliary_loss_clip": 0.01160893,
      "auxiliary_loss_mlp": 0.01029482,
      "balance_loss_clip": 1.04234219,
      "balance_loss_mlp": 1.02198386,
      "epoch": 0.8483135934587868,
      "flos": 20741177888640.0,
      "grad_norm": 2.3212145058970353,
      "language_loss": 0.85465109,
      "learning_rate": 2.3640998776159593e-07,
      "loss": 0.87655485,
      "num_input_tokens_seen": 152632770,
      "step": 7055,
      "time_per_iteration": 3.553556203842163
    },
    {
      "auxiliary_loss_clip": 0.01173761,
      "auxiliary_loss_mlp": 0.01020585,
      "balance_loss_clip": 1.04711127,
      "balance_loss_mlp": 1.01423967,
      "epoch": 0.8484338363494258,
      "flos": 21653057485440.0,
      "grad_norm": 2.0781034403196594,
      "language_loss": 0.8136037,
      "learning_rate": 2.3604273290224253e-07,
      "loss": 0.83554721,
      "num_input_tokens_seen": 152653485,
      "step": 7056,
      "time_per_iteration": 2.6156160831451416
    },
    {
      "auxiliary_loss_clip": 0.01174979,
      "auxiliary_loss_mlp": 0.01027586,
      "balance_loss_clip": 1.04660773,
      "balance_loss_mlp": 1.01943767,
      "epoch": 0.848554079240065,
      "flos": 15013964926080.0,
      "grad_norm": 3.4871526966435717,
      "language_loss": 0.7495786,
      "learning_rate": 2.356757456338039e-07,
      "loss": 0.77160424,
      "num_input_tokens_seen": 152670970,
      "step": 7057,
      "time_per_iteration": 2.6450376510620117
    },
    {
      "auxiliary_loss_clip": 0.01075534,
      "auxiliary_loss_mlp": 0.01001096,
      "balance_loss_clip": 1.01507008,
      "balance_loss_mlp": 1.00026166,
      "epoch": 0.848674322130704,
      "flos": 68060453742720.0,
      "grad_norm": 0.7594565759645941,
      "language_loss": 0.58984005,
      "learning_rate": 2.3530902601195147e-07,
      "loss": 0.61060631,
      "num_input_tokens_seen": 152739460,
      "step": 7058,
      "time_per_iteration": 3.249131202697754
    },
    {
      "auxiliary_loss_clip": 0.01166309,
      "auxiliary_loss_mlp": 0.01027912,
      "balance_loss_clip": 1.04444242,
      "balance_loss_mlp": 1.02023482,
      "epoch": 0.8487945650213431,
      "flos": 18475788977280.0,
      "grad_norm": 2.2533896694264923,
      "language_loss": 0.79276383,
      "learning_rate": 2.34942574092317e-07,
      "loss": 0.81470609,
      "num_input_tokens_seen": 152754710,
      "step": 7059,
      "time_per_iteration": 3.5280730724334717
    },
    {
      "auxiliary_loss_clip": 0.01173463,
      "auxiliary_loss_mlp": 0.01027224,
      "balance_loss_clip": 1.04673767,
      "balance_loss_mlp": 1.01940966,
      "epoch": 0.8489148079119821,
      "flos": 23473189405440.0,
      "grad_norm": 1.8966845688731222,
      "language_loss": 0.77164149,
      "learning_rate": 2.3457638993049045e-07,
      "loss": 0.79364836,
      "num_input_tokens_seen": 152772700,
      "step": 7060,
      "time_per_iteration": 2.702631950378418
    },
    {
      "auxiliary_loss_clip": 0.01186398,
      "auxiliary_loss_mlp": 0.01023449,
      "balance_loss_clip": 1.04522562,
      "balance_loss_mlp": 1.01578951,
      "epoch": 0.8490350508026213,
      "flos": 19937604775680.0,
      "grad_norm": 1.9907335975062321,
      "language_loss": 0.64697939,
      "learning_rate": 2.3421047358202252e-07,
      "loss": 0.66907787,
      "num_input_tokens_seen": 152791550,
      "step": 7061,
      "time_per_iteration": 2.7087883949279785
    },
    {
      "auxiliary_loss_clip": 0.01171627,
      "auxiliary_loss_mlp": 0.01025199,
      "balance_loss_clip": 1.0477016,
      "balance_loss_mlp": 1.01774824,
      "epoch": 0.8491552936932604,
      "flos": 24279958828800.0,
      "grad_norm": 2.847888298610926,
      "language_loss": 0.83577484,
      "learning_rate": 2.3384482510242144e-07,
      "loss": 0.85774308,
      "num_input_tokens_seen": 152809410,
      "step": 7062,
      "time_per_iteration": 2.717771530151367
    },
    {
      "auxiliary_loss_clip": 0.01168718,
      "auxiliary_loss_mlp": 0.01026132,
      "balance_loss_clip": 1.0468514,
      "balance_loss_mlp": 1.01827586,
      "epoch": 0.8492755365838994,
      "flos": 22522526098560.0,
      "grad_norm": 3.7976816269559617,
      "language_loss": 0.77046031,
      "learning_rate": 2.3347944454715575e-07,
      "loss": 0.79240876,
      "num_input_tokens_seen": 152825800,
      "step": 7063,
      "time_per_iteration": 2.668525457382202
    },
    {
      "auxiliary_loss_clip": 0.01170673,
      "auxiliary_loss_mlp": 0.01024213,
      "balance_loss_clip": 1.04799902,
      "balance_loss_mlp": 1.01643729,
      "epoch": 0.8493957794745386,
      "flos": 26980441182720.0,
      "grad_norm": 2.384544612464124,
      "language_loss": 0.67599237,
      "learning_rate": 2.331143319716542e-07,
      "loss": 0.69794118,
      "num_input_tokens_seen": 152845330,
      "step": 7064,
      "time_per_iteration": 2.670663356781006
    },
    {
      "auxiliary_loss_clip": 0.01176796,
      "auxiliary_loss_mlp": 0.01023534,
      "balance_loss_clip": 1.04214239,
      "balance_loss_mlp": 1.01599979,
      "epoch": 0.8495160223651776,
      "flos": 29861985018240.0,
      "grad_norm": 2.5465376011139833,
      "language_loss": 0.65918308,
      "learning_rate": 2.3274948743130363e-07,
      "loss": 0.68118638,
      "num_input_tokens_seen": 152865165,
      "step": 7065,
      "time_per_iteration": 2.756451368331909
    },
    {
      "auxiliary_loss_clip": 0.01166454,
      "auxiliary_loss_mlp": 0.01023677,
      "balance_loss_clip": 1.04564834,
      "balance_loss_mlp": 1.01574326,
      "epoch": 0.8496362652558167,
      "flos": 23075443128960.0,
      "grad_norm": 2.282244283567434,
      "language_loss": 0.79227889,
      "learning_rate": 2.3238491098145085e-07,
      "loss": 0.81418025,
      "num_input_tokens_seen": 152884695,
      "step": 7066,
      "time_per_iteration": 2.7060062885284424
    },
    {
      "auxiliary_loss_clip": 0.01168165,
      "auxiliary_loss_mlp": 0.01021635,
      "balance_loss_clip": 1.04594636,
      "balance_loss_mlp": 1.01421118,
      "epoch": 0.8497565081464559,
      "flos": 14609107756800.0,
      "grad_norm": 2.2219123312225064,
      "language_loss": 0.73583126,
      "learning_rate": 2.3202060267740141e-07,
      "loss": 0.75772923,
      "num_input_tokens_seen": 152902220,
      "step": 7067,
      "time_per_iteration": 2.5310239791870117
    },
    {
      "auxiliary_loss_clip": 0.0117004,
      "auxiliary_loss_mlp": 0.01018963,
      "balance_loss_clip": 1.03750515,
      "balance_loss_mlp": 1.01188779,
      "epoch": 0.8498767510370949,
      "flos": 21136446126720.0,
      "grad_norm": 3.0089127942446177,
      "language_loss": 0.77323961,
      "learning_rate": 2.3165656257442044e-07,
      "loss": 0.7951296,
      "num_input_tokens_seen": 152920740,
      "step": 7068,
      "time_per_iteration": 2.687053680419922
    },
    {
      "auxiliary_loss_clip": 0.01165447,
      "auxiliary_loss_mlp": 0.01022097,
      "balance_loss_clip": 1.04547405,
      "balance_loss_mlp": 1.0152638,
      "epoch": 0.849996993927734,
      "flos": 23654538195840.0,
      "grad_norm": 1.9562575357357384,
      "language_loss": 0.90605807,
      "learning_rate": 2.31292790727734e-07,
      "loss": 0.92793345,
      "num_input_tokens_seen": 152938305,
      "step": 7069,
      "time_per_iteration": 2.6127078533172607
    },
    {
      "auxiliary_loss_clip": 0.01163305,
      "auxiliary_loss_mlp": 0.01023103,
      "balance_loss_clip": 1.04531479,
      "balance_loss_mlp": 1.01630855,
      "epoch": 0.8501172368183731,
      "flos": 20558069331840.0,
      "grad_norm": 2.268383923330982,
      "language_loss": 0.80161548,
      "learning_rate": 2.3092928719252392e-07,
      "loss": 0.82347953,
      "num_input_tokens_seen": 152956705,
      "step": 7070,
      "time_per_iteration": 2.6623480319976807
    },
    {
      "auxiliary_loss_clip": 0.01166362,
      "auxiliary_loss_mlp": 0.01024816,
      "balance_loss_clip": 1.04442561,
      "balance_loss_mlp": 1.01729357,
      "epoch": 0.8502374797090122,
      "flos": 22272624201600.0,
      "grad_norm": 25.506108253317638,
      "language_loss": 0.78507572,
      "learning_rate": 2.3056605202393475e-07,
      "loss": 0.80698752,
      "num_input_tokens_seen": 152974265,
      "step": 7071,
      "time_per_iteration": 2.602137565612793
    },
    {
      "auxiliary_loss_clip": 0.01164326,
      "auxiliary_loss_mlp": 0.01208132,
      "balance_loss_clip": 1.0426259,
      "balance_loss_mlp": 1.00067353,
      "epoch": 0.8503577225996513,
      "flos": 23659817495040.0,
      "grad_norm": 1.7527717280717618,
      "language_loss": 0.66919881,
      "learning_rate": 2.3020308527706888e-07,
      "loss": 0.69292343,
      "num_input_tokens_seen": 152993680,
      "step": 7072,
      "time_per_iteration": 2.6263720989227295
    },
    {
      "auxiliary_loss_clip": 0.01177522,
      "auxiliary_loss_mlp": 0.0102717,
      "balance_loss_clip": 1.04377985,
      "balance_loss_mlp": 1.01964498,
      "epoch": 0.8504779654902904,
      "flos": 26758513002240.0,
      "grad_norm": 1.620052821847607,
      "language_loss": 0.89035285,
      "learning_rate": 2.2984038700698715e-07,
      "loss": 0.91239977,
      "num_input_tokens_seen": 153012990,
      "step": 7073,
      "time_per_iteration": 3.5903780460357666
    },
    {
      "auxiliary_loss_clip": 0.0116604,
      "auxiliary_loss_mlp": 0.01028135,
      "balance_loss_clip": 1.04697883,
      "balance_loss_mlp": 1.02055955,
      "epoch": 0.8505982083809295,
      "flos": 26468247196800.0,
      "grad_norm": 1.6387015377089864,
      "language_loss": 0.79341727,
      "learning_rate": 2.2947795726871222e-07,
      "loss": 0.815359,
      "num_input_tokens_seen": 153034015,
      "step": 7074,
      "time_per_iteration": 2.5695621967315674
    },
    {
      "auxiliary_loss_clip": 0.01167261,
      "auxiliary_loss_mlp": 0.01208185,
      "balance_loss_clip": 1.04876208,
      "balance_loss_mlp": 1.00090384,
      "epoch": 0.8507184512715685,
      "flos": 20303390926080.0,
      "grad_norm": 1.6895561174436262,
      "language_loss": 0.85676348,
      "learning_rate": 2.2911579611722253e-07,
      "loss": 0.8805179,
      "num_input_tokens_seen": 153053160,
      "step": 7075,
      "time_per_iteration": 2.6135733127593994
    },
    {
      "auxiliary_loss_clip": 0.01169666,
      "auxiliary_loss_mlp": 0.01027173,
      "balance_loss_clip": 1.04393506,
      "balance_loss_mlp": 1.01964164,
      "epoch": 0.8508386941622077,
      "flos": 19025186474880.0,
      "grad_norm": 1.6783382988443536,
      "language_loss": 0.87372184,
      "learning_rate": 2.2875390360745905e-07,
      "loss": 0.8956902,
      "num_input_tokens_seen": 153072565,
      "step": 7076,
      "time_per_iteration": 2.6456897258758545
    },
    {
      "auxiliary_loss_clip": 0.01178901,
      "auxiliary_loss_mlp": 0.01027512,
      "balance_loss_clip": 1.04244828,
      "balance_loss_mlp": 1.02013874,
      "epoch": 0.8509589370528468,
      "flos": 16433405654400.0,
      "grad_norm": 1.7639588255546996,
      "language_loss": 0.77642155,
      "learning_rate": 2.2839227979432008e-07,
      "loss": 0.7984857,
      "num_input_tokens_seen": 153090215,
      "step": 7077,
      "time_per_iteration": 2.6855008602142334
    },
    {
      "auxiliary_loss_clip": 0.01174037,
      "auxiliary_loss_mlp": 0.01027188,
      "balance_loss_clip": 1.04440236,
      "balance_loss_mlp": 1.01971054,
      "epoch": 0.8510791799434858,
      "flos": 18259714713600.0,
      "grad_norm": 1.839386576311624,
      "language_loss": 0.85268271,
      "learning_rate": 2.2803092473266373e-07,
      "loss": 0.87469494,
      "num_input_tokens_seen": 153107740,
      "step": 7078,
      "time_per_iteration": 2.556206464767456
    },
    {
      "auxiliary_loss_clip": 0.01169327,
      "auxiliary_loss_mlp": 0.01026839,
      "balance_loss_clip": 1.04933417,
      "balance_loss_mlp": 1.01983213,
      "epoch": 0.851199422834125,
      "flos": 23441372933760.0,
      "grad_norm": 2.719130599543163,
      "language_loss": 0.86782205,
      "learning_rate": 2.2766983847730724e-07,
      "loss": 0.88978374,
      "num_input_tokens_seen": 153127410,
      "step": 7079,
      "time_per_iteration": 2.697603940963745
    },
    {
      "auxiliary_loss_clip": 0.01182303,
      "auxiliary_loss_mlp": 0.01025862,
      "balance_loss_clip": 1.04219759,
      "balance_loss_mlp": 1.01801157,
      "epoch": 0.851319665724764,
      "flos": 16289404030080.0,
      "grad_norm": 2.6408086900918653,
      "language_loss": 0.67062247,
      "learning_rate": 2.2730902108302663e-07,
      "loss": 0.69270408,
      "num_input_tokens_seen": 153144325,
      "step": 7080,
      "time_per_iteration": 3.587029218673706
    },
    {
      "auxiliary_loss_clip": 0.0116443,
      "auxiliary_loss_mlp": 0.0102483,
      "balance_loss_clip": 1.04135394,
      "balance_loss_mlp": 1.01702797,
      "epoch": 0.8514399086154031,
      "flos": 18989347680000.0,
      "grad_norm": 1.7039534551300322,
      "language_loss": 0.68711609,
      "learning_rate": 2.269484726045583e-07,
      "loss": 0.70900869,
      "num_input_tokens_seen": 153163240,
      "step": 7081,
      "time_per_iteration": 2.599762439727783
    },
    {
      "auxiliary_loss_clip": 0.01177943,
      "auxiliary_loss_mlp": 0.01029726,
      "balance_loss_clip": 1.04393125,
      "balance_loss_mlp": 1.02213264,
      "epoch": 0.8515601515060423,
      "flos": 24571194301440.0,
      "grad_norm": 1.947864106415915,
      "language_loss": 0.78816092,
      "learning_rate": 2.2658819309659672e-07,
      "loss": 0.81023765,
      "num_input_tokens_seen": 153183440,
      "step": 7082,
      "time_per_iteration": 3.6081581115722656
    },
    {
      "auxiliary_loss_clip": 0.01168184,
      "auxiliary_loss_mlp": 0.01021673,
      "balance_loss_clip": 1.04565227,
      "balance_loss_mlp": 1.01507473,
      "epoch": 0.8516803943966813,
      "flos": 19529443555200.0,
      "grad_norm": 2.0549957710975666,
      "language_loss": 0.84988725,
      "learning_rate": 2.2622818261379706e-07,
      "loss": 0.87178588,
      "num_input_tokens_seen": 153200460,
      "step": 7083,
      "time_per_iteration": 2.606545925140381
    },
    {
      "auxiliary_loss_clip": 0.01169157,
      "auxiliary_loss_mlp": 0.01025089,
      "balance_loss_clip": 1.04324317,
      "balance_loss_mlp": 1.01748955,
      "epoch": 0.8518006372873204,
      "flos": 20265792364800.0,
      "grad_norm": 1.7482106190243287,
      "language_loss": 0.74929893,
      "learning_rate": 2.2586844121077142e-07,
      "loss": 0.77124143,
      "num_input_tokens_seen": 153218970,
      "step": 7084,
      "time_per_iteration": 2.6358532905578613
    },
    {
      "auxiliary_loss_clip": 0.01177561,
      "auxiliary_loss_mlp": 0.0103077,
      "balance_loss_clip": 1.03966868,
      "balance_loss_mlp": 1.02273512,
      "epoch": 0.8519208801779595,
      "flos": 24133227770880.0,
      "grad_norm": 1.794091305307738,
      "language_loss": 0.72159028,
      "learning_rate": 2.2550896894209215e-07,
      "loss": 0.74367356,
      "num_input_tokens_seen": 153238485,
      "step": 7085,
      "time_per_iteration": 2.654695510864258
    },
    {
      "auxiliary_loss_clip": 0.01096612,
      "auxiliary_loss_mlp": 0.01001677,
      "balance_loss_clip": 1.01133394,
      "balance_loss_mlp": 1.00091684,
      "epoch": 0.8520411230685986,
      "flos": 63035223252480.0,
      "grad_norm": 0.6775033064717096,
      "language_loss": 0.56580734,
      "learning_rate": 2.2514976586229184e-07,
      "loss": 0.5867902,
      "num_input_tokens_seen": 153306430,
      "step": 7086,
      "time_per_iteration": 4.14595627784729
    },
    {
      "auxiliary_loss_clip": 0.01071847,
      "auxiliary_loss_mlp": 0.01002069,
      "balance_loss_clip": 1.01214874,
      "balance_loss_mlp": 1.0013268,
      "epoch": 0.8521613659592376,
      "flos": 65836865283840.0,
      "grad_norm": 0.7672851066296433,
      "language_loss": 0.54635,
      "learning_rate": 2.247908320258609e-07,
      "loss": 0.5670892,
      "num_input_tokens_seen": 153366520,
      "step": 7087,
      "time_per_iteration": 3.168478488922119
    },
    {
      "auxiliary_loss_clip": 0.01171281,
      "auxiliary_loss_mlp": 0.01024863,
      "balance_loss_clip": 1.04211247,
      "balance_loss_mlp": 1.01694751,
      "epoch": 0.8522816088498768,
      "flos": 23112323418240.0,
      "grad_norm": 1.9528891456814474,
      "language_loss": 0.79955477,
      "learning_rate": 2.2443216748724914e-07,
      "loss": 0.82151622,
      "num_input_tokens_seen": 153387230,
      "step": 7088,
      "time_per_iteration": 2.6556930541992188
    },
    {
      "auxiliary_loss_clip": 0.0117368,
      "auxiliary_loss_mlp": 0.01207615,
      "balance_loss_clip": 1.04754424,
      "balance_loss_mlp": 1.0005846,
      "epoch": 0.8524018517405159,
      "flos": 31758140073600.0,
      "grad_norm": 1.9436887970044046,
      "language_loss": 0.74634647,
      "learning_rate": 2.2407377230086588e-07,
      "loss": 0.77015942,
      "num_input_tokens_seen": 153409585,
      "step": 7089,
      "time_per_iteration": 2.780372142791748
    },
    {
      "auxiliary_loss_clip": 0.0117067,
      "auxiliary_loss_mlp": 0.01019244,
      "balance_loss_clip": 1.04329467,
      "balance_loss_mlp": 1.01206493,
      "epoch": 0.8525220946311549,
      "flos": 18690318956160.0,
      "grad_norm": 2.295359907842936,
      "language_loss": 0.83273965,
      "learning_rate": 2.23715646521079e-07,
      "loss": 0.85463881,
      "num_input_tokens_seen": 153427105,
      "step": 7090,
      "time_per_iteration": 2.64875864982605
    },
    {
      "auxiliary_loss_clip": 0.01173588,
      "auxiliary_loss_mlp": 0.01207868,
      "balance_loss_clip": 1.04590058,
      "balance_loss_mlp": 1.00066662,
      "epoch": 0.852642337521794,
      "flos": 21793216354560.0,
      "grad_norm": 2.2360524636427237,
      "language_loss": 0.84221256,
      "learning_rate": 2.2335779020221724e-07,
      "loss": 0.86602712,
      "num_input_tokens_seen": 153443725,
      "step": 7091,
      "time_per_iteration": 2.559251070022583
    },
    {
      "auxiliary_loss_clip": 0.01065172,
      "auxiliary_loss_mlp": 0.00998619,
      "balance_loss_clip": 1.0129621,
      "balance_loss_mlp": 0.99780291,
      "epoch": 0.8527625804124331,
      "flos": 69040132260480.0,
      "grad_norm": 0.7998569106408824,
      "language_loss": 0.56434143,
      "learning_rate": 2.2300020339856497e-07,
      "loss": 0.5849793,
      "num_input_tokens_seen": 153506410,
      "step": 7092,
      "time_per_iteration": 3.17834734916687
    },
    {
      "auxiliary_loss_clip": 0.01166619,
      "auxiliary_loss_mlp": 0.0102379,
      "balance_loss_clip": 1.04247248,
      "balance_loss_mlp": 1.01638687,
      "epoch": 0.8528828233030722,
      "flos": 26979399688320.0,
      "grad_norm": 2.4383046310969694,
      "language_loss": 0.77917272,
      "learning_rate": 2.2264288616436966e-07,
      "loss": 0.80107683,
      "num_input_tokens_seen": 153526665,
      "step": 7093,
      "time_per_iteration": 2.6571123600006104
    },
    {
      "auxiliary_loss_clip": 0.0116491,
      "auxiliary_loss_mlp": 0.01024128,
      "balance_loss_clip": 1.04289019,
      "balance_loss_mlp": 1.01701093,
      "epoch": 0.8530030661937112,
      "flos": 17487598936320.0,
      "grad_norm": 1.9382136861138883,
      "language_loss": 0.72544968,
      "learning_rate": 2.222858385538351e-07,
      "loss": 0.74734008,
      "num_input_tokens_seen": 153543465,
      "step": 7094,
      "time_per_iteration": 2.647331476211548
    },
    {
      "auxiliary_loss_clip": 0.01164968,
      "auxiliary_loss_mlp": 0.01026635,
      "balance_loss_clip": 1.04414475,
      "balance_loss_mlp": 1.01915765,
      "epoch": 0.8531233090843504,
      "flos": 22160798184960.0,
      "grad_norm": 1.7705151941818198,
      "language_loss": 0.68234599,
      "learning_rate": 2.2192906062112527e-07,
      "loss": 0.70426202,
      "num_input_tokens_seen": 153563340,
      "step": 7095,
      "time_per_iteration": 2.616774797439575
    },
    {
      "auxiliary_loss_clip": 0.01165467,
      "auxiliary_loss_mlp": 0.01023529,
      "balance_loss_clip": 1.0456903,
      "balance_loss_mlp": 1.01663291,
      "epoch": 0.8532435519749895,
      "flos": 37635388145280.0,
      "grad_norm": 1.7063511845284187,
      "language_loss": 0.70869589,
      "learning_rate": 2.2157255242036377e-07,
      "loss": 0.73058581,
      "num_input_tokens_seen": 153587005,
      "step": 7096,
      "time_per_iteration": 2.750498056411743
    },
    {
      "auxiliary_loss_clip": 0.01171698,
      "auxiliary_loss_mlp": 0.01027773,
      "balance_loss_clip": 1.04215026,
      "balance_loss_mlp": 1.02041149,
      "epoch": 0.8533637948656285,
      "flos": 21398163598080.0,
      "grad_norm": 1.5755984792279232,
      "language_loss": 0.74458253,
      "learning_rate": 2.2121631400563135e-07,
      "loss": 0.76657724,
      "num_input_tokens_seen": 153606835,
      "step": 7097,
      "time_per_iteration": 2.6330225467681885
    },
    {
      "auxiliary_loss_clip": 0.01067132,
      "auxiliary_loss_mlp": 0.01000276,
      "balance_loss_clip": 1.0133481,
      "balance_loss_mlp": 0.99944782,
      "epoch": 0.8534840377562677,
      "flos": 53345122490880.0,
      "grad_norm": 0.7597013907881722,
      "language_loss": 0.52936399,
      "learning_rate": 2.208603454309701e-07,
      "loss": 0.5500381,
      "num_input_tokens_seen": 153664925,
      "step": 7098,
      "time_per_iteration": 3.1400206089019775
    },
    {
      "auxiliary_loss_clip": 0.01177421,
      "auxiliary_loss_mlp": 0.01024903,
      "balance_loss_clip": 1.04189992,
      "balance_loss_mlp": 1.0167197,
      "epoch": 0.8536042806469067,
      "flos": 20814148368000.0,
      "grad_norm": 1.836949023653484,
      "language_loss": 0.71114361,
      "learning_rate": 2.2050464675037994e-07,
      "loss": 0.73316681,
      "num_input_tokens_seen": 153683550,
      "step": 7099,
      "time_per_iteration": 3.620943307876587
    },
    {
      "auxiliary_loss_clip": 0.0117274,
      "auxiliary_loss_mlp": 0.01026471,
      "balance_loss_clip": 1.0461663,
      "balance_loss_mlp": 1.0187583,
      "epoch": 0.8537245235375458,
      "flos": 24681368292480.0,
      "grad_norm": 2.291568835978473,
      "language_loss": 0.73056412,
      "learning_rate": 2.2014921801782016e-07,
      "loss": 0.75255626,
      "num_input_tokens_seen": 153703040,
      "step": 7100,
      "time_per_iteration": 2.699664831161499
    },
    {
      "auxiliary_loss_clip": 0.01171805,
      "auxiliary_loss_mlp": 0.0102316,
      "balance_loss_clip": 1.04054928,
      "balance_loss_mlp": 1.01592445,
      "epoch": 0.853844766428185,
      "flos": 24384817607040.0,
      "grad_norm": 1.9509058450210004,
      "language_loss": 0.73920238,
      "learning_rate": 2.1979405928720872e-07,
      "loss": 0.76115203,
      "num_input_tokens_seen": 153722695,
      "step": 7101,
      "time_per_iteration": 2.745309829711914
    },
    {
      "auxiliary_loss_clip": 0.01173227,
      "auxiliary_loss_mlp": 0.01023399,
      "balance_loss_clip": 1.04397404,
      "balance_loss_mlp": 1.01635361,
      "epoch": 0.853965009318824,
      "flos": 20955707867520.0,
      "grad_norm": 1.6484930627328545,
      "language_loss": 0.79320771,
      "learning_rate": 2.1943917061242257e-07,
      "loss": 0.81517398,
      "num_input_tokens_seen": 153742550,
      "step": 7102,
      "time_per_iteration": 2.6544923782348633
    },
    {
      "auxiliary_loss_clip": 0.01178811,
      "auxiliary_loss_mlp": 0.01207698,
      "balance_loss_clip": 1.04810667,
      "balance_loss_mlp": 1.00069547,
      "epoch": 0.8540852522094631,
      "flos": 24201816791040.0,
      "grad_norm": 2.0755516907003733,
      "language_loss": 0.66447318,
      "learning_rate": 2.1908455204729903e-07,
      "loss": 0.68833828,
      "num_input_tokens_seen": 153761700,
      "step": 7103,
      "time_per_iteration": 2.7232251167297363
    },
    {
      "auxiliary_loss_clip": 0.01172613,
      "auxiliary_loss_mlp": 0.01026555,
      "balance_loss_clip": 1.04361606,
      "balance_loss_mlp": 1.01900935,
      "epoch": 0.8542054951001022,
      "flos": 25082921410560.0,
      "grad_norm": 23.42643570586118,
      "language_loss": 0.78461015,
      "learning_rate": 2.1873020364563265e-07,
      "loss": 0.80660188,
      "num_input_tokens_seen": 153780765,
      "step": 7104,
      "time_per_iteration": 2.723378896713257
    },
    {
      "auxiliary_loss_clip": 0.01165712,
      "auxiliary_loss_mlp": 0.01024365,
      "balance_loss_clip": 1.04620075,
      "balance_loss_mlp": 1.01697147,
      "epoch": 0.8543257379907413,
      "flos": 24316551809280.0,
      "grad_norm": 2.9039353148556177,
      "language_loss": 0.76069069,
      "learning_rate": 2.183761254611789e-07,
      "loss": 0.78259146,
      "num_input_tokens_seen": 153801090,
      "step": 7105,
      "time_per_iteration": 2.6045217514038086
    },
    {
      "auxiliary_loss_clip": 0.01168739,
      "auxiliary_loss_mlp": 0.01025191,
      "balance_loss_clip": 1.04658461,
      "balance_loss_mlp": 1.01787746,
      "epoch": 0.8544459808813804,
      "flos": 55286630467200.0,
      "grad_norm": 2.146254097438898,
      "language_loss": 0.70838273,
      "learning_rate": 2.1802231754764987e-07,
      "loss": 0.73032212,
      "num_input_tokens_seen": 153826530,
      "step": 7106,
      "time_per_iteration": 2.869489908218384
    },
    {
      "auxiliary_loss_clip": 0.01172379,
      "auxiliary_loss_mlp": 0.0102522,
      "balance_loss_clip": 1.04093266,
      "balance_loss_mlp": 1.01744485,
      "epoch": 0.8545662237720195,
      "flos": 25776248705280.0,
      "grad_norm": 1.9325634694428067,
      "language_loss": 0.76568711,
      "learning_rate": 2.17668779958718e-07,
      "loss": 0.7876631,
      "num_input_tokens_seen": 153849110,
      "step": 7107,
      "time_per_iteration": 3.5751707553863525
    },
    {
      "auxiliary_loss_clip": 0.0116726,
      "auxiliary_loss_mlp": 0.01024967,
      "balance_loss_clip": 1.0472362,
      "balance_loss_mlp": 1.01745677,
      "epoch": 0.8546864666626586,
      "flos": 11108320427520.0,
      "grad_norm": 2.3050381371324695,
      "language_loss": 0.80537838,
      "learning_rate": 2.1731551274801553e-07,
      "loss": 0.82730067,
      "num_input_tokens_seen": 153865550,
      "step": 7108,
      "time_per_iteration": 2.6164209842681885
    },
    {
      "auxiliary_loss_clip": 0.01172527,
      "auxiliary_loss_mlp": 0.01024727,
      "balance_loss_clip": 1.04544544,
      "balance_loss_mlp": 1.01721668,
      "epoch": 0.8548067095532976,
      "flos": 25520169669120.0,
      "grad_norm": 2.188595840658495,
      "language_loss": 0.61793137,
      "learning_rate": 2.169625159691324e-07,
      "loss": 0.63990396,
      "num_input_tokens_seen": 153885425,
      "step": 7109,
      "time_per_iteration": 3.5273277759552
    },
    {
      "auxiliary_loss_clip": 0.01187197,
      "auxiliary_loss_mlp": 0.0102314,
      "balance_loss_clip": 1.04205859,
      "balance_loss_mlp": 1.01573682,
      "epoch": 0.8549269524439368,
      "flos": 24717853532160.0,
      "grad_norm": 6.024787400497944,
      "language_loss": 0.74256015,
      "learning_rate": 2.1660978967561784e-07,
      "loss": 0.76466352,
      "num_input_tokens_seen": 153904760,
      "step": 7110,
      "time_per_iteration": 2.8049607276916504
    },
    {
      "auxiliary_loss_clip": 0.01164826,
      "auxiliary_loss_mlp": 0.01020286,
      "balance_loss_clip": 1.04478431,
      "balance_loss_mlp": 1.01311898,
      "epoch": 0.8550471953345758,
      "flos": 19825599191040.0,
      "grad_norm": 2.838511183306623,
      "language_loss": 0.79323113,
      "learning_rate": 2.1625733392098035e-07,
      "loss": 0.81508231,
      "num_input_tokens_seen": 153920370,
      "step": 7111,
      "time_per_iteration": 2.6245365142822266
    },
    {
      "auxiliary_loss_clip": 0.01166419,
      "auxiliary_loss_mlp": 0.01023326,
      "balance_loss_clip": 1.04651523,
      "balance_loss_mlp": 1.01602411,
      "epoch": 0.8551674382252149,
      "flos": 22820441500800.0,
      "grad_norm": 2.637061013840521,
      "language_loss": 0.79406452,
      "learning_rate": 2.159051487586867e-07,
      "loss": 0.81596196,
      "num_input_tokens_seen": 153940500,
      "step": 7112,
      "time_per_iteration": 2.6482884883880615
    },
    {
      "auxiliary_loss_clip": 0.01176919,
      "auxiliary_loss_mlp": 0.01031952,
      "balance_loss_clip": 1.04782796,
      "balance_loss_mlp": 1.02374458,
      "epoch": 0.8552876811158541,
      "flos": 20631255292800.0,
      "grad_norm": 2.7461425663533214,
      "language_loss": 0.72515303,
      "learning_rate": 2.155532342421642e-07,
      "loss": 0.74724174,
      "num_input_tokens_seen": 153958500,
      "step": 7113,
      "time_per_iteration": 3.560548782348633
    },
    {
      "auxiliary_loss_clip": 0.01172835,
      "auxiliary_loss_mlp": 0.01027797,
      "balance_loss_clip": 1.04628515,
      "balance_loss_mlp": 1.02006364,
      "epoch": 0.8554079240064931,
      "flos": 23112359331840.0,
      "grad_norm": 2.0342660267500885,
      "language_loss": 0.78120279,
      "learning_rate": 2.1520159042479636e-07,
      "loss": 0.80320907,
      "num_input_tokens_seen": 153976790,
      "step": 7114,
      "time_per_iteration": 2.581678867340088
    },
    {
      "auxiliary_loss_clip": 0.01169018,
      "auxiliary_loss_mlp": 0.01026758,
      "balance_loss_clip": 1.04719567,
      "balance_loss_mlp": 1.0193131,
      "epoch": 0.8555281668971322,
      "flos": 22128047959680.0,
      "grad_norm": 2.1909414110989625,
      "language_loss": 0.70898855,
      "learning_rate": 2.148502173599287e-07,
      "loss": 0.7309463,
      "num_input_tokens_seen": 153994930,
      "step": 7115,
      "time_per_iteration": 2.643247604370117
    },
    {
      "auxiliary_loss_clip": 0.01166211,
      "auxiliary_loss_mlp": 0.01020266,
      "balance_loss_clip": 1.04340339,
      "balance_loss_mlp": 1.01217151,
      "epoch": 0.8556484097877713,
      "flos": 31139040234240.0,
      "grad_norm": 4.868790605902118,
      "language_loss": 0.65977782,
      "learning_rate": 2.1449911510086372e-07,
      "loss": 0.68164253,
      "num_input_tokens_seen": 154014400,
      "step": 7116,
      "time_per_iteration": 2.655488967895508
    },
    {
      "auxiliary_loss_clip": 0.01167144,
      "auxiliary_loss_mlp": 0.01024514,
      "balance_loss_clip": 1.0461247,
      "balance_loss_mlp": 1.0170877,
      "epoch": 0.8557686526784104,
      "flos": 24316551809280.0,
      "grad_norm": 1.9933592138435925,
      "language_loss": 0.76643521,
      "learning_rate": 2.141482837008628e-07,
      "loss": 0.78835177,
      "num_input_tokens_seen": 154034940,
      "step": 7117,
      "time_per_iteration": 2.6247074604034424
    },
    {
      "auxiliary_loss_clip": 0.011606,
      "auxiliary_loss_mlp": 0.01027641,
      "balance_loss_clip": 1.04343235,
      "balance_loss_mlp": 1.01973724,
      "epoch": 0.8558888955690495,
      "flos": 17712723427200.0,
      "grad_norm": 5.7448687727001415,
      "language_loss": 0.71903378,
      "learning_rate": 2.1379772321314826e-07,
      "loss": 0.74091613,
      "num_input_tokens_seen": 154052985,
      "step": 7118,
      "time_per_iteration": 2.538745880126953
    },
    {
      "auxiliary_loss_clip": 0.01171809,
      "auxiliary_loss_mlp": 0.01027928,
      "balance_loss_clip": 1.04078686,
      "balance_loss_mlp": 1.01994133,
      "epoch": 0.8560091384596886,
      "flos": 19171702051200.0,
      "grad_norm": 2.032826578815368,
      "language_loss": 0.81896418,
      "learning_rate": 2.1344743369089802e-07,
      "loss": 0.84096158,
      "num_input_tokens_seen": 154068765,
      "step": 7119,
      "time_per_iteration": 2.648212432861328
    },
    {
      "auxiliary_loss_clip": 0.01168938,
      "auxiliary_loss_mlp": 0.01021235,
      "balance_loss_clip": 1.04416895,
      "balance_loss_mlp": 1.01408243,
      "epoch": 0.8561293813503277,
      "flos": 23914855036800.0,
      "grad_norm": 1.7027953246546494,
      "language_loss": 0.82025009,
      "learning_rate": 2.130974151872522e-07,
      "loss": 0.84215176,
      "num_input_tokens_seen": 154089100,
      "step": 7120,
      "time_per_iteration": 2.6350057125091553
    },
    {
      "auxiliary_loss_clip": 0.01175764,
      "auxiliary_loss_mlp": 0.01021554,
      "balance_loss_clip": 1.04354715,
      "balance_loss_mlp": 1.01437807,
      "epoch": 0.8562496242409667,
      "flos": 22529206028160.0,
      "grad_norm": 1.875907333650519,
      "language_loss": 0.78754705,
      "learning_rate": 2.1274766775530773e-07,
      "loss": 0.80952024,
      "num_input_tokens_seen": 154108965,
      "step": 7121,
      "time_per_iteration": 2.696748971939087
    },
    {
      "auxiliary_loss_clip": 0.01168102,
      "auxiliary_loss_mlp": 0.01022559,
      "balance_loss_clip": 1.04594851,
      "balance_loss_mlp": 1.01509321,
      "epoch": 0.8563698671316058,
      "flos": 14712745472640.0,
      "grad_norm": 2.4679427492799575,
      "language_loss": 0.79818666,
      "learning_rate": 2.1239819144812077e-07,
      "loss": 0.82009321,
      "num_input_tokens_seen": 154123425,
      "step": 7122,
      "time_per_iteration": 2.568272829055786
    },
    {
      "auxiliary_loss_clip": 0.01168075,
      "auxiliary_loss_mlp": 0.01024617,
      "balance_loss_clip": 1.04030061,
      "balance_loss_mlp": 1.01710057,
      "epoch": 0.856490110022245,
      "flos": 39167768211840.0,
      "grad_norm": 1.7376659894560522,
      "language_loss": 0.7002269,
      "learning_rate": 2.1204898631870716e-07,
      "loss": 0.72215378,
      "num_input_tokens_seen": 154148315,
      "step": 7123,
      "time_per_iteration": 2.852647304534912
    },
    {
      "auxiliary_loss_clip": 0.01173437,
      "auxiliary_loss_mlp": 0.01021753,
      "balance_loss_clip": 1.04692888,
      "balance_loss_mlp": 1.01479673,
      "epoch": 0.856610352912884,
      "flos": 29059345658880.0,
      "grad_norm": 1.9369943955764124,
      "language_loss": 0.76040912,
      "learning_rate": 2.1170005242004006e-07,
      "loss": 0.78236097,
      "num_input_tokens_seen": 154169665,
      "step": 7124,
      "time_per_iteration": 2.719127655029297
    },
    {
      "auxiliary_loss_clip": 0.01176323,
      "auxiliary_loss_mlp": 0.01018993,
      "balance_loss_clip": 1.04389477,
      "balance_loss_mlp": 1.01176333,
      "epoch": 0.8567305958035231,
      "flos": 23878333883520.0,
      "grad_norm": 1.752903481952701,
      "language_loss": 0.78179204,
      "learning_rate": 2.1135138980505384e-07,
      "loss": 0.80374515,
      "num_input_tokens_seen": 154190335,
      "step": 7125,
      "time_per_iteration": 2.6386430263519287
    },
    {
      "auxiliary_loss_clip": 0.01168038,
      "auxiliary_loss_mlp": 0.01022541,
      "balance_loss_clip": 1.04520619,
      "balance_loss_mlp": 1.01506627,
      "epoch": 0.8568508386941622,
      "flos": 22200120599040.0,
      "grad_norm": 3.328338878962523,
      "language_loss": 0.72122657,
      "learning_rate": 2.110029985266395e-07,
      "loss": 0.74313235,
      "num_input_tokens_seen": 154210040,
      "step": 7126,
      "time_per_iteration": 3.599090814590454
    },
    {
      "auxiliary_loss_clip": 0.01175086,
      "auxiliary_loss_mlp": 0.01021904,
      "balance_loss_clip": 1.04308319,
      "balance_loss_mlp": 1.01463842,
      "epoch": 0.8569710815848013,
      "flos": 17307507121920.0,
      "grad_norm": 2.111406269558244,
      "language_loss": 0.74094599,
      "learning_rate": 2.1065487863764787e-07,
      "loss": 0.76291591,
      "num_input_tokens_seen": 154228385,
      "step": 7127,
      "time_per_iteration": 2.649991750717163
    },
    {
      "auxiliary_loss_clip": 0.01167315,
      "auxiliary_loss_mlp": 0.01021721,
      "balance_loss_clip": 1.03671145,
      "balance_loss_mlp": 1.01368046,
      "epoch": 0.8570913244754403,
      "flos": 23732285184000.0,
      "grad_norm": 4.378309563604398,
      "language_loss": 0.85897529,
      "learning_rate": 2.1030703019088846e-07,
      "loss": 0.88086569,
      "num_input_tokens_seen": 154249015,
      "step": 7128,
      "time_per_iteration": 2.7283637523651123
    },
    {
      "auxiliary_loss_clip": 0.01164569,
      "auxiliary_loss_mlp": 0.01021992,
      "balance_loss_clip": 1.04541039,
      "balance_loss_mlp": 1.01465154,
      "epoch": 0.8572115673660795,
      "flos": 20048748433920.0,
      "grad_norm": 1.8219351332749802,
      "language_loss": 0.70797038,
      "learning_rate": 2.099594532391291e-07,
      "loss": 0.72983599,
      "num_input_tokens_seen": 154267700,
      "step": 7129,
      "time_per_iteration": 2.701162815093994
    },
    {
      "auxiliary_loss_clip": 0.01160079,
      "auxiliary_loss_mlp": 0.0102309,
      "balance_loss_clip": 1.04437065,
      "balance_loss_mlp": 1.01565123,
      "epoch": 0.8573318102567186,
      "flos": 27160389342720.0,
      "grad_norm": 1.7066280755855938,
      "language_loss": 0.79074085,
      "learning_rate": 2.0961214783509806e-07,
      "loss": 0.81257248,
      "num_input_tokens_seen": 154290580,
      "step": 7130,
      "time_per_iteration": 2.676452159881592
    },
    {
      "auxiliary_loss_clip": 0.01176672,
      "auxiliary_loss_mlp": 0.01022292,
      "balance_loss_clip": 1.04382825,
      "balance_loss_mlp": 1.01466894,
      "epoch": 0.8574520531473576,
      "flos": 24936585402240.0,
      "grad_norm": 1.72593703433913,
      "language_loss": 0.74878442,
      "learning_rate": 2.0926511403148051e-07,
      "loss": 0.77077407,
      "num_input_tokens_seen": 154309545,
      "step": 7131,
      "time_per_iteration": 2.663382053375244
    },
    {
      "auxiliary_loss_clip": 0.01182368,
      "auxiliary_loss_mlp": 0.01029257,
      "balance_loss_clip": 1.04542422,
      "balance_loss_mlp": 1.02208924,
      "epoch": 0.8575722960379968,
      "flos": 18771154513920.0,
      "grad_norm": 2.0324892713103404,
      "language_loss": 0.76422894,
      "learning_rate": 2.0891835188092143e-07,
      "loss": 0.78634524,
      "num_input_tokens_seen": 154326545,
      "step": 7132,
      "time_per_iteration": 2.633444309234619
    },
    {
      "auxiliary_loss_clip": 0.0118083,
      "auxiliary_loss_mlp": 0.0102591,
      "balance_loss_clip": 1.04265571,
      "balance_loss_mlp": 1.01807785,
      "epoch": 0.8576925389286358,
      "flos": 22200300167040.0,
      "grad_norm": 2.0506737298643096,
      "language_loss": 0.81492043,
      "learning_rate": 2.0857186143602434e-07,
      "loss": 0.83698779,
      "num_input_tokens_seen": 154345190,
      "step": 7133,
      "time_per_iteration": 2.8222920894622803
    },
    {
      "auxiliary_loss_clip": 0.01163448,
      "auxiliary_loss_mlp": 0.01030969,
      "balance_loss_clip": 1.03978562,
      "balance_loss_mlp": 1.02305984,
      "epoch": 0.8578127818192749,
      "flos": 22894345733760.0,
      "grad_norm": 2.0162146804934675,
      "language_loss": 0.67466557,
      "learning_rate": 2.0822564274935094e-07,
      "loss": 0.69660974,
      "num_input_tokens_seen": 154364615,
      "step": 7134,
      "time_per_iteration": 3.558295249938965
    },
    {
      "auxiliary_loss_clip": 0.01171853,
      "auxiliary_loss_mlp": 0.01024657,
      "balance_loss_clip": 1.04658699,
      "balance_loss_mlp": 1.01676261,
      "epoch": 0.8579330247099141,
      "flos": 34824839541120.0,
      "grad_norm": 13.92690885304023,
      "language_loss": 0.67075783,
      "learning_rate": 2.078796958734239e-07,
      "loss": 0.69272292,
      "num_input_tokens_seen": 154387335,
      "step": 7135,
      "time_per_iteration": 3.679636240005493
    },
    {
      "auxiliary_loss_clip": 0.01169077,
      "auxiliary_loss_mlp": 0.01023488,
      "balance_loss_clip": 1.04588616,
      "balance_loss_mlp": 1.016294,
      "epoch": 0.8580532676005531,
      "flos": 19755681367680.0,
      "grad_norm": 3.1386229306325686,
      "language_loss": 0.75157428,
      "learning_rate": 2.0753402086072124e-07,
      "loss": 0.77349997,
      "num_input_tokens_seen": 154405965,
      "step": 7136,
      "time_per_iteration": 2.6406421661376953
    },
    {
      "auxiliary_loss_clip": 0.01193122,
      "auxiliary_loss_mlp": 0.01026177,
      "balance_loss_clip": 1.04131627,
      "balance_loss_mlp": 1.01877379,
      "epoch": 0.8581735104911922,
      "flos": 22739318634240.0,
      "grad_norm": 2.584655587862011,
      "language_loss": 0.7524175,
      "learning_rate": 2.071886177636828e-07,
      "loss": 0.77461052,
      "num_input_tokens_seen": 154422750,
      "step": 7137,
      "time_per_iteration": 2.7238929271698
    },
    {
      "auxiliary_loss_clip": 0.01167492,
      "auxiliary_loss_mlp": 0.01023174,
      "balance_loss_clip": 1.04631972,
      "balance_loss_mlp": 1.01603043,
      "epoch": 0.8582937533818313,
      "flos": 23149131880320.0,
      "grad_norm": 1.787047573851759,
      "language_loss": 0.83139265,
      "learning_rate": 2.0684348663470575e-07,
      "loss": 0.85329938,
      "num_input_tokens_seen": 154442930,
      "step": 7138,
      "time_per_iteration": 2.635164260864258
    },
    {
      "auxiliary_loss_clip": 0.01169558,
      "auxiliary_loss_mlp": 0.01027771,
      "balance_loss_clip": 1.04017961,
      "balance_loss_mlp": 1.02007008,
      "epoch": 0.8584139962724704,
      "flos": 19498668577920.0,
      "grad_norm": 5.527892761437455,
      "language_loss": 0.61302066,
      "learning_rate": 2.0649862752614555e-07,
      "loss": 0.63499403,
      "num_input_tokens_seen": 154461640,
      "step": 7139,
      "time_per_iteration": 3.5666399002075195
    },
    {
      "auxiliary_loss_clip": 0.01077692,
      "auxiliary_loss_mlp": 0.01002294,
      "balance_loss_clip": 1.01051283,
      "balance_loss_mlp": 1.00156045,
      "epoch": 0.8585342391631094,
      "flos": 71276577788160.0,
      "grad_norm": 0.7510285727690093,
      "language_loss": 0.57000315,
      "learning_rate": 2.0615404049031838e-07,
      "loss": 0.59080303,
      "num_input_tokens_seen": 154518610,
      "step": 7140,
      "time_per_iteration": 3.1625585556030273
    },
    {
      "auxiliary_loss_clip": 0.01170292,
      "auxiliary_loss_mlp": 0.01024567,
      "balance_loss_clip": 1.04600894,
      "balance_loss_mlp": 1.01652646,
      "epoch": 0.8586544820537486,
      "flos": 10815432929280.0,
      "grad_norm": 2.776663234793304,
      "language_loss": 0.78119153,
      "learning_rate": 2.0580972557949616e-07,
      "loss": 0.80314016,
      "num_input_tokens_seen": 154533700,
      "step": 7141,
      "time_per_iteration": 2.7078747749328613
    },
    {
      "auxiliary_loss_clip": 0.01071672,
      "auxiliary_loss_mlp": 0.01004448,
      "balance_loss_clip": 1.01054764,
      "balance_loss_mlp": 1.00370336,
      "epoch": 0.8587747249443877,
      "flos": 64811184422400.0,
      "grad_norm": 0.7929863448831773,
      "language_loss": 0.54186761,
      "learning_rate": 2.054656828459125e-07,
      "loss": 0.56262887,
      "num_input_tokens_seen": 154597810,
      "step": 7142,
      "time_per_iteration": 3.1606268882751465
    },
    {
      "auxiliary_loss_clip": 0.01173721,
      "auxiliary_loss_mlp": 0.01024494,
      "balance_loss_clip": 1.0412221,
      "balance_loss_mlp": 1.01660848,
      "epoch": 0.8588949678350267,
      "flos": 26834607964800.0,
      "grad_norm": 1.7902219041932124,
      "language_loss": 0.77238262,
      "learning_rate": 2.051219123417578e-07,
      "loss": 0.79436475,
      "num_input_tokens_seen": 154617870,
      "step": 7143,
      "time_per_iteration": 2.8706281185150146
    },
    {
      "auxiliary_loss_clip": 0.01169296,
      "auxiliary_loss_mlp": 0.01023499,
      "balance_loss_clip": 1.04665804,
      "balance_loss_mlp": 1.01492786,
      "epoch": 0.8590152107256659,
      "flos": 26104256726400.0,
      "grad_norm": 2.401740065310604,
      "language_loss": 0.60512769,
      "learning_rate": 2.0477841411918196e-07,
      "loss": 0.62705564,
      "num_input_tokens_seen": 154637395,
      "step": 7144,
      "time_per_iteration": 2.6262362003326416
    },
    {
      "auxiliary_loss_clip": 0.01161617,
      "auxiliary_loss_mlp": 0.01022346,
      "balance_loss_clip": 1.04322422,
      "balance_loss_mlp": 1.01477623,
      "epoch": 0.859135453616305,
      "flos": 26140885620480.0,
      "grad_norm": 1.8628453244032845,
      "language_loss": 0.74474245,
      "learning_rate": 2.0443518823029326e-07,
      "loss": 0.76658213,
      "num_input_tokens_seen": 154657935,
      "step": 7145,
      "time_per_iteration": 2.689995288848877
    },
    {
      "auxiliary_loss_clip": 0.01167397,
      "auxiliary_loss_mlp": 0.01026791,
      "balance_loss_clip": 1.04016328,
      "balance_loss_mlp": 1.01913822,
      "epoch": 0.859255696506944,
      "flos": 12969319046400.0,
      "grad_norm": 2.3382241042742433,
      "language_loss": 0.76789331,
      "learning_rate": 2.0409223472715854e-07,
      "loss": 0.78983521,
      "num_input_tokens_seen": 154675080,
      "step": 7146,
      "time_per_iteration": 2.6748383045196533
    },
    {
      "auxiliary_loss_clip": 0.01176261,
      "auxiliary_loss_mlp": 0.01207269,
      "balance_loss_clip": 1.04420221,
      "balance_loss_mlp": 1.00075686,
      "epoch": 0.8593759393975832,
      "flos": 18475753063680.0,
      "grad_norm": 2.2890229808179687,
      "language_loss": 0.74834597,
      "learning_rate": 2.0374955366180434e-07,
      "loss": 0.77218127,
      "num_input_tokens_seen": 154692720,
      "step": 7147,
      "time_per_iteration": 2.639108896255493
    },
    {
      "auxiliary_loss_clip": 0.01178287,
      "auxiliary_loss_mlp": 0.01020312,
      "balance_loss_clip": 1.04196525,
      "balance_loss_mlp": 1.01267326,
      "epoch": 0.8594961822882222,
      "flos": 22200156512640.0,
      "grad_norm": 2.295225081531143,
      "language_loss": 0.72733235,
      "learning_rate": 2.034071450862147e-07,
      "loss": 0.74931836,
      "num_input_tokens_seen": 154710190,
      "step": 7148,
      "time_per_iteration": 2.6659486293792725
    },
    {
      "auxiliary_loss_clip": 0.01174275,
      "auxiliary_loss_mlp": 0.01024989,
      "balance_loss_clip": 1.04338431,
      "balance_loss_mlp": 1.01716912,
      "epoch": 0.8596164251788613,
      "flos": 23294749616640.0,
      "grad_norm": 1.9488091293071634,
      "language_loss": 0.77074462,
      "learning_rate": 2.030650090523327e-07,
      "loss": 0.79273725,
      "num_input_tokens_seen": 154729380,
      "step": 7149,
      "time_per_iteration": 2.7641098499298096
    },
    {
      "auxiliary_loss_clip": 0.01171534,
      "auxiliary_loss_mlp": 0.01021897,
      "balance_loss_clip": 1.04124737,
      "balance_loss_mlp": 1.01413035,
      "epoch": 0.8597366680695004,
      "flos": 31649905416960.0,
      "grad_norm": 1.750404352246518,
      "language_loss": 0.59512258,
      "learning_rate": 2.0272314561205995e-07,
      "loss": 0.61705691,
      "num_input_tokens_seen": 154749775,
      "step": 7150,
      "time_per_iteration": 2.7128751277923584
    },
    {
      "auxiliary_loss_clip": 0.01167557,
      "auxiliary_loss_mlp": 0.01018742,
      "balance_loss_clip": 1.0397321,
      "balance_loss_mlp": 1.01165497,
      "epoch": 0.8598569109601395,
      "flos": 21287738211840.0,
      "grad_norm": 1.900139134339894,
      "language_loss": 0.73500395,
      "learning_rate": 2.023815548172567e-07,
      "loss": 0.75686693,
      "num_input_tokens_seen": 154769845,
      "step": 7151,
      "time_per_iteration": 2.7222630977630615
    },
    {
      "auxiliary_loss_clip": 0.01169876,
      "auxiliary_loss_mlp": 0.01023388,
      "balance_loss_clip": 1.04557443,
      "balance_loss_mlp": 1.01565087,
      "epoch": 0.8599771538507786,
      "flos": 25447809720960.0,
      "grad_norm": 1.8836417067501938,
      "language_loss": 0.6611796,
      "learning_rate": 2.0204023671974267e-07,
      "loss": 0.6831122,
      "num_input_tokens_seen": 154789230,
      "step": 7152,
      "time_per_iteration": 3.59348201751709
    },
    {
      "auxiliary_loss_clip": 0.01161047,
      "auxiliary_loss_mlp": 0.0102487,
      "balance_loss_clip": 1.04270053,
      "balance_loss_mlp": 1.01709723,
      "epoch": 0.8600973967414177,
      "flos": 16723958768640.0,
      "grad_norm": 3.0048319847442877,
      "language_loss": 0.80581856,
      "learning_rate": 2.0169919137129532e-07,
      "loss": 0.82767773,
      "num_input_tokens_seen": 154807670,
      "step": 7153,
      "time_per_iteration": 2.540822744369507
    },
    {
      "auxiliary_loss_clip": 0.01172205,
      "auxiliary_loss_mlp": 0.01024723,
      "balance_loss_clip": 1.04820478,
      "balance_loss_mlp": 1.01659882,
      "epoch": 0.8602176396320568,
      "flos": 25227928615680.0,
      "grad_norm": 2.4898017817831355,
      "language_loss": 0.70534015,
      "learning_rate": 2.013584188236508e-07,
      "loss": 0.72730947,
      "num_input_tokens_seen": 154825575,
      "step": 7154,
      "time_per_iteration": 2.728067636489868
    },
    {
      "auxiliary_loss_clip": 0.01169199,
      "auxiliary_loss_mlp": 0.01023954,
      "balance_loss_clip": 1.04794884,
      "balance_loss_mlp": 1.01596677,
      "epoch": 0.8603378825226958,
      "flos": 20412236113920.0,
      "grad_norm": 1.6459665352985178,
      "language_loss": 0.79125512,
      "learning_rate": 2.0101791912850396e-07,
      "loss": 0.81318665,
      "num_input_tokens_seen": 154845115,
      "step": 7155,
      "time_per_iteration": 2.623875379562378
    },
    {
      "auxiliary_loss_clip": 0.01173577,
      "auxiliary_loss_mlp": 0.01018907,
      "balance_loss_clip": 1.04686928,
      "balance_loss_mlp": 1.01168323,
      "epoch": 0.8604581254133349,
      "flos": 34930201109760.0,
      "grad_norm": 2.647442283257097,
      "language_loss": 0.64069647,
      "learning_rate": 2.006776923375082e-07,
      "loss": 0.66262138,
      "num_input_tokens_seen": 154866770,
      "step": 7156,
      "time_per_iteration": 2.7338459491729736
    },
    {
      "auxiliary_loss_clip": 0.01168107,
      "auxiliary_loss_mlp": 0.0102236,
      "balance_loss_clip": 1.04787099,
      "balance_loss_mlp": 1.01480842,
      "epoch": 0.860578368303974,
      "flos": 22596538072320.0,
      "grad_norm": 1.622177606401057,
      "language_loss": 0.7120955,
      "learning_rate": 2.003377385022764e-07,
      "loss": 0.73400015,
      "num_input_tokens_seen": 154885595,
      "step": 7157,
      "time_per_iteration": 2.586143970489502
    },
    {
      "auxiliary_loss_clip": 0.0117295,
      "auxiliary_loss_mlp": 0.01026862,
      "balance_loss_clip": 1.04299974,
      "balance_loss_mlp": 1.01973319,
      "epoch": 0.8606986111946131,
      "flos": 21324331192320.0,
      "grad_norm": 2.0973949772039817,
      "language_loss": 0.77509952,
      "learning_rate": 1.9999805767437826e-07,
      "loss": 0.79709768,
      "num_input_tokens_seen": 154904485,
      "step": 7158,
      "time_per_iteration": 2.8031747341156006
    },
    {
      "auxiliary_loss_clip": 0.01165744,
      "auxiliary_loss_mlp": 0.010222,
      "balance_loss_clip": 1.04184902,
      "balance_loss_mlp": 1.01511598,
      "epoch": 0.8608188540852522,
      "flos": 28877206769280.0,
      "grad_norm": 2.078721699339898,
      "language_loss": 0.71884871,
      "learning_rate": 1.9965864990534386e-07,
      "loss": 0.74072808,
      "num_input_tokens_seen": 154925010,
      "step": 7159,
      "time_per_iteration": 2.731905221939087
    },
    {
      "auxiliary_loss_clip": 0.01167481,
      "auxiliary_loss_mlp": 0.01021919,
      "balance_loss_clip": 1.03947282,
      "balance_loss_mlp": 1.0150671,
      "epoch": 0.8609390969758913,
      "flos": 29716187713920.0,
      "grad_norm": 1.7201809883104497,
      "language_loss": 0.77541882,
      "learning_rate": 1.9931951524666092e-07,
      "loss": 0.79731274,
      "num_input_tokens_seen": 154946100,
      "step": 7160,
      "time_per_iteration": 3.6907200813293457
    },
    {
      "auxiliary_loss_clip": 0.01173109,
      "auxiliary_loss_mlp": 0.01207656,
      "balance_loss_clip": 1.04751527,
      "balance_loss_mlp": 1.00073373,
      "epoch": 0.8610593398665304,
      "flos": 21249349551360.0,
      "grad_norm": 1.617312944998265,
      "language_loss": 0.81410325,
      "learning_rate": 1.9898065374977534e-07,
      "loss": 0.83791089,
      "num_input_tokens_seen": 154966305,
      "step": 7161,
      "time_per_iteration": 2.6764025688171387
    },
    {
      "auxiliary_loss_clip": 0.01169134,
      "auxiliary_loss_mlp": 0.01019925,
      "balance_loss_clip": 1.04048228,
      "balance_loss_mlp": 1.01346421,
      "epoch": 0.8611795827571694,
      "flos": 14830102183680.0,
      "grad_norm": 2.113776995940045,
      "language_loss": 0.73269594,
      "learning_rate": 1.9864206546609342e-07,
      "loss": 0.75458646,
      "num_input_tokens_seen": 154985145,
      "step": 7162,
      "time_per_iteration": 3.6062796115875244
    },
    {
      "auxiliary_loss_clip": 0.01166589,
      "auxiliary_loss_mlp": 0.0102044,
      "balance_loss_clip": 1.04669034,
      "balance_loss_mlp": 1.01310551,
      "epoch": 0.8612998256478086,
      "flos": 24243258107520.0,
      "grad_norm": 1.8756910686081831,
      "language_loss": 0.84611708,
      "learning_rate": 1.983037504469771e-07,
      "loss": 0.86798739,
      "num_input_tokens_seen": 155003855,
      "step": 7163,
      "time_per_iteration": 2.616257667541504
    },
    {
      "auxiliary_loss_clip": 0.01174709,
      "auxiliary_loss_mlp": 0.01024651,
      "balance_loss_clip": 1.04963326,
      "balance_loss_mlp": 1.01684332,
      "epoch": 0.8614200685384477,
      "flos": 21252653602560.0,
      "grad_norm": 2.8335715788941345,
      "language_loss": 0.66849738,
      "learning_rate": 1.9796570874374984e-07,
      "loss": 0.69049096,
      "num_input_tokens_seen": 155023960,
      "step": 7164,
      "time_per_iteration": 2.591139078140259
    },
    {
      "auxiliary_loss_clip": 0.01173349,
      "auxiliary_loss_mlp": 0.01020258,
      "balance_loss_clip": 1.04401517,
      "balance_loss_mlp": 1.01262259,
      "epoch": 0.8615403114290867,
      "flos": 20007738080640.0,
      "grad_norm": 2.17588354386085,
      "language_loss": 0.77808201,
      "learning_rate": 1.976279404076917e-07,
      "loss": 0.80001807,
      "num_input_tokens_seen": 155043360,
      "step": 7165,
      "time_per_iteration": 2.6590490341186523
    },
    {
      "auxiliary_loss_clip": 0.01175437,
      "auxiliary_loss_mlp": 0.01025371,
      "balance_loss_clip": 1.04505539,
      "balance_loss_mlp": 1.01794386,
      "epoch": 0.8616605543197259,
      "flos": 29789373674880.0,
      "grad_norm": 2.270265107581676,
      "language_loss": 0.76031148,
      "learning_rate": 1.9729044549004193e-07,
      "loss": 0.78231955,
      "num_input_tokens_seen": 155064745,
      "step": 7166,
      "time_per_iteration": 3.6734116077423096
    },
    {
      "auxiliary_loss_clip": 0.01168052,
      "auxiliary_loss_mlp": 0.01022738,
      "balance_loss_clip": 1.0461458,
      "balance_loss_mlp": 1.01551962,
      "epoch": 0.8617807972103649,
      "flos": 28911609020160.0,
      "grad_norm": 1.8573517865583624,
      "language_loss": 0.7021758,
      "learning_rate": 1.9695322404199822e-07,
      "loss": 0.72408366,
      "num_input_tokens_seen": 155086790,
      "step": 7167,
      "time_per_iteration": 2.6640567779541016
    },
    {
      "auxiliary_loss_clip": 0.01170711,
      "auxiliary_loss_mlp": 0.01026448,
      "balance_loss_clip": 1.04541469,
      "balance_loss_mlp": 1.01891983,
      "epoch": 0.861901040101004,
      "flos": 27673804391040.0,
      "grad_norm": 3.119698745458783,
      "language_loss": 0.82598615,
      "learning_rate": 1.9661627611471654e-07,
      "loss": 0.84795773,
      "num_input_tokens_seen": 155106585,
      "step": 7168,
      "time_per_iteration": 2.7176225185394287
    },
    {
      "auxiliary_loss_clip": 0.01178679,
      "auxiliary_loss_mlp": 0.01021679,
      "balance_loss_clip": 1.04525483,
      "balance_loss_mlp": 1.0136627,
      "epoch": 0.8620212829916432,
      "flos": 49748056755840.0,
      "grad_norm": 2.6462227639134404,
      "language_loss": 0.70299071,
      "learning_rate": 1.9627960175931246e-07,
      "loss": 0.7249943,
      "num_input_tokens_seen": 155131285,
      "step": 7169,
      "time_per_iteration": 2.8723011016845703
    },
    {
      "auxiliary_loss_clip": 0.01169442,
      "auxiliary_loss_mlp": 0.01027992,
      "balance_loss_clip": 1.04816508,
      "balance_loss_mlp": 1.02085745,
      "epoch": 0.8621415258822822,
      "flos": 21138672769920.0,
      "grad_norm": 2.187477149530627,
      "language_loss": 0.74178004,
      "learning_rate": 1.9594320102685847e-07,
      "loss": 0.76375431,
      "num_input_tokens_seen": 155150555,
      "step": 7170,
      "time_per_iteration": 2.596162796020508
    },
    {
      "auxiliary_loss_clip": 0.01161594,
      "auxiliary_loss_mlp": 0.01207376,
      "balance_loss_clip": 1.04276705,
      "balance_loss_mlp": 1.00059867,
      "epoch": 0.8622617687729213,
      "flos": 21689039934720.0,
      "grad_norm": 2.102153513997858,
      "language_loss": 0.64146578,
      "learning_rate": 1.956070739683864e-07,
      "loss": 0.66515553,
      "num_input_tokens_seen": 155169890,
      "step": 7171,
      "time_per_iteration": 2.657956123352051
    },
    {
      "auxiliary_loss_clip": 0.0115982,
      "auxiliary_loss_mlp": 0.0102001,
      "balance_loss_clip": 1.03952205,
      "balance_loss_mlp": 1.01249957,
      "epoch": 0.8623820116635604,
      "flos": 26250592734720.0,
      "grad_norm": 1.583153246665297,
      "language_loss": 0.74348044,
      "learning_rate": 1.9527122063488678e-07,
      "loss": 0.76527882,
      "num_input_tokens_seen": 155191005,
      "step": 7172,
      "time_per_iteration": 2.734034776687622
    },
    {
      "auxiliary_loss_clip": 0.01167328,
      "auxiliary_loss_mlp": 0.01019742,
      "balance_loss_clip": 1.03987646,
      "balance_loss_mlp": 1.0129981,
      "epoch": 0.8625022545541995,
      "flos": 19647554451840.0,
      "grad_norm": 2.508984867710035,
      "language_loss": 0.80510223,
      "learning_rate": 1.9493564107730755e-07,
      "loss": 0.8269729,
      "num_input_tokens_seen": 155211005,
      "step": 7173,
      "time_per_iteration": 2.6334216594696045
    },
    {
      "auxiliary_loss_clip": 0.01165098,
      "auxiliary_loss_mlp": 0.0102425,
      "balance_loss_clip": 1.04059029,
      "balance_loss_mlp": 1.0174129,
      "epoch": 0.8626224974448385,
      "flos": 21908382336000.0,
      "grad_norm": 3.009396260281493,
      "language_loss": 0.61084211,
      "learning_rate": 1.9460033534655684e-07,
      "loss": 0.63273561,
      "num_input_tokens_seen": 155230365,
      "step": 7174,
      "time_per_iteration": 2.5555238723754883
    },
    {
      "auxiliary_loss_clip": 0.01164039,
      "auxiliary_loss_mlp": 0.01023936,
      "balance_loss_clip": 1.03903878,
      "balance_loss_mlp": 1.01646781,
      "epoch": 0.8627427403354777,
      "flos": 23331198942720.0,
      "grad_norm": 1.5119307060221363,
      "language_loss": 0.84251881,
      "learning_rate": 1.9426530349349978e-07,
      "loss": 0.86439854,
      "num_input_tokens_seen": 155250815,
      "step": 7175,
      "time_per_iteration": 2.678060293197632
    },
    {
      "auxiliary_loss_clip": 0.01167473,
      "auxiliary_loss_mlp": 0.01207617,
      "balance_loss_clip": 1.04484415,
      "balance_loss_mlp": 1.00080371,
      "epoch": 0.8628629832261168,
      "flos": 16362877299840.0,
      "grad_norm": 1.9209245823041456,
      "language_loss": 0.64952308,
      "learning_rate": 1.9393054556896038e-07,
      "loss": 0.67327392,
      "num_input_tokens_seen": 155268515,
      "step": 7176,
      "time_per_iteration": 2.5937626361846924
    },
    {
      "auxiliary_loss_clip": 0.01171258,
      "auxiliary_loss_mlp": 0.01022857,
      "balance_loss_clip": 1.04057598,
      "balance_loss_mlp": 1.01500094,
      "epoch": 0.8629832261167558,
      "flos": 28103941756800.0,
      "grad_norm": 2.618255470916982,
      "language_loss": 0.69028342,
      "learning_rate": 1.9359606162372133e-07,
      "loss": 0.71222454,
      "num_input_tokens_seen": 155290120,
      "step": 7177,
      "time_per_iteration": 2.743732452392578
    },
    {
      "auxiliary_loss_clip": 0.01166426,
      "auxiliary_loss_mlp": 0.01023709,
      "balance_loss_clip": 1.04778314,
      "balance_loss_mlp": 1.01624346,
      "epoch": 0.863103469007395,
      "flos": 20230061310720.0,
      "grad_norm": 1.7565973019838574,
      "language_loss": 0.70636719,
      "learning_rate": 1.9326185170852293e-07,
      "loss": 0.7282685,
      "num_input_tokens_seen": 155309085,
      "step": 7178,
      "time_per_iteration": 2.5718119144439697
    },
    {
      "auxiliary_loss_clip": 0.01167255,
      "auxiliary_loss_mlp": 0.01022801,
      "balance_loss_clip": 1.04517984,
      "balance_loss_mlp": 1.01569581,
      "epoch": 0.863223711898034,
      "flos": 24498547044480.0,
      "grad_norm": 2.0500476900638325,
      "language_loss": 0.72416276,
      "learning_rate": 1.9292791587406598e-07,
      "loss": 0.74606335,
      "num_input_tokens_seen": 155327945,
      "step": 7179,
      "time_per_iteration": 3.6297216415405273
    },
    {
      "auxiliary_loss_clip": 0.01166283,
      "auxiliary_loss_mlp": 0.01207919,
      "balance_loss_clip": 1.04379296,
      "balance_loss_mlp": 1.0007658,
      "epoch": 0.8633439547886731,
      "flos": 17675376261120.0,
      "grad_norm": 3.277180140705791,
      "language_loss": 0.86882144,
      "learning_rate": 1.9259425417100661e-07,
      "loss": 0.89256346,
      "num_input_tokens_seen": 155344060,
      "step": 7180,
      "time_per_iteration": 2.6376943588256836
    },
    {
      "auxiliary_loss_clip": 0.01174477,
      "auxiliary_loss_mlp": 0.0102728,
      "balance_loss_clip": 1.03413296,
      "balance_loss_mlp": 1.01935863,
      "epoch": 0.8634641976793123,
      "flos": 12895055677440.0,
      "grad_norm": 2.254058584350962,
      "language_loss": 0.75461608,
      "learning_rate": 1.9226086664996234e-07,
      "loss": 0.77663362,
      "num_input_tokens_seen": 155362305,
      "step": 7181,
      "time_per_iteration": 2.6635866165161133
    },
    {
      "auxiliary_loss_clip": 0.01175636,
      "auxiliary_loss_mlp": 0.0102591,
      "balance_loss_clip": 1.04678583,
      "balance_loss_mlp": 1.0187335,
      "epoch": 0.8635844405699513,
      "flos": 23878980328320.0,
      "grad_norm": 3.258341946945112,
      "language_loss": 0.74003983,
      "learning_rate": 1.9192775336150712e-07,
      "loss": 0.76205528,
      "num_input_tokens_seen": 155382605,
      "step": 7182,
      "time_per_iteration": 2.6910557746887207
    },
    {
      "auxiliary_loss_clip": 0.01066068,
      "auxiliary_loss_mlp": 0.01001651,
      "balance_loss_clip": 1.01053572,
      "balance_loss_mlp": 1.00089407,
      "epoch": 0.8637046834605904,
      "flos": 60453387521280.0,
      "grad_norm": 0.7675568120005937,
      "language_loss": 0.56197625,
      "learning_rate": 1.915949143561739e-07,
      "loss": 0.5826534,
      "num_input_tokens_seen": 155437280,
      "step": 7183,
      "time_per_iteration": 3.140069007873535
    },
    {
      "auxiliary_loss_clip": 0.01168998,
      "auxiliary_loss_mlp": 0.01027421,
      "balance_loss_clip": 1.04673004,
      "balance_loss_mlp": 1.0205636,
      "epoch": 0.8638249263512295,
      "flos": 20558751690240.0,
      "grad_norm": 2.036673873625124,
      "language_loss": 0.77863503,
      "learning_rate": 1.9126234968445498e-07,
      "loss": 0.80059922,
      "num_input_tokens_seen": 155456970,
      "step": 7184,
      "time_per_iteration": 2.6292991638183594
    },
    {
      "auxiliary_loss_clip": 0.01168435,
      "auxiliary_loss_mlp": 0.01028285,
      "balance_loss_clip": 1.04860866,
      "balance_loss_mlp": 1.02063227,
      "epoch": 0.8639451692418686,
      "flos": 26615768353920.0,
      "grad_norm": 1.9851667200466747,
      "language_loss": 0.67703366,
      "learning_rate": 1.9093005939679884e-07,
      "loss": 0.6990009,
      "num_input_tokens_seen": 155478925,
      "step": 7185,
      "time_per_iteration": 2.636950731277466
    },
    {
      "auxiliary_loss_clip": 0.01168363,
      "auxiliary_loss_mlp": 0.01028752,
      "balance_loss_clip": 1.04551589,
      "balance_loss_mlp": 1.0211587,
      "epoch": 0.8640654121325076,
      "flos": 15122450977920.0,
      "grad_norm": 20.894466286853152,
      "language_loss": 0.76478803,
      "learning_rate": 1.9059804354361452e-07,
      "loss": 0.78675914,
      "num_input_tokens_seen": 155496700,
      "step": 7186,
      "time_per_iteration": 3.556633234024048
    },
    {
      "auxiliary_loss_clip": 0.01164038,
      "auxiliary_loss_mlp": 0.01022223,
      "balance_loss_clip": 1.04008341,
      "balance_loss_mlp": 1.01449537,
      "epoch": 0.8641856550231467,
      "flos": 31869068250240.0,
      "grad_norm": 1.7135482740764545,
      "language_loss": 0.70124811,
      "learning_rate": 1.902663021752684e-07,
      "loss": 0.72311068,
      "num_input_tokens_seen": 155518130,
      "step": 7187,
      "time_per_iteration": 2.7195305824279785
    },
    {
      "auxiliary_loss_clip": 0.01171182,
      "auxiliary_loss_mlp": 0.01022303,
      "balance_loss_clip": 1.05028486,
      "balance_loss_mlp": 1.01489723,
      "epoch": 0.8643058979137859,
      "flos": 14976545932800.0,
      "grad_norm": 2.131650440034889,
      "language_loss": 0.82611263,
      "learning_rate": 1.8993483534208556e-07,
      "loss": 0.84804749,
      "num_input_tokens_seen": 155537040,
      "step": 7188,
      "time_per_iteration": 3.4799716472625732
    },
    {
      "auxiliary_loss_clip": 0.0116461,
      "auxiliary_loss_mlp": 0.01026839,
      "balance_loss_clip": 1.04369879,
      "balance_loss_mlp": 1.01899791,
      "epoch": 0.8644261408044249,
      "flos": 13115726881920.0,
      "grad_norm": 2.6844699291840874,
      "language_loss": 0.74952185,
      "learning_rate": 1.8960364309434884e-07,
      "loss": 0.77143633,
      "num_input_tokens_seen": 155554535,
      "step": 7189,
      "time_per_iteration": 2.6205103397369385
    },
    {
      "auxiliary_loss_clip": 0.01175188,
      "auxiliary_loss_mlp": 0.0120752,
      "balance_loss_clip": 1.03877866,
      "balance_loss_mlp": 1.00078225,
      "epoch": 0.864546383695064,
      "flos": 20850920916480.0,
      "grad_norm": 1.8724436825543451,
      "language_loss": 0.78474158,
      "learning_rate": 1.8927272548229967e-07,
      "loss": 0.80856866,
      "num_input_tokens_seen": 155574225,
      "step": 7190,
      "time_per_iteration": 2.6933984756469727
    },
    {
      "auxiliary_loss_clip": 0.01178304,
      "auxiliary_loss_mlp": 0.01022007,
      "balance_loss_clip": 1.04271007,
      "balance_loss_mlp": 1.01466322,
      "epoch": 0.8646666265857031,
      "flos": 21324582587520.0,
      "grad_norm": 1.5722080347832708,
      "language_loss": 0.82824039,
      "learning_rate": 1.8894208255613876e-07,
      "loss": 0.85024345,
      "num_input_tokens_seen": 155593540,
      "step": 7191,
      "time_per_iteration": 2.7613253593444824
    },
    {
      "auxiliary_loss_clip": 0.01167413,
      "auxiliary_loss_mlp": 0.01022325,
      "balance_loss_clip": 1.04857969,
      "balance_loss_mlp": 1.0148387,
      "epoch": 0.8647868694763422,
      "flos": 19750833031680.0,
      "grad_norm": 2.63553714983274,
      "language_loss": 0.77814955,
      "learning_rate": 1.8861171436602397e-07,
      "loss": 0.80004692,
      "num_input_tokens_seen": 155610655,
      "step": 7192,
      "time_per_iteration": 3.565246105194092
    },
    {
      "auxiliary_loss_clip": 0.01170133,
      "auxiliary_loss_mlp": 0.01026503,
      "balance_loss_clip": 1.04610801,
      "balance_loss_mlp": 1.01876009,
      "epoch": 0.8649071123669813,
      "flos": 26176760328960.0,
      "grad_norm": 4.276014190443245,
      "language_loss": 0.8055644,
      "learning_rate": 1.882816209620719e-07,
      "loss": 0.82753074,
      "num_input_tokens_seen": 155627365,
      "step": 7193,
      "time_per_iteration": 2.6525638103485107
    },
    {
      "auxiliary_loss_clip": 0.01174397,
      "auxiliary_loss_mlp": 0.01022882,
      "balance_loss_clip": 1.04747736,
      "balance_loss_mlp": 1.01504445,
      "epoch": 0.8650273552576204,
      "flos": 20302888135680.0,
      "grad_norm": 1.8705106041551214,
      "language_loss": 0.76886249,
      "learning_rate": 1.8795180239435738e-07,
      "loss": 0.79083532,
      "num_input_tokens_seen": 155646220,
      "step": 7194,
      "time_per_iteration": 2.6091015338897705
    },
    {
      "auxiliary_loss_clip": 0.01176305,
      "auxiliary_loss_mlp": 0.01024771,
      "balance_loss_clip": 1.04456854,
      "balance_loss_mlp": 1.01713824,
      "epoch": 0.8651475981482595,
      "flos": 23951088881280.0,
      "grad_norm": 2.516450857019715,
      "language_loss": 0.75131881,
      "learning_rate": 1.8762225871291348e-07,
      "loss": 0.77332962,
      "num_input_tokens_seen": 155662095,
      "step": 7195,
      "time_per_iteration": 2.630721092224121
    },
    {
      "auxiliary_loss_clip": 0.01168456,
      "auxiliary_loss_mlp": 0.01207714,
      "balance_loss_clip": 1.04825544,
      "balance_loss_mlp": 1.00076461,
      "epoch": 0.8652678410388985,
      "flos": 21684622561920.0,
      "grad_norm": 1.7036318904924073,
      "language_loss": 0.81184149,
      "learning_rate": 1.8729298996773201e-07,
      "loss": 0.83560324,
      "num_input_tokens_seen": 155680845,
      "step": 7196,
      "time_per_iteration": 2.589351177215576
    },
    {
      "auxiliary_loss_clip": 0.01067789,
      "auxiliary_loss_mlp": 0.01002557,
      "balance_loss_clip": 1.0117867,
      "balance_loss_mlp": 1.00177574,
      "epoch": 0.8653880839295377,
      "flos": 65224660855680.0,
      "grad_norm": 0.8325741040602815,
      "language_loss": 0.60898757,
      "learning_rate": 1.8696399620876301e-07,
      "loss": 0.629691,
      "num_input_tokens_seen": 155737875,
      "step": 7197,
      "time_per_iteration": 3.0794849395751953
    },
    {
      "auxiliary_loss_clip": 0.01172632,
      "auxiliary_loss_mlp": 0.01024769,
      "balance_loss_clip": 1.03988504,
      "balance_loss_mlp": 1.01679683,
      "epoch": 0.8655083268201768,
      "flos": 17749172753280.0,
      "grad_norm": 1.971235992831256,
      "language_loss": 0.79524422,
      "learning_rate": 1.866352774859141e-07,
      "loss": 0.81721824,
      "num_input_tokens_seen": 155753100,
      "step": 7198,
      "time_per_iteration": 2.591992139816284
    },
    {
      "auxiliary_loss_clip": 0.01176902,
      "auxiliary_loss_mlp": 0.01024391,
      "balance_loss_clip": 1.04082203,
      "balance_loss_mlp": 1.01757193,
      "epoch": 0.8656285697108158,
      "flos": 20703974376960.0,
      "grad_norm": 2.277130030721523,
      "language_loss": 0.68749642,
      "learning_rate": 1.8630683384905188e-07,
      "loss": 0.70950937,
      "num_input_tokens_seen": 155772430,
      "step": 7199,
      "time_per_iteration": 2.6684629917144775
    },
    {
      "auxiliary_loss_clip": 0.01169653,
      "auxiliary_loss_mlp": 0.01207899,
      "balance_loss_clip": 1.04988933,
      "balance_loss_mlp": 1.00072098,
      "epoch": 0.865748812601455,
      "flos": 18653833716480.0,
      "grad_norm": 2.9449386828834117,
      "language_loss": 0.8875525,
      "learning_rate": 1.8597866534800045e-07,
      "loss": 0.91132796,
      "num_input_tokens_seen": 155787545,
      "step": 7200,
      "time_per_iteration": 2.6639745235443115
    },
    {
      "auxiliary_loss_clip": 0.01172625,
      "auxiliary_loss_mlp": 0.01207917,
      "balance_loss_clip": 1.04683185,
      "balance_loss_mlp": 1.00059521,
      "epoch": 0.865869055492094,
      "flos": 70652554807680.0,
      "grad_norm": 1.8875395177548067,
      "language_loss": 0.74600255,
      "learning_rate": 1.8565077203254398e-07,
      "loss": 0.76980799,
      "num_input_tokens_seen": 155813005,
      "step": 7201,
      "time_per_iteration": 2.970611810684204
    },
    {
      "auxiliary_loss_clip": 0.01175328,
      "auxiliary_loss_mlp": 0.01027957,
      "balance_loss_clip": 1.04648507,
      "balance_loss_mlp": 1.02039301,
      "epoch": 0.8659892983827331,
      "flos": 17383961220480.0,
      "grad_norm": 2.650887189466114,
      "language_loss": 0.73044777,
      "learning_rate": 1.8532315395242203e-07,
      "loss": 0.75248057,
      "num_input_tokens_seen": 155829455,
      "step": 7202,
      "time_per_iteration": 2.7280099391937256
    },
    {
      "auxiliary_loss_clip": 0.01174997,
      "auxiliary_loss_mlp": 0.01023328,
      "balance_loss_clip": 1.04195213,
      "balance_loss_mlp": 1.01617503,
      "epoch": 0.8661095412733723,
      "flos": 17895221452800.0,
      "grad_norm": 4.150486544486586,
      "language_loss": 0.72753465,
      "learning_rate": 1.849958111573353e-07,
      "loss": 0.74951792,
      "num_input_tokens_seen": 155848060,
      "step": 7203,
      "time_per_iteration": 2.6487762928009033
    },
    {
      "auxiliary_loss_clip": 0.01163935,
      "auxiliary_loss_mlp": 0.01022839,
      "balance_loss_clip": 1.04611039,
      "balance_loss_mlp": 1.01587737,
      "epoch": 0.8662297841640113,
      "flos": 18224163227520.0,
      "grad_norm": 1.6587810766282396,
      "language_loss": 0.64494205,
      "learning_rate": 1.8466874369694074e-07,
      "loss": 0.6668098,
      "num_input_tokens_seen": 155865755,
      "step": 7204,
      "time_per_iteration": 2.638248920440674
    },
    {
      "auxiliary_loss_clip": 0.01174165,
      "auxiliary_loss_mlp": 0.01022779,
      "balance_loss_clip": 1.03920233,
      "balance_loss_mlp": 1.01561475,
      "epoch": 0.8663500270546504,
      "flos": 16362159027840.0,
      "grad_norm": 3.382664271379569,
      "language_loss": 0.70435131,
      "learning_rate": 1.843419516208542e-07,
      "loss": 0.72632074,
      "num_input_tokens_seen": 155882680,
      "step": 7205,
      "time_per_iteration": 2.7097737789154053
    },
    {
      "auxiliary_loss_clip": 0.01172382,
      "auxiliary_loss_mlp": 0.01022429,
      "balance_loss_clip": 1.04879558,
      "balance_loss_mlp": 1.01454961,
      "epoch": 0.8664702699452895,
      "flos": 17894431353600.0,
      "grad_norm": 2.3450214136741954,
      "language_loss": 0.79912776,
      "learning_rate": 1.8401543497865047e-07,
      "loss": 0.82107592,
      "num_input_tokens_seen": 155900680,
      "step": 7206,
      "time_per_iteration": 3.5186870098114014
    },
    {
      "auxiliary_loss_clip": 0.01172066,
      "auxiliary_loss_mlp": 0.0120759,
      "balance_loss_clip": 1.04689538,
      "balance_loss_mlp": 1.00073934,
      "epoch": 0.8665905128359286,
      "flos": 30736373794560.0,
      "grad_norm": 2.5259054284400295,
      "language_loss": 0.64659083,
      "learning_rate": 1.836891938198608e-07,
      "loss": 0.67038739,
      "num_input_tokens_seen": 155921105,
      "step": 7207,
      "time_per_iteration": 2.693228006362915
    },
    {
      "auxiliary_loss_clip": 0.01172972,
      "auxiliary_loss_mlp": 0.01026139,
      "balance_loss_clip": 1.04620743,
      "balance_loss_mlp": 1.01866448,
      "epoch": 0.8667107557265676,
      "flos": 18656419495680.0,
      "grad_norm": 3.3740624064627576,
      "language_loss": 0.71228021,
      "learning_rate": 1.8336322819397677e-07,
      "loss": 0.73427135,
      "num_input_tokens_seen": 155938640,
      "step": 7208,
      "time_per_iteration": 2.566537380218506
    },
    {
      "auxiliary_loss_clip": 0.011784,
      "auxiliary_loss_mlp": 0.0102549,
      "balance_loss_clip": 1.04068112,
      "balance_loss_mlp": 1.01820278,
      "epoch": 0.8668309986172068,
      "flos": 20083725302400.0,
      "grad_norm": 2.827490338688224,
      "language_loss": 0.62644768,
      "learning_rate": 1.8303753815044654e-07,
      "loss": 0.64848661,
      "num_input_tokens_seen": 155957945,
      "step": 7209,
      "time_per_iteration": 2.713981866836548
    },
    {
      "auxiliary_loss_clip": 0.01180455,
      "auxiliary_loss_mlp": 0.01025476,
      "balance_loss_clip": 1.04433393,
      "balance_loss_mlp": 1.01745296,
      "epoch": 0.8669512415078459,
      "flos": 21615099788160.0,
      "grad_norm": 2.831794948208614,
      "language_loss": 0.70948493,
      "learning_rate": 1.827121237386773e-07,
      "loss": 0.73154426,
      "num_input_tokens_seen": 155975390,
      "step": 7210,
      "time_per_iteration": 2.7489302158355713
    },
    {
      "auxiliary_loss_clip": 0.01175255,
      "auxiliary_loss_mlp": 0.0102961,
      "balance_loss_clip": 1.04489887,
      "balance_loss_mlp": 1.02199829,
      "epoch": 0.8670714843984849,
      "flos": 17703601372800.0,
      "grad_norm": 3.0653366493432994,
      "language_loss": 0.75409865,
      "learning_rate": 1.8238698500803374e-07,
      "loss": 0.77614725,
      "num_input_tokens_seen": 155988155,
      "step": 7211,
      "time_per_iteration": 2.666839599609375
    },
    {
      "auxiliary_loss_clip": 0.01070922,
      "auxiliary_loss_mlp": 0.01001838,
      "balance_loss_clip": 1.01062346,
      "balance_loss_mlp": 1.00102699,
      "epoch": 0.8671917272891241,
      "flos": 60705483125760.0,
      "grad_norm": 0.7146664363997064,
      "language_loss": 0.56251347,
      "learning_rate": 1.820621220078391e-07,
      "loss": 0.58324111,
      "num_input_tokens_seen": 156052065,
      "step": 7212,
      "time_per_iteration": 3.2311594486236572
    },
    {
      "auxiliary_loss_clip": 0.01166442,
      "auxiliary_loss_mlp": 0.01021097,
      "balance_loss_clip": 1.04671192,
      "balance_loss_mlp": 1.01337266,
      "epoch": 0.8673119701797631,
      "flos": 20451881750400.0,
      "grad_norm": 1.7237913491500851,
      "language_loss": 0.68019974,
      "learning_rate": 1.8173753478737553e-07,
      "loss": 0.70207512,
      "num_input_tokens_seen": 156072500,
      "step": 7213,
      "time_per_iteration": 3.4327077865600586
    },
    {
      "auxiliary_loss_clip": 0.0117104,
      "auxiliary_loss_mlp": 0.01025655,
      "balance_loss_clip": 1.04889679,
      "balance_loss_mlp": 1.01816833,
      "epoch": 0.8674322130704022,
      "flos": 19647410797440.0,
      "grad_norm": 4.955772168042472,
      "language_loss": 0.8016817,
      "learning_rate": 1.8141322339588205e-07,
      "loss": 0.82364863,
      "num_input_tokens_seen": 156089840,
      "step": 7214,
      "time_per_iteration": 2.6122171878814697
    },
    {
      "auxiliary_loss_clip": 0.01167618,
      "auxiliary_loss_mlp": 0.01025005,
      "balance_loss_clip": 1.04891109,
      "balance_loss_mlp": 1.01776862,
      "epoch": 0.8675524559610414,
      "flos": 26025001367040.0,
      "grad_norm": 9.942059940034865,
      "language_loss": 0.70179379,
      "learning_rate": 1.810891878825569e-07,
      "loss": 0.72372001,
      "num_input_tokens_seen": 156109815,
      "step": 7215,
      "time_per_iteration": 3.5990395545959473
    },
    {
      "auxiliary_loss_clip": 0.01169133,
      "auxiliary_loss_mlp": 0.01023169,
      "balance_loss_clip": 1.04102457,
      "balance_loss_mlp": 1.01558685,
      "epoch": 0.8676726988516804,
      "flos": 15049444584960.0,
      "grad_norm": 2.0490422906463825,
      "language_loss": 0.71867466,
      "learning_rate": 1.8076542829655561e-07,
      "loss": 0.74059767,
      "num_input_tokens_seen": 156128620,
      "step": 7216,
      "time_per_iteration": 2.5785324573516846
    },
    {
      "auxiliary_loss_clip": 0.01172653,
      "auxiliary_loss_mlp": 0.0102563,
      "balance_loss_clip": 1.04614353,
      "balance_loss_mlp": 1.01732707,
      "epoch": 0.8677929417423195,
      "flos": 16288111140480.0,
      "grad_norm": 4.291639258123586,
      "language_loss": 0.79606783,
      "learning_rate": 1.8044194468699203e-07,
      "loss": 0.81805074,
      "num_input_tokens_seen": 156145930,
      "step": 7217,
      "time_per_iteration": 2.6183624267578125
    },
    {
      "auxiliary_loss_clip": 0.01168621,
      "auxiliary_loss_mlp": 0.01024565,
      "balance_loss_clip": 1.0461576,
      "balance_loss_mlp": 1.01721239,
      "epoch": 0.8679131846329585,
      "flos": 18844160906880.0,
      "grad_norm": 3.8140507889508863,
      "language_loss": 0.75811559,
      "learning_rate": 1.8011873710293912e-07,
      "loss": 0.78004742,
      "num_input_tokens_seen": 156164435,
      "step": 7218,
      "time_per_iteration": 2.602316379547119
    },
    {
      "auxiliary_loss_clip": 0.01165084,
      "auxiliary_loss_mlp": 0.01023436,
      "balance_loss_clip": 1.04522562,
      "balance_loss_mlp": 1.01575899,
      "epoch": 0.8680334275235977,
      "flos": 33620718890880.0,
      "grad_norm": 1.8727475268100675,
      "language_loss": 0.69794631,
      "learning_rate": 1.7979580559342677e-07,
      "loss": 0.71983147,
      "num_input_tokens_seen": 156185165,
      "step": 7219,
      "time_per_iteration": 3.637603282928467
    },
    {
      "auxiliary_loss_clip": 0.01170437,
      "auxiliary_loss_mlp": 0.01024568,
      "balance_loss_clip": 1.04498053,
      "balance_loss_mlp": 1.01732588,
      "epoch": 0.8681536704142367,
      "flos": 24681152810880.0,
      "grad_norm": 2.755425137499788,
      "language_loss": 0.66683471,
      "learning_rate": 1.7947315020744358e-07,
      "loss": 0.68878472,
      "num_input_tokens_seen": 156206260,
      "step": 7220,
      "time_per_iteration": 2.669787883758545
    },
    {
      "auxiliary_loss_clip": 0.01171097,
      "auxiliary_loss_mlp": 0.01020977,
      "balance_loss_clip": 1.04406381,
      "balance_loss_mlp": 1.0135504,
      "epoch": 0.8682739133048758,
      "flos": 20011042131840.0,
      "grad_norm": 2.350499520742462,
      "language_loss": 0.80369759,
      "learning_rate": 1.7915077099393594e-07,
      "loss": 0.82561833,
      "num_input_tokens_seen": 156222860,
      "step": 7221,
      "time_per_iteration": 2.6419999599456787
    },
    {
      "auxiliary_loss_clip": 0.01171877,
      "auxiliary_loss_mlp": 0.01027389,
      "balance_loss_clip": 1.04494953,
      "balance_loss_mlp": 1.01967883,
      "epoch": 0.868394156195515,
      "flos": 16654759217280.0,
      "grad_norm": 1.7770184617078442,
      "language_loss": 0.73243058,
      "learning_rate": 1.788286680018083e-07,
      "loss": 0.75442326,
      "num_input_tokens_seen": 156241570,
      "step": 7222,
      "time_per_iteration": 2.6154088973999023
    },
    {
      "auxiliary_loss_clip": 0.01176154,
      "auxiliary_loss_mlp": 0.01025076,
      "balance_loss_clip": 1.04521024,
      "balance_loss_mlp": 1.01801574,
      "epoch": 0.868514399086154,
      "flos": 28001381448960.0,
      "grad_norm": 1.816623440942801,
      "language_loss": 0.72731566,
      "learning_rate": 1.7850684127992443e-07,
      "loss": 0.74932796,
      "num_input_tokens_seen": 156261315,
      "step": 7223,
      "time_per_iteration": 2.6634814739227295
    },
    {
      "auxiliary_loss_clip": 0.01175663,
      "auxiliary_loss_mlp": 0.01025492,
      "balance_loss_clip": 1.04559982,
      "balance_loss_mlp": 1.01800549,
      "epoch": 0.8686346419767931,
      "flos": 20084587228800.0,
      "grad_norm": 1.69120158776189,
      "language_loss": 0.70597118,
      "learning_rate": 1.7818529087710378e-07,
      "loss": 0.7279827,
      "num_input_tokens_seen": 156281670,
      "step": 7224,
      "time_per_iteration": 2.684068202972412
    },
    {
      "auxiliary_loss_clip": 0.01166336,
      "auxiliary_loss_mlp": 0.01207681,
      "balance_loss_clip": 1.04405236,
      "balance_loss_mlp": 1.00060868,
      "epoch": 0.8687548848674322,
      "flos": 18223516782720.0,
      "grad_norm": 1.7803180742181457,
      "language_loss": 0.84060729,
      "learning_rate": 1.7786401684212637e-07,
      "loss": 0.86434746,
      "num_input_tokens_seen": 156300500,
      "step": 7225,
      "time_per_iteration": 2.613680601119995
    },
    {
      "auxiliary_loss_clip": 0.01080008,
      "auxiliary_loss_mlp": 0.01000153,
      "balance_loss_clip": 1.01271558,
      "balance_loss_mlp": 0.99949706,
      "epoch": 0.8688751277580713,
      "flos": 70457885049600.0,
      "grad_norm": 0.742026027451984,
      "language_loss": 0.55958784,
      "learning_rate": 1.7754301922372883e-07,
      "loss": 0.58038944,
      "num_input_tokens_seen": 156350145,
      "step": 7226,
      "time_per_iteration": 3.0499775409698486
    },
    {
      "auxiliary_loss_clip": 0.01184354,
      "auxiliary_loss_mlp": 0.01025564,
      "balance_loss_clip": 1.04060459,
      "balance_loss_mlp": 1.01783371,
      "epoch": 0.8689953706487104,
      "flos": 26906788344960.0,
      "grad_norm": 1.8981591583857693,
      "language_loss": 0.80969071,
      "learning_rate": 1.7722229807060617e-07,
      "loss": 0.83178991,
      "num_input_tokens_seen": 156368725,
      "step": 7227,
      "time_per_iteration": 2.7131454944610596
    },
    {
      "auxiliary_loss_clip": 0.01163952,
      "auxiliary_loss_mlp": 0.01023034,
      "balance_loss_clip": 1.03895152,
      "balance_loss_mlp": 1.01596475,
      "epoch": 0.8691156135393495,
      "flos": 34637385438720.0,
      "grad_norm": 2.230393453855591,
      "language_loss": 0.81964236,
      "learning_rate": 1.7690185343141172e-07,
      "loss": 0.8415122,
      "num_input_tokens_seen": 156388640,
      "step": 7228,
      "time_per_iteration": 2.7527735233306885
    },
    {
      "auxiliary_loss_clip": 0.01170671,
      "auxiliary_loss_mlp": 0.01021909,
      "balance_loss_clip": 1.04257441,
      "balance_loss_mlp": 1.01510811,
      "epoch": 0.8692358564299886,
      "flos": 18989814556800.0,
      "grad_norm": 2.2958498989877705,
      "language_loss": 0.69570667,
      "learning_rate": 1.7658168535475615e-07,
      "loss": 0.71763247,
      "num_input_tokens_seen": 156406425,
      "step": 7229,
      "time_per_iteration": 2.599151134490967
    },
    {
      "auxiliary_loss_clip": 0.0117747,
      "auxiliary_loss_mlp": 0.01025935,
      "balance_loss_clip": 1.04654312,
      "balance_loss_mlp": 1.01835942,
      "epoch": 0.8693560993206276,
      "flos": 30370839039360.0,
      "grad_norm": 1.6202816564681484,
      "language_loss": 0.64396369,
      "learning_rate": 1.7626179388920948e-07,
      "loss": 0.66599774,
      "num_input_tokens_seen": 156427705,
      "step": 7230,
      "time_per_iteration": 2.751286029815674
    },
    {
      "auxiliary_loss_clip": 0.01171242,
      "auxiliary_loss_mlp": 0.01207269,
      "balance_loss_clip": 1.04503608,
      "balance_loss_mlp": 1.00073409,
      "epoch": 0.8694763422112668,
      "flos": 27200430028800.0,
      "grad_norm": 1.7493545887727602,
      "language_loss": 0.80771756,
      "learning_rate": 1.7594217908329866e-07,
      "loss": 0.83150268,
      "num_input_tokens_seen": 156449890,
      "step": 7231,
      "time_per_iteration": 2.6548070907592773
    },
    {
      "auxiliary_loss_clip": 0.01163608,
      "auxiliary_loss_mlp": 0.01021244,
      "balance_loss_clip": 1.04390919,
      "balance_loss_mlp": 1.0139811,
      "epoch": 0.8695965851019059,
      "flos": 26139161767680.0,
      "grad_norm": 1.9612173120177492,
      "language_loss": 0.74338657,
      "learning_rate": 1.7562284098550895e-07,
      "loss": 0.76523513,
      "num_input_tokens_seen": 156469600,
      "step": 7232,
      "time_per_iteration": 3.587873697280884
    },
    {
      "auxiliary_loss_clip": 0.01072156,
      "auxiliary_loss_mlp": 0.01000113,
      "balance_loss_clip": 1.01289845,
      "balance_loss_mlp": 0.99929619,
      "epoch": 0.8697168279925449,
      "flos": 67332616456320.0,
      "grad_norm": 0.837362816753486,
      "language_loss": 0.62220812,
      "learning_rate": 1.753037796442838e-07,
      "loss": 0.64293075,
      "num_input_tokens_seen": 156529040,
      "step": 7233,
      "time_per_iteration": 3.1525590419769287
    },
    {
      "auxiliary_loss_clip": 0.01166866,
      "auxiliary_loss_mlp": 0.01022681,
      "balance_loss_clip": 1.04649305,
      "balance_loss_mlp": 1.01464593,
      "epoch": 0.8698370708831841,
      "flos": 19718693337600.0,
      "grad_norm": 6.59577645198553,
      "language_loss": 0.75143468,
      "learning_rate": 1.74984995108024e-07,
      "loss": 0.77333009,
      "num_input_tokens_seen": 156546970,
      "step": 7234,
      "time_per_iteration": 2.5490007400512695
    },
    {
      "auxiliary_loss_clip": 0.01171174,
      "auxiliary_loss_mlp": 0.01021374,
      "balance_loss_clip": 1.04605258,
      "balance_loss_mlp": 1.0141921,
      "epoch": 0.8699573137738231,
      "flos": 12859971068160.0,
      "grad_norm": 1.9566925963722144,
      "language_loss": 0.83637142,
      "learning_rate": 1.7466648742508981e-07,
      "loss": 0.85829687,
      "num_input_tokens_seen": 156563155,
      "step": 7235,
      "time_per_iteration": 2.474792718887329
    },
    {
      "auxiliary_loss_clip": 0.01171628,
      "auxiliary_loss_mlp": 0.01026749,
      "balance_loss_clip": 1.04527938,
      "balance_loss_mlp": 1.01907158,
      "epoch": 0.8700775566644622,
      "flos": 17420733768960.0,
      "grad_norm": 1.798265956257319,
      "language_loss": 0.84496963,
      "learning_rate": 1.7434825664379837e-07,
      "loss": 0.86695337,
      "num_input_tokens_seen": 156581660,
      "step": 7236,
      "time_per_iteration": 2.511645555496216
    },
    {
      "auxiliary_loss_clip": 0.01171387,
      "auxiliary_loss_mlp": 0.01021767,
      "balance_loss_clip": 1.04629588,
      "balance_loss_mlp": 1.01410222,
      "epoch": 0.8701977995551013,
      "flos": 13735221770880.0,
      "grad_norm": 7.543871017277182,
      "language_loss": 0.86347193,
      "learning_rate": 1.740303028124246e-07,
      "loss": 0.88540351,
      "num_input_tokens_seen": 156597720,
      "step": 7237,
      "time_per_iteration": 2.5212230682373047
    },
    {
      "auxiliary_loss_clip": 0.0116925,
      "auxiliary_loss_mlp": 0.01022065,
      "balance_loss_clip": 1.03661764,
      "balance_loss_mlp": 1.01490617,
      "epoch": 0.8703180424457404,
      "flos": 30555707362560.0,
      "grad_norm": 1.8427742616423792,
      "language_loss": 0.75855595,
      "learning_rate": 1.7371262597920212e-07,
      "loss": 0.78046906,
      "num_input_tokens_seen": 156619780,
      "step": 7238,
      "time_per_iteration": 2.7050514221191406
    },
    {
      "auxiliary_loss_clip": 0.01173044,
      "auxiliary_loss_mlp": 0.01030652,
      "balance_loss_clip": 1.04201674,
      "balance_loss_mlp": 1.0231688,
      "epoch": 0.8704382853363795,
      "flos": 19608986223360.0,
      "grad_norm": 1.5419336947853362,
      "language_loss": 0.76550114,
      "learning_rate": 1.7339522619232195e-07,
      "loss": 0.78753817,
      "num_input_tokens_seen": 156638160,
      "step": 7239,
      "time_per_iteration": 2.7178077697753906
    },
    {
      "auxiliary_loss_clip": 0.01178625,
      "auxiliary_loss_mlp": 0.01024642,
      "balance_loss_clip": 1.04305649,
      "balance_loss_mlp": 1.01664937,
      "epoch": 0.8705585282270186,
      "flos": 26613900846720.0,
      "grad_norm": 2.059381780124815,
      "language_loss": 0.7551533,
      "learning_rate": 1.730781034999338e-07,
      "loss": 0.77718604,
      "num_input_tokens_seen": 156659740,
      "step": 7240,
      "time_per_iteration": 3.5720036029815674
    },
    {
      "auxiliary_loss_clip": 0.01163848,
      "auxiliary_loss_mlp": 0.01025407,
      "balance_loss_clip": 1.04885912,
      "balance_loss_mlp": 1.01802814,
      "epoch": 0.8706787711176577,
      "flos": 34090465979520.0,
      "grad_norm": 1.8851264603118392,
      "language_loss": 0.7367205,
      "learning_rate": 1.7276125795014497e-07,
      "loss": 0.75861305,
      "num_input_tokens_seen": 156678190,
      "step": 7241,
      "time_per_iteration": 3.599154472351074
    },
    {
      "auxiliary_loss_clip": 0.01174803,
      "auxiliary_loss_mlp": 0.01024435,
      "balance_loss_clip": 1.04213727,
      "balance_loss_mlp": 1.01654947,
      "epoch": 0.8707990140082967,
      "flos": 14611513968000.0,
      "grad_norm": 2.556938860565998,
      "language_loss": 0.67502105,
      "learning_rate": 1.7244468959102054e-07,
      "loss": 0.69701338,
      "num_input_tokens_seen": 156695245,
      "step": 7242,
      "time_per_iteration": 2.576009750366211
    },
    {
      "auxiliary_loss_clip": 0.01170969,
      "auxiliary_loss_mlp": 0.01019841,
      "balance_loss_clip": 1.04763412,
      "balance_loss_mlp": 1.01231849,
      "epoch": 0.8709192568989359,
      "flos": 20084156265600.0,
      "grad_norm": 2.100109295765189,
      "language_loss": 0.85596132,
      "learning_rate": 1.7212839847058348e-07,
      "loss": 0.87786943,
      "num_input_tokens_seen": 156710375,
      "step": 7243,
      "time_per_iteration": 2.5840487480163574
    },
    {
      "auxiliary_loss_clip": 0.01186369,
      "auxiliary_loss_mlp": 0.01023807,
      "balance_loss_clip": 1.03903544,
      "balance_loss_mlp": 1.01708353,
      "epoch": 0.871039499789575,
      "flos": 16727083251840.0,
      "grad_norm": 4.273186473765322,
      "language_loss": 0.73893094,
      "learning_rate": 1.718123846368147e-07,
      "loss": 0.7610327,
      "num_input_tokens_seen": 156729420,
      "step": 7244,
      "time_per_iteration": 2.710348129272461
    },
    {
      "auxiliary_loss_clip": 0.01173005,
      "auxiliary_loss_mlp": 0.01207442,
      "balance_loss_clip": 1.04726505,
      "balance_loss_mlp": 1.00069928,
      "epoch": 0.871159742680214,
      "flos": 21068790860160.0,
      "grad_norm": 2.413060858510725,
      "language_loss": 0.71456957,
      "learning_rate": 1.714966481376543e-07,
      "loss": 0.73837399,
      "num_input_tokens_seen": 156746100,
      "step": 7245,
      "time_per_iteration": 2.559544324874878
    },
    {
      "auxiliary_loss_clip": 0.0116821,
      "auxiliary_loss_mlp": 0.01026042,
      "balance_loss_clip": 1.04434717,
      "balance_loss_mlp": 1.0188421,
      "epoch": 0.8712799855708532,
      "flos": 28256526731520.0,
      "grad_norm": 3.144497321237223,
      "language_loss": 0.83048826,
      "learning_rate": 1.7118118902099797e-07,
      "loss": 0.85243076,
      "num_input_tokens_seen": 156764185,
      "step": 7246,
      "time_per_iteration": 3.600125551223755
    },
    {
      "auxiliary_loss_clip": 0.01170141,
      "auxiliary_loss_mlp": 0.01030207,
      "balance_loss_clip": 1.04660058,
      "balance_loss_mlp": 1.02278066,
      "epoch": 0.8714002284614922,
      "flos": 22236677665920.0,
      "grad_norm": 2.5154762687189023,
      "language_loss": 0.8066045,
      "learning_rate": 1.7086600733470146e-07,
      "loss": 0.82860798,
      "num_input_tokens_seen": 156784855,
      "step": 7247,
      "time_per_iteration": 2.6311962604522705
    },
    {
      "auxiliary_loss_clip": 0.01165012,
      "auxiliary_loss_mlp": 0.01024915,
      "balance_loss_clip": 1.04477429,
      "balance_loss_mlp": 1.01789999,
      "epoch": 0.8715204713521313,
      "flos": 21431919404160.0,
      "grad_norm": 1.7843829139636282,
      "language_loss": 0.7719239,
      "learning_rate": 1.7055110312657738e-07,
      "loss": 0.79382324,
      "num_input_tokens_seen": 156804350,
      "step": 7248,
      "time_per_iteration": 2.6427063941955566
    },
    {
      "auxiliary_loss_clip": 0.01167477,
      "auxiliary_loss_mlp": 0.0102714,
      "balance_loss_clip": 1.04483819,
      "balance_loss_mlp": 1.01923621,
      "epoch": 0.8716407142427703,
      "flos": 23440439180160.0,
      "grad_norm": 2.294765463545443,
      "language_loss": 0.7439974,
      "learning_rate": 1.702364764443962e-07,
      "loss": 0.76594365,
      "num_input_tokens_seen": 156823425,
      "step": 7249,
      "time_per_iteration": 2.7245519161224365
    },
    {
      "auxiliary_loss_clip": 0.01175426,
      "auxiliary_loss_mlp": 0.01022108,
      "balance_loss_clip": 1.03720605,
      "balance_loss_mlp": 1.01421356,
      "epoch": 0.8717609571334095,
      "flos": 27958683156480.0,
      "grad_norm": 1.9685229643323314,
      "language_loss": 0.72336781,
      "learning_rate": 1.6992212733588685e-07,
      "loss": 0.74534321,
      "num_input_tokens_seen": 156843090,
      "step": 7250,
      "time_per_iteration": 2.767129421234131
    },
    {
      "auxiliary_loss_clip": 0.01168879,
      "auxiliary_loss_mlp": 0.01024009,
      "balance_loss_clip": 1.04357338,
      "balance_loss_mlp": 1.01657915,
      "epoch": 0.8718812000240486,
      "flos": 25479482538240.0,
      "grad_norm": 2.3008344624102697,
      "language_loss": 0.75166631,
      "learning_rate": 1.6960805584873538e-07,
      "loss": 0.77359509,
      "num_input_tokens_seen": 156861090,
      "step": 7251,
      "time_per_iteration": 2.8096859455108643
    },
    {
      "auxiliary_loss_clip": 0.01179234,
      "auxiliary_loss_mlp": 0.01023666,
      "balance_loss_clip": 1.03968167,
      "balance_loss_mlp": 1.01673079,
      "epoch": 0.8720014429146876,
      "flos": 23403056100480.0,
      "grad_norm": 2.0806776976270513,
      "language_loss": 0.78248858,
      "learning_rate": 1.6929426203058684e-07,
      "loss": 0.80451763,
      "num_input_tokens_seen": 156881515,
      "step": 7252,
      "time_per_iteration": 2.7966060638427734
    },
    {
      "auxiliary_loss_clip": 0.01170181,
      "auxiliary_loss_mlp": 0.01208328,
      "balance_loss_clip": 1.04632723,
      "balance_loss_mlp": 1.0006789,
      "epoch": 0.8721216858053268,
      "flos": 24352821567360.0,
      "grad_norm": 3.6107798934658977,
      "language_loss": 0.7969662,
      "learning_rate": 1.689807459290431e-07,
      "loss": 0.82075131,
      "num_input_tokens_seen": 156900170,
      "step": 7253,
      "time_per_iteration": 2.6608994007110596
    },
    {
      "auxiliary_loss_clip": 0.01170381,
      "auxiliary_loss_mlp": 0.01024858,
      "balance_loss_clip": 1.04282475,
      "balance_loss_mlp": 1.01804245,
      "epoch": 0.8722419286959658,
      "flos": 33869687034240.0,
      "grad_norm": 2.1336060849192164,
      "language_loss": 0.70991945,
      "learning_rate": 1.6866750759166437e-07,
      "loss": 0.73187178,
      "num_input_tokens_seen": 156920150,
      "step": 7254,
      "time_per_iteration": 2.7876815795898438
    },
    {
      "auxiliary_loss_clip": 0.01169023,
      "auxiliary_loss_mlp": 0.01021887,
      "balance_loss_clip": 1.03889108,
      "balance_loss_mlp": 1.0141145,
      "epoch": 0.8723621715866049,
      "flos": 18369385914240.0,
      "grad_norm": 2.4223477399125164,
      "language_loss": 0.77111185,
      "learning_rate": 1.6835454706596865e-07,
      "loss": 0.7930209,
      "num_input_tokens_seen": 156937980,
      "step": 7255,
      "time_per_iteration": 2.7947888374328613
    },
    {
      "auxiliary_loss_clip": 0.01167506,
      "auxiliary_loss_mlp": 0.01026816,
      "balance_loss_clip": 1.04784584,
      "balance_loss_mlp": 1.01932931,
      "epoch": 0.8724824144772441,
      "flos": 22013348855040.0,
      "grad_norm": 1.7064416781488025,
      "language_loss": 0.73654664,
      "learning_rate": 1.680418643994317e-07,
      "loss": 0.75848985,
      "num_input_tokens_seen": 156956550,
      "step": 7256,
      "time_per_iteration": 2.6204028129577637
    },
    {
      "auxiliary_loss_clip": 0.01063046,
      "auxiliary_loss_mlp": 0.01000305,
      "balance_loss_clip": 1.01033318,
      "balance_loss_mlp": 0.99951249,
      "epoch": 0.8726026573678831,
      "flos": 66698720213760.0,
      "grad_norm": 0.8912602280628109,
      "language_loss": 0.64495409,
      "learning_rate": 1.6772945963948738e-07,
      "loss": 0.66558754,
      "num_input_tokens_seen": 157014715,
      "step": 7257,
      "time_per_iteration": 3.168891191482544
    },
    {
      "auxiliary_loss_clip": 0.01167105,
      "auxiliary_loss_mlp": 0.01025069,
      "balance_loss_clip": 1.04435861,
      "balance_loss_mlp": 1.01754665,
      "epoch": 0.8727229002585222,
      "flos": 13370908078080.0,
      "grad_norm": 2.2271190638308735,
      "language_loss": 0.77392828,
      "learning_rate": 1.6741733283352733e-07,
      "loss": 0.79584998,
      "num_input_tokens_seen": 157032320,
      "step": 7258,
      "time_per_iteration": 2.819324254989624
    },
    {
      "auxiliary_loss_clip": 0.0118391,
      "auxiliary_loss_mlp": 0.01027848,
      "balance_loss_clip": 1.04365826,
      "balance_loss_mlp": 1.02046943,
      "epoch": 0.8728431431491613,
      "flos": 21796987282560.0,
      "grad_norm": 1.7738966257516229,
      "language_loss": 0.83977842,
      "learning_rate": 1.6710548402890102e-07,
      "loss": 0.86189604,
      "num_input_tokens_seen": 157052845,
      "step": 7259,
      "time_per_iteration": 3.656242847442627
    },
    {
      "auxiliary_loss_clip": 0.01171625,
      "auxiliary_loss_mlp": 0.0102367,
      "balance_loss_clip": 1.04800797,
      "balance_loss_mlp": 1.01576686,
      "epoch": 0.8729633860398004,
      "flos": 36173823742080.0,
      "grad_norm": 2.054098951547381,
      "language_loss": 0.66891462,
      "learning_rate": 1.6679391327291527e-07,
      "loss": 0.69086754,
      "num_input_tokens_seen": 157074050,
      "step": 7260,
      "time_per_iteration": 2.664679527282715
    },
    {
      "auxiliary_loss_clip": 0.01171813,
      "auxiliary_loss_mlp": 0.01026742,
      "balance_loss_clip": 1.0427897,
      "balance_loss_mlp": 1.01959848,
      "epoch": 0.8730836289304394,
      "flos": 16359680989440.0,
      "grad_norm": 3.0569222522594286,
      "language_loss": 0.67866659,
      "learning_rate": 1.6648262061283492e-07,
      "loss": 0.70065218,
      "num_input_tokens_seen": 157089350,
      "step": 7261,
      "time_per_iteration": 2.5704774856567383
    },
    {
      "auxiliary_loss_clip": 0.01172634,
      "auxiliary_loss_mlp": 0.01023544,
      "balance_loss_clip": 1.03966451,
      "balance_loss_mlp": 1.01662421,
      "epoch": 0.8732038718210786,
      "flos": 21215126868480.0,
      "grad_norm": 2.7637268705098075,
      "language_loss": 0.73920846,
      "learning_rate": 1.6617160609588353e-07,
      "loss": 0.76117027,
      "num_input_tokens_seen": 157108525,
      "step": 7262,
      "time_per_iteration": 2.6112453937530518
    },
    {
      "auxiliary_loss_clip": 0.01177518,
      "auxiliary_loss_mlp": 0.01025341,
      "balance_loss_clip": 1.04517138,
      "balance_loss_mlp": 1.01745546,
      "epoch": 0.8733241147117177,
      "flos": 16610696208000.0,
      "grad_norm": 3.202248185499551,
      "language_loss": 0.72134179,
      "learning_rate": 1.6586086976924163e-07,
      "loss": 0.74337041,
      "num_input_tokens_seen": 157124025,
      "step": 7263,
      "time_per_iteration": 2.5490875244140625
    },
    {
      "auxiliary_loss_clip": 0.01167766,
      "auxiliary_loss_mlp": 0.01022406,
      "balance_loss_clip": 1.04386163,
      "balance_loss_mlp": 1.01536703,
      "epoch": 0.8734443576023567,
      "flos": 20193935207040.0,
      "grad_norm": 2.107608404810007,
      "language_loss": 0.78404844,
      "learning_rate": 1.6555041168004747e-07,
      "loss": 0.80595016,
      "num_input_tokens_seen": 157143345,
      "step": 7264,
      "time_per_iteration": 2.481076240539551
    },
    {
      "auxiliary_loss_clip": 0.01166728,
      "auxiliary_loss_mlp": 0.01023737,
      "balance_loss_clip": 1.04354596,
      "balance_loss_mlp": 1.01674557,
      "epoch": 0.8735646004929959,
      "flos": 18041162411520.0,
      "grad_norm": 1.9913116273007794,
      "language_loss": 0.68919837,
      "learning_rate": 1.6524023187539715e-07,
      "loss": 0.71110308,
      "num_input_tokens_seen": 157161630,
      "step": 7265,
      "time_per_iteration": 2.519596815109253
    },
    {
      "auxiliary_loss_clip": 0.01173016,
      "auxiliary_loss_mlp": 0.01023421,
      "balance_loss_clip": 1.04428434,
      "balance_loss_mlp": 1.01622653,
      "epoch": 0.873684843383635,
      "flos": 20262344659200.0,
      "grad_norm": 1.8004252126648312,
      "language_loss": 0.7493692,
      "learning_rate": 1.649303304023446e-07,
      "loss": 0.77133358,
      "num_input_tokens_seen": 157181385,
      "step": 7266,
      "time_per_iteration": 3.3108105659484863
    },
    {
      "auxiliary_loss_clip": 0.01169822,
      "auxiliary_loss_mlp": 0.01020739,
      "balance_loss_clip": 1.04349291,
      "balance_loss_mlp": 1.01336563,
      "epoch": 0.873805086274274,
      "flos": 16947287579520.0,
      "grad_norm": 2.2206864879725083,
      "language_loss": 0.78794599,
      "learning_rate": 1.6462070730790246e-07,
      "loss": 0.80985159,
      "num_input_tokens_seen": 157200545,
      "step": 7267,
      "time_per_iteration": 2.609184503555298
    },
    {
      "auxiliary_loss_clip": 0.01166632,
      "auxiliary_loss_mlp": 0.01024249,
      "balance_loss_clip": 1.04046988,
      "balance_loss_mlp": 1.0166254,
      "epoch": 0.8739253291649132,
      "flos": 18041270152320.0,
      "grad_norm": 16.319917217947236,
      "language_loss": 0.79041713,
      "learning_rate": 1.6431136263903912e-07,
      "loss": 0.81232595,
      "num_input_tokens_seen": 157219545,
      "step": 7268,
      "time_per_iteration": 3.466888427734375
    },
    {
      "auxiliary_loss_clip": 0.01172683,
      "auxiliary_loss_mlp": 0.01207343,
      "balance_loss_clip": 1.04428816,
      "balance_loss_mlp": 1.00070953,
      "epoch": 0.8740455720555522,
      "flos": 21325085377920.0,
      "grad_norm": 5.257612726967932,
      "language_loss": 0.73748851,
      "learning_rate": 1.6400229644268282e-07,
      "loss": 0.76128876,
      "num_input_tokens_seen": 157237900,
      "step": 7269,
      "time_per_iteration": 2.6111598014831543
    },
    {
      "auxiliary_loss_clip": 0.01167038,
      "auxiliary_loss_mlp": 0.01026593,
      "balance_loss_clip": 1.04432809,
      "balance_loss_mlp": 1.01870394,
      "epoch": 0.8741658149461913,
      "flos": 15158684822400.0,
      "grad_norm": 2.2081572339119497,
      "language_loss": 0.80986273,
      "learning_rate": 1.6369350876571852e-07,
      "loss": 0.83179903,
      "num_input_tokens_seen": 157256055,
      "step": 7270,
      "time_per_iteration": 2.596433639526367
    },
    {
      "auxiliary_loss_clip": 0.01171165,
      "auxiliary_loss_mlp": 0.01022819,
      "balance_loss_clip": 1.03874576,
      "balance_loss_mlp": 1.01532686,
      "epoch": 0.8742860578368304,
      "flos": 23039855729280.0,
      "grad_norm": 2.240314444584283,
      "language_loss": 0.81325507,
      "learning_rate": 1.6338499965498874e-07,
      "loss": 0.83519495,
      "num_input_tokens_seen": 157274785,
      "step": 7271,
      "time_per_iteration": 2.6793501377105713
    },
    {
      "auxiliary_loss_clip": 0.01168754,
      "auxiliary_loss_mlp": 0.01026987,
      "balance_loss_clip": 1.04194129,
      "balance_loss_mlp": 1.01908684,
      "epoch": 0.8744063007274695,
      "flos": 28145347159680.0,
      "grad_norm": 2.3167532365297094,
      "language_loss": 0.77448779,
      "learning_rate": 1.630767691572943e-07,
      "loss": 0.79644525,
      "num_input_tokens_seen": 157294805,
      "step": 7272,
      "time_per_iteration": 2.699564218521118
    },
    {
      "auxiliary_loss_clip": 0.0107832,
      "auxiliary_loss_mlp": 0.01001684,
      "balance_loss_clip": 1.01112914,
      "balance_loss_mlp": 1.00091493,
      "epoch": 0.8745265436181086,
      "flos": 64034076654720.0,
      "grad_norm": 0.746503508825896,
      "language_loss": 0.5350244,
      "learning_rate": 1.6276881731939306e-07,
      "loss": 0.5558244,
      "num_input_tokens_seen": 157356695,
      "step": 7273,
      "time_per_iteration": 4.122012615203857
    },
    {
      "auxiliary_loss_clip": 0.01166764,
      "auxiliary_loss_mlp": 0.01021714,
      "balance_loss_clip": 1.0468154,
      "balance_loss_mlp": 1.01433468,
      "epoch": 0.8746467865087477,
      "flos": 28658618553600.0,
      "grad_norm": 2.07722208407423,
      "language_loss": 0.75046945,
      "learning_rate": 1.6246114418800193e-07,
      "loss": 0.77235425,
      "num_input_tokens_seen": 157376975,
      "step": 7274,
      "time_per_iteration": 2.679231882095337
    },
    {
      "auxiliary_loss_clip": 0.01163953,
      "auxiliary_loss_mlp": 0.01026978,
      "balance_loss_clip": 1.0446806,
      "balance_loss_mlp": 1.01909244,
      "epoch": 0.8747670293993868,
      "flos": 23985850268160.0,
      "grad_norm": 1.8302156240067406,
      "language_loss": 0.76898623,
      "learning_rate": 1.6215374980979423e-07,
      "loss": 0.79089558,
      "num_input_tokens_seen": 157397385,
      "step": 7275,
      "time_per_iteration": 2.6085915565490723
    },
    {
      "auxiliary_loss_clip": 0.01163687,
      "auxiliary_loss_mlp": 0.01028072,
      "balance_loss_clip": 1.04632807,
      "balance_loss_mlp": 1.02126229,
      "epoch": 0.8748872722900258,
      "flos": 45221624478720.0,
      "grad_norm": 5.680243344951327,
      "language_loss": 0.69110155,
      "learning_rate": 1.6184663423140133e-07,
      "loss": 0.71301919,
      "num_input_tokens_seen": 157417685,
      "step": 7276,
      "time_per_iteration": 2.755587339401245
    },
    {
      "auxiliary_loss_clip": 0.01179306,
      "auxiliary_loss_mlp": 0.01029214,
      "balance_loss_clip": 1.04229927,
      "balance_loss_mlp": 1.02189493,
      "epoch": 0.875007515180665,
      "flos": 19754280737280.0,
      "grad_norm": 2.1272560085438923,
      "language_loss": 0.64394546,
      "learning_rate": 1.615397974994126e-07,
      "loss": 0.66603065,
      "num_input_tokens_seen": 157435490,
      "step": 7277,
      "time_per_iteration": 2.660926342010498
    },
    {
      "auxiliary_loss_clip": 0.01164418,
      "auxiliary_loss_mlp": 0.01024068,
      "balance_loss_clip": 1.04771852,
      "balance_loss_mlp": 1.01711249,
      "epoch": 0.875127758071304,
      "flos": 22710734386560.0,
      "grad_norm": 1.5158191791785285,
      "language_loss": 0.80765975,
      "learning_rate": 1.6123323966037438e-07,
      "loss": 0.82954466,
      "num_input_tokens_seen": 157454010,
      "step": 7278,
      "time_per_iteration": 2.5753173828125
    },
    {
      "auxiliary_loss_clip": 0.01166849,
      "auxiliary_loss_mlp": 0.01028513,
      "balance_loss_clip": 1.04825568,
      "balance_loss_mlp": 1.02114868,
      "epoch": 0.8752480009619431,
      "flos": 23403846199680.0,
      "grad_norm": 2.5418547262731233,
      "language_loss": 0.7888993,
      "learning_rate": 1.6092696076079216e-07,
      "loss": 0.81085294,
      "num_input_tokens_seen": 157472385,
      "step": 7279,
      "time_per_iteration": 2.572460412979126
    },
    {
      "auxiliary_loss_clip": 0.01162541,
      "auxiliary_loss_mlp": 0.01021476,
      "balance_loss_clip": 1.0410217,
      "balance_loss_mlp": 1.01423454,
      "epoch": 0.8753682438525822,
      "flos": 26213101914240.0,
      "grad_norm": 1.7837792752686998,
      "language_loss": 0.74095601,
      "learning_rate": 1.6062096084712785e-07,
      "loss": 0.76279616,
      "num_input_tokens_seen": 157493735,
      "step": 7280,
      "time_per_iteration": 2.683140277862549
    },
    {
      "auxiliary_loss_clip": 0.01163011,
      "auxiliary_loss_mlp": 0.01207324,
      "balance_loss_clip": 1.04032981,
      "balance_loss_mlp": 1.00085998,
      "epoch": 0.8754884867432213,
      "flos": 23326745656320.0,
      "grad_norm": 1.6734210860650083,
      "language_loss": 0.70699745,
      "learning_rate": 1.6031523996580098e-07,
      "loss": 0.73070085,
      "num_input_tokens_seen": 157511295,
      "step": 7281,
      "time_per_iteration": 2.674173593521118
    },
    {
      "auxiliary_loss_clip": 0.01183677,
      "auxiliary_loss_mlp": 0.01025012,
      "balance_loss_clip": 1.04494298,
      "balance_loss_mlp": 1.0173887,
      "epoch": 0.8756087296338604,
      "flos": 12495226412160.0,
      "grad_norm": 2.345854485694051,
      "language_loss": 0.66241956,
      "learning_rate": 1.6000979816318981e-07,
      "loss": 0.68450642,
      "num_input_tokens_seen": 157529760,
      "step": 7282,
      "time_per_iteration": 2.5931077003479004
    },
    {
      "auxiliary_loss_clip": 0.01161241,
      "auxiliary_loss_mlp": 0.01022144,
      "balance_loss_clip": 1.04438889,
      "balance_loss_mlp": 1.01477098,
      "epoch": 0.8757289725244994,
      "flos": 18952898353920.0,
      "grad_norm": 9.64707199575436,
      "language_loss": 0.74926913,
      "learning_rate": 1.5970463548562886e-07,
      "loss": 0.77110302,
      "num_input_tokens_seen": 157548915,
      "step": 7283,
      "time_per_iteration": 2.615353584289551
    },
    {
      "auxiliary_loss_clip": 0.01169822,
      "auxiliary_loss_mlp": 0.01022149,
      "balance_loss_clip": 1.04471695,
      "balance_loss_mlp": 1.01521981,
      "epoch": 0.8758492154151386,
      "flos": 25265958140160.0,
      "grad_norm": 1.9118018785291135,
      "language_loss": 0.71230066,
      "learning_rate": 1.5939975197941192e-07,
      "loss": 0.73422033,
      "num_input_tokens_seen": 157570570,
      "step": 7284,
      "time_per_iteration": 2.6302077770233154
    },
    {
      "auxiliary_loss_clip": 0.01077546,
      "auxiliary_loss_mlp": 0.01002693,
      "balance_loss_clip": 1.01034164,
      "balance_loss_mlp": 1.0019536,
      "epoch": 0.8759694583057777,
      "flos": 65571664193280.0,
      "grad_norm": 0.8186696564091919,
      "language_loss": 0.5335958,
      "learning_rate": 1.5909514769078892e-07,
      "loss": 0.55439818,
      "num_input_tokens_seen": 157635675,
      "step": 7285,
      "time_per_iteration": 3.227816343307495
    },
    {
      "auxiliary_loss_clip": 0.01163771,
      "auxiliary_loss_mlp": 0.01024125,
      "balance_loss_clip": 1.04327595,
      "balance_loss_mlp": 1.01704395,
      "epoch": 0.8760897011964167,
      "flos": 25446193608960.0,
      "grad_norm": 1.577411954293477,
      "language_loss": 0.77534777,
      "learning_rate": 1.5879082266596867e-07,
      "loss": 0.79722679,
      "num_input_tokens_seen": 157657015,
      "step": 7286,
      "time_per_iteration": 3.6290693283081055
    },
    {
      "auxiliary_loss_clip": 0.01163349,
      "auxiliary_loss_mlp": 0.01021571,
      "balance_loss_clip": 1.03928435,
      "balance_loss_mlp": 1.01393843,
      "epoch": 0.8762099440870559,
      "flos": 28984830894720.0,
      "grad_norm": 2.016863892132037,
      "language_loss": 0.72096944,
      "learning_rate": 1.5848677695111645e-07,
      "loss": 0.74281865,
      "num_input_tokens_seen": 157678615,
      "step": 7287,
      "time_per_iteration": 2.684361696243286
    },
    {
      "auxiliary_loss_clip": 0.01180224,
      "auxiliary_loss_mlp": 0.0102623,
      "balance_loss_clip": 1.0438261,
      "balance_loss_mlp": 1.0185585,
      "epoch": 0.8763301869776949,
      "flos": 21609461352960.0,
      "grad_norm": 2.2201717425495384,
      "language_loss": 0.69582665,
      "learning_rate": 1.5818301059235562e-07,
      "loss": 0.71789116,
      "num_input_tokens_seen": 157693790,
      "step": 7288,
      "time_per_iteration": 2.7487738132476807
    },
    {
      "auxiliary_loss_clip": 0.01173967,
      "auxiliary_loss_mlp": 0.01022655,
      "balance_loss_clip": 1.04703677,
      "balance_loss_mlp": 1.01521945,
      "epoch": 0.876450429868334,
      "flos": 24644416176000.0,
      "grad_norm": 1.9598522102737044,
      "language_loss": 0.81326187,
      "learning_rate": 1.578795236357684e-07,
      "loss": 0.83522809,
      "num_input_tokens_seen": 157715255,
      "step": 7289,
      "time_per_iteration": 2.6329939365386963
    },
    {
      "auxiliary_loss_clip": 0.01172744,
      "auxiliary_loss_mlp": 0.01021704,
      "balance_loss_clip": 1.04619718,
      "balance_loss_mlp": 1.01459885,
      "epoch": 0.8765706727589732,
      "flos": 20260046188800.0,
      "grad_norm": 2.2784639567507106,
      "language_loss": 0.85807335,
      "learning_rate": 1.5757631612739218e-07,
      "loss": 0.88001776,
      "num_input_tokens_seen": 157728800,
      "step": 7290,
      "time_per_iteration": 2.6019065380096436
    },
    {
      "auxiliary_loss_clip": 0.01062758,
      "auxiliary_loss_mlp": 0.01001985,
      "balance_loss_clip": 1.01007175,
      "balance_loss_mlp": 1.0012517,
      "epoch": 0.8766909156496122,
      "flos": 71371165276800.0,
      "grad_norm": 0.7950866230886253,
      "language_loss": 0.61421436,
      "learning_rate": 1.572733881132242e-07,
      "loss": 0.63486177,
      "num_input_tokens_seen": 157789445,
      "step": 7291,
      "time_per_iteration": 3.232935667037964
    },
    {
      "auxiliary_loss_clip": 0.01079278,
      "auxiliary_loss_mlp": 0.00999538,
      "balance_loss_clip": 1.01299477,
      "balance_loss_mlp": 0.99876302,
      "epoch": 0.8768111585402513,
      "flos": 69523490603520.0,
      "grad_norm": 0.7794097366080731,
      "language_loss": 0.58481205,
      "learning_rate": 1.5697073963921814e-07,
      "loss": 0.60560018,
      "num_input_tokens_seen": 157848685,
      "step": 7292,
      "time_per_iteration": 3.0636954307556152
    },
    {
      "auxiliary_loss_clip": 0.01173544,
      "auxiliary_loss_mlp": 0.01018982,
      "balance_loss_clip": 1.04879165,
      "balance_loss_mlp": 1.01149011,
      "epoch": 0.8769314014308904,
      "flos": 18838558385280.0,
      "grad_norm": 2.5490263182139223,
      "language_loss": 0.84829402,
      "learning_rate": 1.566683707512857e-07,
      "loss": 0.87021923,
      "num_input_tokens_seen": 157866360,
      "step": 7293,
      "time_per_iteration": 3.4331271648406982
    },
    {
      "auxiliary_loss_clip": 0.01167406,
      "auxiliary_loss_mlp": 0.01026899,
      "balance_loss_clip": 1.04306865,
      "balance_loss_mlp": 1.01894212,
      "epoch": 0.8770516443215295,
      "flos": 14976402278400.0,
      "grad_norm": 2.1737267322403877,
      "language_loss": 0.79314208,
      "learning_rate": 1.5636628149529553e-07,
      "loss": 0.81508517,
      "num_input_tokens_seen": 157884150,
      "step": 7294,
      "time_per_iteration": 2.607283592224121
    },
    {
      "auxiliary_loss_clip": 0.01167141,
      "auxiliary_loss_mlp": 0.0102169,
      "balance_loss_clip": 1.04173756,
      "balance_loss_mlp": 1.01465118,
      "epoch": 0.8771718872121685,
      "flos": 31649654021760.0,
      "grad_norm": 2.1269405005505333,
      "language_loss": 0.79879445,
      "learning_rate": 1.560644719170743e-07,
      "loss": 0.82068276,
      "num_input_tokens_seen": 157905020,
      "step": 7295,
      "time_per_iteration": 3.5869665145874023
    },
    {
      "auxiliary_loss_clip": 0.01172481,
      "auxiliary_loss_mlp": 0.0102411,
      "balance_loss_clip": 1.041538,
      "balance_loss_mlp": 1.01602721,
      "epoch": 0.8772921301028077,
      "flos": 36095466222720.0,
      "grad_norm": 1.9610966578605862,
      "language_loss": 0.72396123,
      "learning_rate": 1.5576294206240692e-07,
      "loss": 0.74592716,
      "num_input_tokens_seen": 157924545,
      "step": 7296,
      "time_per_iteration": 2.7822883129119873
    },
    {
      "auxiliary_loss_clip": 0.01165404,
      "auxiliary_loss_mlp": 0.01023669,
      "balance_loss_clip": 1.04264843,
      "balance_loss_mlp": 1.01654625,
      "epoch": 0.8774123729934468,
      "flos": 57116961849600.0,
      "grad_norm": 1.7055729376854079,
      "language_loss": 0.67590415,
      "learning_rate": 1.5546169197703507e-07,
      "loss": 0.69779491,
      "num_input_tokens_seen": 157950820,
      "step": 7297,
      "time_per_iteration": 3.000260829925537
    },
    {
      "auxiliary_loss_clip": 0.01175946,
      "auxiliary_loss_mlp": 0.01027318,
      "balance_loss_clip": 1.04210603,
      "balance_loss_mlp": 1.02007294,
      "epoch": 0.8775326158840858,
      "flos": 23914495900800.0,
      "grad_norm": 2.5879737610531013,
      "language_loss": 0.77748835,
      "learning_rate": 1.5516072170665774e-07,
      "loss": 0.79952097,
      "num_input_tokens_seen": 157968790,
      "step": 7298,
      "time_per_iteration": 2.6551804542541504
    },
    {
      "auxiliary_loss_clip": 0.01169744,
      "auxiliary_loss_mlp": 0.01020765,
      "balance_loss_clip": 1.04576254,
      "balance_loss_mlp": 1.01358294,
      "epoch": 0.877652858774725,
      "flos": 17123285243520.0,
      "grad_norm": 2.4307263781120736,
      "language_loss": 0.87088668,
      "learning_rate": 1.5486003129693214e-07,
      "loss": 0.89279175,
      "num_input_tokens_seen": 157986155,
      "step": 7299,
      "time_per_iteration": 3.606801986694336
    },
    {
      "auxiliary_loss_clip": 0.01170652,
      "auxiliary_loss_mlp": 0.01020557,
      "balance_loss_clip": 1.04693639,
      "balance_loss_mlp": 1.01327288,
      "epoch": 0.877773101665364,
      "flos": 16508961912960.0,
      "grad_norm": 1.9540250174029299,
      "language_loss": 0.78339756,
      "learning_rate": 1.545596207934725e-07,
      "loss": 0.80530971,
      "num_input_tokens_seen": 158004640,
      "step": 7300,
      "time_per_iteration": 2.6634371280670166
    },
    {
      "auxiliary_loss_clip": 0.01164217,
      "auxiliary_loss_mlp": 0.01021476,
      "balance_loss_clip": 1.04277587,
      "balance_loss_mlp": 1.01407015,
      "epoch": 0.8778933445560031,
      "flos": 22053209973120.0,
      "grad_norm": 1.9649698803307696,
      "language_loss": 0.77603912,
      "learning_rate": 1.5425949024185147e-07,
      "loss": 0.79789603,
      "num_input_tokens_seen": 158024665,
      "step": 7301,
      "time_per_iteration": 2.6541922092437744
    },
    {
      "auxiliary_loss_clip": 0.01173628,
      "auxiliary_loss_mlp": 0.01022616,
      "balance_loss_clip": 1.04364705,
      "balance_loss_mlp": 1.01557684,
      "epoch": 0.8780135874466423,
      "flos": 22564757514240.0,
      "grad_norm": 1.86608392094648,
      "language_loss": 0.67818308,
      "learning_rate": 1.5395963968759818e-07,
      "loss": 0.7001456,
      "num_input_tokens_seen": 158044940,
      "step": 7302,
      "time_per_iteration": 2.7025411128997803
    },
    {
      "auxiliary_loss_clip": 0.01170904,
      "auxiliary_loss_mlp": 0.01020485,
      "balance_loss_clip": 1.04165435,
      "balance_loss_mlp": 1.01324892,
      "epoch": 0.8781338303372813,
      "flos": 61531999073280.0,
      "grad_norm": 1.5102175761621737,
      "language_loss": 0.64197481,
      "learning_rate": 1.536600691761998e-07,
      "loss": 0.66388869,
      "num_input_tokens_seen": 158070770,
      "step": 7303,
      "time_per_iteration": 2.9232068061828613
    },
    {
      "auxiliary_loss_clip": 0.01177592,
      "auxiliary_loss_mlp": 0.01025485,
      "balance_loss_clip": 1.04556727,
      "balance_loss_mlp": 1.01880348,
      "epoch": 0.8782540732279204,
      "flos": 22674751937280.0,
      "grad_norm": 1.7929673928552383,
      "language_loss": 0.71722209,
      "learning_rate": 1.5336077875310084e-07,
      "loss": 0.73925281,
      "num_input_tokens_seen": 158089995,
      "step": 7304,
      "time_per_iteration": 2.6784415245056152
    },
    {
      "auxiliary_loss_clip": 0.01181368,
      "auxiliary_loss_mlp": 0.01021337,
      "balance_loss_clip": 1.04278708,
      "balance_loss_mlp": 1.01418126,
      "epoch": 0.8783743161185595,
      "flos": 16070348937600.0,
      "grad_norm": 2.0307381806251685,
      "language_loss": 0.73789144,
      "learning_rate": 1.5306176846370321e-07,
      "loss": 0.75991845,
      "num_input_tokens_seen": 158108140,
      "step": 7305,
      "time_per_iteration": 2.667680025100708
    },
    {
      "auxiliary_loss_clip": 0.01178252,
      "auxiliary_loss_mlp": 0.01032101,
      "balance_loss_clip": 1.04340565,
      "balance_loss_mlp": 1.02448332,
      "epoch": 0.8784945590091986,
      "flos": 26067879227520.0,
      "grad_norm": 2.1083928428705816,
      "language_loss": 0.73913836,
      "learning_rate": 1.5276303835336712e-07,
      "loss": 0.76124191,
      "num_input_tokens_seen": 158128680,
      "step": 7306,
      "time_per_iteration": 2.6539864540100098
    },
    {
      "auxiliary_loss_clip": 0.01069971,
      "auxiliary_loss_mlp": 0.01000937,
      "balance_loss_clip": 1.00973248,
      "balance_loss_mlp": 1.00018907,
      "epoch": 0.8786148018998376,
      "flos": 62720643939840.0,
      "grad_norm": 0.760212276720951,
      "language_loss": 0.5351606,
      "learning_rate": 1.524645884674094e-07,
      "loss": 0.5558697,
      "num_input_tokens_seen": 158185610,
      "step": 7307,
      "time_per_iteration": 3.1435940265655518
    },
    {
      "auxiliary_loss_clip": 0.01166916,
      "auxiliary_loss_mlp": 0.01208483,
      "balance_loss_clip": 1.04571247,
      "balance_loss_mlp": 1.0006671,
      "epoch": 0.8787350447904768,
      "flos": 21652734263040.0,
      "grad_norm": 3.8745667253857383,
      "language_loss": 0.79444015,
      "learning_rate": 1.521664188511047e-07,
      "loss": 0.81819415,
      "num_input_tokens_seen": 158205635,
      "step": 7308,
      "time_per_iteration": 2.7659404277801514
    },
    {
      "auxiliary_loss_clip": 0.01171243,
      "auxiliary_loss_mlp": 0.01207339,
      "balance_loss_clip": 1.04705143,
      "balance_loss_mlp": 1.00077343,
      "epoch": 0.8788552876811159,
      "flos": 25478476957440.0,
      "grad_norm": 2.093175258663539,
      "language_loss": 0.80419946,
      "learning_rate": 1.518685295496851e-07,
      "loss": 0.82798529,
      "num_input_tokens_seen": 158223495,
      "step": 7309,
      "time_per_iteration": 2.729419469833374
    },
    {
      "auxiliary_loss_clip": 0.01168868,
      "auxiliary_loss_mlp": 0.01025241,
      "balance_loss_clip": 1.04384935,
      "balance_loss_mlp": 1.01821709,
      "epoch": 0.8789755305717549,
      "flos": 22310222762880.0,
      "grad_norm": 1.6784427124980998,
      "language_loss": 0.85484529,
      "learning_rate": 1.5157092060833975e-07,
      "loss": 0.87678641,
      "num_input_tokens_seen": 158243145,
      "step": 7310,
      "time_per_iteration": 2.5894246101379395
    },
    {
      "auxiliary_loss_clip": 0.01169489,
      "auxiliary_loss_mlp": 0.0101808,
      "balance_loss_clip": 1.0429467,
      "balance_loss_mlp": 1.01112676,
      "epoch": 0.879095773462394,
      "flos": 29310971408640.0,
      "grad_norm": 2.1988872055492283,
      "language_loss": 0.66171014,
      "learning_rate": 1.5127359207221658e-07,
      "loss": 0.68358582,
      "num_input_tokens_seen": 158262625,
      "step": 7311,
      "time_per_iteration": 2.654534339904785
    },
    {
      "auxiliary_loss_clip": 0.01168565,
      "auxiliary_loss_mlp": 0.01023098,
      "balance_loss_clip": 1.03542805,
      "balance_loss_mlp": 1.01515293,
      "epoch": 0.8792160163530331,
      "flos": 16690023394560.0,
      "grad_norm": 1.8367335114790349,
      "language_loss": 0.73449308,
      "learning_rate": 1.5097654398641923e-07,
      "loss": 0.7564097,
      "num_input_tokens_seen": 158280530,
      "step": 7312,
      "time_per_iteration": 3.6239142417907715
    },
    {
      "auxiliary_loss_clip": 0.0117374,
      "auxiliary_loss_mlp": 0.01025967,
      "balance_loss_clip": 1.04726338,
      "balance_loss_mlp": 1.01843321,
      "epoch": 0.8793362592436722,
      "flos": 24499301230080.0,
      "grad_norm": 1.6139449274378788,
      "language_loss": 0.73100483,
      "learning_rate": 1.5067977639601014e-07,
      "loss": 0.75300193,
      "num_input_tokens_seen": 158303290,
      "step": 7313,
      "time_per_iteration": 2.632854700088501
    },
    {
      "auxiliary_loss_clip": 0.01165859,
      "auxiliary_loss_mlp": 0.01024793,
      "balance_loss_clip": 1.0428946,
      "balance_loss_mlp": 1.01737273,
      "epoch": 0.8794565021343113,
      "flos": 14538399834240.0,
      "grad_norm": 2.245416975650776,
      "language_loss": 0.71070731,
      "learning_rate": 1.5038328934600864e-07,
      "loss": 0.7326138,
      "num_input_tokens_seen": 158319925,
      "step": 7314,
      "time_per_iteration": 2.584162950515747
    },
    {
      "auxiliary_loss_clip": 0.01168951,
      "auxiliary_loss_mlp": 0.01023461,
      "balance_loss_clip": 1.04443264,
      "balance_loss_mlp": 1.01657677,
      "epoch": 0.8795767450249504,
      "flos": 39530286224640.0,
      "grad_norm": 2.386862116316859,
      "language_loss": 0.69695479,
      "learning_rate": 1.5008708288139161e-07,
      "loss": 0.71887892,
      "num_input_tokens_seen": 158342285,
      "step": 7315,
      "time_per_iteration": 2.7600886821746826
    },
    {
      "auxiliary_loss_clip": 0.01167079,
      "auxiliary_loss_mlp": 0.0102454,
      "balance_loss_clip": 1.04565001,
      "balance_loss_mlp": 1.01692557,
      "epoch": 0.8796969879155895,
      "flos": 22960672197120.0,
      "grad_norm": 2.2543851651085203,
      "language_loss": 0.73101878,
      "learning_rate": 1.497911570470931e-07,
      "loss": 0.75293499,
      "num_input_tokens_seen": 158362290,
      "step": 7316,
      "time_per_iteration": 2.6248650550842285
    },
    {
      "auxiliary_loss_clip": 0.01164011,
      "auxiliary_loss_mlp": 0.01029448,
      "balance_loss_clip": 1.04261255,
      "balance_loss_mlp": 1.02205718,
      "epoch": 0.8798172308062285,
      "flos": 28362427004160.0,
      "grad_norm": 1.7605754389869106,
      "language_loss": 0.85546124,
      "learning_rate": 1.494955118880048e-07,
      "loss": 0.87739587,
      "num_input_tokens_seen": 158383275,
      "step": 7317,
      "time_per_iteration": 2.6614420413970947
    },
    {
      "auxiliary_loss_clip": 0.01169402,
      "auxiliary_loss_mlp": 0.01026065,
      "balance_loss_clip": 1.04518795,
      "balance_loss_mlp": 1.0187608,
      "epoch": 0.8799374736968677,
      "flos": 23988974751360.0,
      "grad_norm": 2.1652101032714497,
      "language_loss": 0.73001254,
      "learning_rate": 1.4920014744897634e-07,
      "loss": 0.75196725,
      "num_input_tokens_seen": 158402690,
      "step": 7318,
      "time_per_iteration": 2.6190693378448486
    },
    {
      "auxiliary_loss_clip": 0.01162909,
      "auxiliary_loss_mlp": 0.01020947,
      "balance_loss_clip": 1.04349065,
      "balance_loss_mlp": 1.01347876,
      "epoch": 0.8800577165875068,
      "flos": 25630271832960.0,
      "grad_norm": 2.02021361146477,
      "language_loss": 0.86332381,
      "learning_rate": 1.4890506377481392e-07,
      "loss": 0.88516235,
      "num_input_tokens_seen": 158421780,
      "step": 7319,
      "time_per_iteration": 2.61780047416687
    },
    {
      "auxiliary_loss_clip": 0.01175134,
      "auxiliary_loss_mlp": 0.01026816,
      "balance_loss_clip": 1.04025805,
      "balance_loss_mlp": 1.0203073,
      "epoch": 0.8801779594781458,
      "flos": 23440331439360.0,
      "grad_norm": 1.6300073674738824,
      "language_loss": 0.63919759,
      "learning_rate": 1.486102609102815e-07,
      "loss": 0.66121709,
      "num_input_tokens_seen": 158442330,
      "step": 7320,
      "time_per_iteration": 3.5344901084899902
    },
    {
      "auxiliary_loss_clip": 0.01162077,
      "auxiliary_loss_mlp": 0.01020777,
      "balance_loss_clip": 1.04169941,
      "balance_loss_mlp": 1.01364517,
      "epoch": 0.880298202368785,
      "flos": 11508580656000.0,
      "grad_norm": 2.476485739156584,
      "language_loss": 0.85555947,
      "learning_rate": 1.483157389001004e-07,
      "loss": 0.877388,
      "num_input_tokens_seen": 158459890,
      "step": 7321,
      "time_per_iteration": 3.326072931289673
    },
    {
      "auxiliary_loss_clip": 0.0117162,
      "auxiliary_loss_mlp": 0.01023683,
      "balance_loss_clip": 1.04212832,
      "balance_loss_mlp": 1.0153625,
      "epoch": 0.880418445259424,
      "flos": 22671447886080.0,
      "grad_norm": 2.4968659243945517,
      "language_loss": 0.79194921,
      "learning_rate": 1.4802149778894933e-07,
      "loss": 0.81390226,
      "num_input_tokens_seen": 158478680,
      "step": 7322,
      "time_per_iteration": 2.6793811321258545
    },
    {
      "auxiliary_loss_clip": 0.01157666,
      "auxiliary_loss_mlp": 0.01021307,
      "balance_loss_clip": 1.04078865,
      "balance_loss_mlp": 1.01466155,
      "epoch": 0.8805386881500631,
      "flos": 20522158709760.0,
      "grad_norm": 1.7727967042604935,
      "language_loss": 0.87353176,
      "learning_rate": 1.4772753762146484e-07,
      "loss": 0.89532155,
      "num_input_tokens_seen": 158497935,
      "step": 7323,
      "time_per_iteration": 2.633235216140747
    },
    {
      "auxiliary_loss_clip": 0.011628,
      "auxiliary_loss_mlp": 0.01021715,
      "balance_loss_clip": 1.04375911,
      "balance_loss_mlp": 1.01386237,
      "epoch": 0.8806589310407023,
      "flos": 36538891620480.0,
      "grad_norm": 1.674763527943844,
      "language_loss": 0.70823592,
      "learning_rate": 1.474338584422401e-07,
      "loss": 0.73008108,
      "num_input_tokens_seen": 158523145,
      "step": 7324,
      "time_per_iteration": 2.700808525085449
    },
    {
      "auxiliary_loss_clip": 0.01164641,
      "auxiliary_loss_mlp": 0.01020527,
      "balance_loss_clip": 1.04546642,
      "balance_loss_mlp": 1.01319301,
      "epoch": 0.8807791739313413,
      "flos": 23440187784960.0,
      "grad_norm": 1.7609265605805364,
      "language_loss": 0.75913548,
      "learning_rate": 1.4714046029582595e-07,
      "loss": 0.78098708,
      "num_input_tokens_seen": 158542210,
      "step": 7325,
      "time_per_iteration": 2.655251979827881
    },
    {
      "auxiliary_loss_clip": 0.01175544,
      "auxiliary_loss_mlp": 0.01021552,
      "balance_loss_clip": 1.04238462,
      "balance_loss_mlp": 1.01416135,
      "epoch": 0.8808994168219804,
      "flos": 25956843310080.0,
      "grad_norm": 2.4251013536362214,
      "language_loss": 0.75476104,
      "learning_rate": 1.46847343226731e-07,
      "loss": 0.77673197,
      "num_input_tokens_seen": 158563250,
      "step": 7326,
      "time_per_iteration": 3.556759834289551
    },
    {
      "auxiliary_loss_clip": 0.01169494,
      "auxiliary_loss_mlp": 0.01023166,
      "balance_loss_clip": 1.04354525,
      "balance_loss_mlp": 1.01543808,
      "epoch": 0.8810196597126195,
      "flos": 17092079303040.0,
      "grad_norm": 3.1693631608062454,
      "language_loss": 0.69665313,
      "learning_rate": 1.465545072794203e-07,
      "loss": 0.71857977,
      "num_input_tokens_seen": 158581125,
      "step": 7327,
      "time_per_iteration": 2.581181526184082
    },
    {
      "auxiliary_loss_clip": 0.01171019,
      "auxiliary_loss_mlp": 0.01023438,
      "balance_loss_clip": 1.04220748,
      "balance_loss_mlp": 1.01605582,
      "epoch": 0.8811399026032586,
      "flos": 23002831785600.0,
      "grad_norm": 1.5867893878833608,
      "language_loss": 0.75824285,
      "learning_rate": 1.4626195249831774e-07,
      "loss": 0.78018743,
      "num_input_tokens_seen": 158602025,
      "step": 7328,
      "time_per_iteration": 2.69834303855896
    },
    {
      "auxiliary_loss_clip": 0.01165702,
      "auxiliary_loss_mlp": 0.01023855,
      "balance_loss_clip": 1.04396069,
      "balance_loss_mlp": 1.01672316,
      "epoch": 0.8812601454938976,
      "flos": 14463813242880.0,
      "grad_norm": 1.9487677071733562,
      "language_loss": 0.71953994,
      "learning_rate": 1.4596967892780244e-07,
      "loss": 0.74143547,
      "num_input_tokens_seen": 158618355,
      "step": 7329,
      "time_per_iteration": 2.5906050205230713
    },
    {
      "auxiliary_loss_clip": 0.01163713,
      "auxiliary_loss_mlp": 0.01023767,
      "balance_loss_clip": 1.04625273,
      "balance_loss_mlp": 1.01648939,
      "epoch": 0.8813803883845368,
      "flos": 22493223578880.0,
      "grad_norm": 1.810421280718181,
      "language_loss": 0.74793035,
      "learning_rate": 1.4567768661221314e-07,
      "loss": 0.76980519,
      "num_input_tokens_seen": 158638925,
      "step": 7330,
      "time_per_iteration": 2.62907338142395
    },
    {
      "auxiliary_loss_clip": 0.01173849,
      "auxiliary_loss_mlp": 0.01207738,
      "balance_loss_clip": 1.04806185,
      "balance_loss_mlp": 1.00069237,
      "epoch": 0.8815006312751759,
      "flos": 21506901045120.0,
      "grad_norm": 2.250314035796842,
      "language_loss": 0.75176698,
      "learning_rate": 1.4538597559584442e-07,
      "loss": 0.77558285,
      "num_input_tokens_seen": 158656715,
      "step": 7331,
      "time_per_iteration": 2.6253085136413574
    },
    {
      "auxiliary_loss_clip": 0.011681,
      "auxiliary_loss_mlp": 0.01028018,
      "balance_loss_clip": 1.04334521,
      "balance_loss_mlp": 1.0201025,
      "epoch": 0.8816208741658149,
      "flos": 22784566792320.0,
      "grad_norm": 1.9686574534255958,
      "language_loss": 0.7886349,
      "learning_rate": 1.4509454592294823e-07,
      "loss": 0.81059605,
      "num_input_tokens_seen": 158677200,
      "step": 7332,
      "time_per_iteration": 2.665950298309326
    },
    {
      "auxiliary_loss_clip": 0.0117566,
      "auxiliary_loss_mlp": 0.01207661,
      "balance_loss_clip": 1.04508936,
      "balance_loss_mlp": 1.00067997,
      "epoch": 0.8817411170564541,
      "flos": 17779409026560.0,
      "grad_norm": 3.6858808553039983,
      "language_loss": 0.79159743,
      "learning_rate": 1.448033976377354e-07,
      "loss": 0.81543064,
      "num_input_tokens_seen": 158692185,
      "step": 7333,
      "time_per_iteration": 2.5676276683807373
    },
    {
      "auxiliary_loss_clip": 0.01168982,
      "auxiliary_loss_mlp": 0.01020127,
      "balance_loss_clip": 1.04376245,
      "balance_loss_mlp": 1.013165,
      "epoch": 0.8818613599470931,
      "flos": 18551812112640.0,
      "grad_norm": 2.186057060342004,
      "language_loss": 0.74387658,
      "learning_rate": 1.445125307843713e-07,
      "loss": 0.76576763,
      "num_input_tokens_seen": 158710410,
      "step": 7334,
      "time_per_iteration": 2.6005098819732666
    },
    {
      "auxiliary_loss_clip": 0.01166337,
      "auxiliary_loss_mlp": 0.01022146,
      "balance_loss_clip": 1.04610348,
      "balance_loss_mlp": 1.01532125,
      "epoch": 0.8819816028377322,
      "flos": 27599792417280.0,
      "grad_norm": 1.7094867283192718,
      "language_loss": 0.75487036,
      "learning_rate": 1.442219454069813e-07,
      "loss": 0.77675515,
      "num_input_tokens_seen": 158731435,
      "step": 7335,
      "time_per_iteration": 2.651862144470215
    },
    {
      "auxiliary_loss_clip": 0.01176376,
      "auxiliary_loss_mlp": 0.0102497,
      "balance_loss_clip": 1.04037714,
      "balance_loss_mlp": 1.0178442,
      "epoch": 0.8821018457283714,
      "flos": 23404600385280.0,
      "grad_norm": 2.4459126759034993,
      "language_loss": 0.66371763,
      "learning_rate": 1.4393164154964676e-07,
      "loss": 0.68573105,
      "num_input_tokens_seen": 158750965,
      "step": 7336,
      "time_per_iteration": 2.741055727005005
    },
    {
      "auxiliary_loss_clip": 0.01164845,
      "auxiliary_loss_mlp": 0.01024212,
      "balance_loss_clip": 1.04581821,
      "balance_loss_mlp": 1.01725638,
      "epoch": 0.8822220886190104,
      "flos": 29132459792640.0,
      "grad_norm": 2.0143054873410224,
      "language_loss": 0.94400388,
      "learning_rate": 1.4364161925640649e-07,
      "loss": 0.96589446,
      "num_input_tokens_seen": 158772365,
      "step": 7337,
      "time_per_iteration": 2.638862133026123
    },
    {
      "auxiliary_loss_clip": 0.01165065,
      "auxiliary_loss_mlp": 0.01021541,
      "balance_loss_clip": 1.04714215,
      "balance_loss_mlp": 1.01451993,
      "epoch": 0.8823423315096495,
      "flos": 20485422074880.0,
      "grad_norm": 2.1610133816055255,
      "language_loss": 0.85051572,
      "learning_rate": 1.4335187857125663e-07,
      "loss": 0.87238181,
      "num_input_tokens_seen": 158791065,
      "step": 7338,
      "time_per_iteration": 2.6482040882110596
    },
    {
      "auxiliary_loss_clip": 0.01169566,
      "auxiliary_loss_mlp": 0.01020271,
      "balance_loss_clip": 1.04554987,
      "balance_loss_mlp": 1.01331544,
      "epoch": 0.8824625744002886,
      "flos": 24206377818240.0,
      "grad_norm": 2.2382747117467794,
      "language_loss": 0.75797856,
      "learning_rate": 1.4306241953815023e-07,
      "loss": 0.77987689,
      "num_input_tokens_seen": 158812125,
      "step": 7339,
      "time_per_iteration": 3.5741097927093506
    },
    {
      "auxiliary_loss_clip": 0.01171993,
      "auxiliary_loss_mlp": 0.01021884,
      "balance_loss_clip": 1.04746974,
      "balance_loss_mlp": 1.0147289,
      "epoch": 0.8825828172909277,
      "flos": 24679500785280.0,
      "grad_norm": 2.134273236128594,
      "language_loss": 0.71040839,
      "learning_rate": 1.4277324220099862e-07,
      "loss": 0.73234713,
      "num_input_tokens_seen": 158834035,
      "step": 7340,
      "time_per_iteration": 2.624271869659424
    },
    {
      "auxiliary_loss_clip": 0.01166948,
      "auxiliary_loss_mlp": 0.01025266,
      "balance_loss_clip": 1.04003,
      "balance_loss_mlp": 1.01827729,
      "epoch": 0.8827030601815667,
      "flos": 22456163721600.0,
      "grad_norm": 2.2882310873674654,
      "language_loss": 0.74304831,
      "learning_rate": 1.4248434660366938e-07,
      "loss": 0.76497048,
      "num_input_tokens_seen": 158853510,
      "step": 7341,
      "time_per_iteration": 2.722785711288452
    },
    {
      "auxiliary_loss_clip": 0.0117001,
      "auxiliary_loss_mlp": 0.01025015,
      "balance_loss_clip": 1.04537606,
      "balance_loss_mlp": 1.01795769,
      "epoch": 0.8828233030722058,
      "flos": 19865639877120.0,
      "grad_norm": 1.9135105208433734,
      "language_loss": 0.70539153,
      "learning_rate": 1.4219573278998808e-07,
      "loss": 0.72734171,
      "num_input_tokens_seen": 158871970,
      "step": 7342,
      "time_per_iteration": 2.730886697769165
    },
    {
      "auxiliary_loss_clip": 0.01168576,
      "auxiliary_loss_mlp": 0.01021055,
      "balance_loss_clip": 1.04118872,
      "balance_loss_mlp": 1.01334214,
      "epoch": 0.882943545962845,
      "flos": 39347213581440.0,
      "grad_norm": 2.075278064329825,
      "language_loss": 0.65082252,
      "learning_rate": 1.4190740080373685e-07,
      "loss": 0.67271882,
      "num_input_tokens_seen": 158892250,
      "step": 7343,
      "time_per_iteration": 2.7624011039733887
    },
    {
      "auxiliary_loss_clip": 0.01177564,
      "auxiliary_loss_mlp": 0.01025486,
      "balance_loss_clip": 1.04409099,
      "balance_loss_mlp": 1.01756477,
      "epoch": 0.883063788853484,
      "flos": 19054524908160.0,
      "grad_norm": 1.879514808462717,
      "language_loss": 0.84246421,
      "learning_rate": 1.4161935068865538e-07,
      "loss": 0.86449468,
      "num_input_tokens_seen": 158907395,
      "step": 7344,
      "time_per_iteration": 2.754115343093872
    },
    {
      "auxiliary_loss_clip": 0.01165453,
      "auxiliary_loss_mlp": 0.01020064,
      "balance_loss_clip": 1.04654217,
      "balance_loss_mlp": 1.01264322,
      "epoch": 0.8831840317441231,
      "flos": 18733196816640.0,
      "grad_norm": 1.974499866958169,
      "language_loss": 0.75552219,
      "learning_rate": 1.4133158248844113e-07,
      "loss": 0.77737737,
      "num_input_tokens_seen": 158926300,
      "step": 7345,
      "time_per_iteration": 2.5824217796325684
    },
    {
      "auxiliary_loss_clip": 0.01177646,
      "auxiliary_loss_mlp": 0.01024967,
      "balance_loss_clip": 1.04369521,
      "balance_loss_mlp": 1.01674128,
      "epoch": 0.8833042746347622,
      "flos": 26827712553600.0,
      "grad_norm": 1.922123343917614,
      "language_loss": 0.73496169,
      "learning_rate": 1.4104409624674785e-07,
      "loss": 0.75698781,
      "num_input_tokens_seen": 158946085,
      "step": 7346,
      "time_per_iteration": 2.6682217121124268
    },
    {
      "auxiliary_loss_clip": 0.01172475,
      "auxiliary_loss_mlp": 0.01019433,
      "balance_loss_clip": 1.04976988,
      "balance_loss_mlp": 1.01272488,
      "epoch": 0.8834245175254013,
      "flos": 26104077158400.0,
      "grad_norm": 1.6765417738449984,
      "language_loss": 0.78587055,
      "learning_rate": 1.407568920071873e-07,
      "loss": 0.80778968,
      "num_input_tokens_seen": 158964950,
      "step": 7347,
      "time_per_iteration": 3.648289442062378
    },
    {
      "auxiliary_loss_clip": 0.01172345,
      "auxiliary_loss_mlp": 0.01024107,
      "balance_loss_clip": 1.04915404,
      "balance_loss_mlp": 1.0162394,
      "epoch": 0.8835447604160404,
      "flos": 30629036977920.0,
      "grad_norm": 2.4901224779759166,
      "language_loss": 0.68653893,
      "learning_rate": 1.4046996981332782e-07,
      "loss": 0.70850343,
      "num_input_tokens_seen": 158984835,
      "step": 7348,
      "time_per_iteration": 3.617302656173706
    },
    {
      "auxiliary_loss_clip": 0.01178049,
      "auxiliary_loss_mlp": 0.01021661,
      "balance_loss_clip": 1.04466033,
      "balance_loss_mlp": 1.01382303,
      "epoch": 0.8836650033066795,
      "flos": 24718356322560.0,
      "grad_norm": 1.8857230731495713,
      "language_loss": 0.78129464,
      "learning_rate": 1.4018332970869516e-07,
      "loss": 0.80329174,
      "num_input_tokens_seen": 159002775,
      "step": 7349,
      "time_per_iteration": 2.666348695755005
    },
    {
      "auxiliary_loss_clip": 0.01162156,
      "auxiliary_loss_mlp": 0.01022474,
      "balance_loss_clip": 1.04258823,
      "balance_loss_mlp": 1.01500845,
      "epoch": 0.8837852461973186,
      "flos": 25413371556480.0,
      "grad_norm": 1.880155931865621,
      "language_loss": 0.85531157,
      "learning_rate": 1.398969717367733e-07,
      "loss": 0.87715793,
      "num_input_tokens_seen": 159024100,
      "step": 7350,
      "time_per_iteration": 2.699498176574707
    },
    {
      "auxiliary_loss_clip": 0.01176904,
      "auxiliary_loss_mlp": 0.01025442,
      "balance_loss_clip": 1.04476333,
      "balance_loss_mlp": 1.01836991,
      "epoch": 0.8839054890879576,
      "flos": 17822574195840.0,
      "grad_norm": 1.7100758042883402,
      "language_loss": 0.76442951,
      "learning_rate": 1.396108959410014e-07,
      "loss": 0.78645295,
      "num_input_tokens_seen": 159043315,
      "step": 7351,
      "time_per_iteration": 2.631516456604004
    },
    {
      "auxiliary_loss_clip": 0.01167702,
      "auxiliary_loss_mlp": 0.01207676,
      "balance_loss_clip": 1.04651797,
      "balance_loss_mlp": 1.00062346,
      "epoch": 0.8840257319785968,
      "flos": 23769021818880.0,
      "grad_norm": 1.9795208711990773,
      "language_loss": 0.81745833,
      "learning_rate": 1.3932510236477745e-07,
      "loss": 0.84121215,
      "num_input_tokens_seen": 159063985,
      "step": 7352,
      "time_per_iteration": 2.641050100326538
    },
    {
      "auxiliary_loss_clip": 0.01167699,
      "auxiliary_loss_mlp": 0.01023557,
      "balance_loss_clip": 1.04369092,
      "balance_loss_mlp": 1.01566577,
      "epoch": 0.8841459748692359,
      "flos": 29059776622080.0,
      "grad_norm": 1.9179123059733758,
      "language_loss": 0.56078589,
      "learning_rate": 1.3903959105145636e-07,
      "loss": 0.5826984,
      "num_input_tokens_seen": 159084475,
      "step": 7353,
      "time_per_iteration": 3.529292583465576
    },
    {
      "auxiliary_loss_clip": 0.01165577,
      "auxiliary_loss_mlp": 0.0102212,
      "balance_loss_clip": 1.04661012,
      "balance_loss_mlp": 1.01482403,
      "epoch": 0.8842662177598749,
      "flos": 24311523905280.0,
      "grad_norm": 2.256918191745554,
      "language_loss": 0.8335703,
      "learning_rate": 1.387543620443492e-07,
      "loss": 0.85544729,
      "num_input_tokens_seen": 159101320,
      "step": 7354,
      "time_per_iteration": 2.645615816116333
    },
    {
      "auxiliary_loss_clip": 0.01166146,
      "auxiliary_loss_mlp": 0.01023877,
      "balance_loss_clip": 1.04772818,
      "balance_loss_mlp": 1.01674867,
      "epoch": 0.8843864606505141,
      "flos": 25007867942400.0,
      "grad_norm": 1.8475514718052217,
      "language_loss": 0.84354919,
      "learning_rate": 1.3846941538672606e-07,
      "loss": 0.86544943,
      "num_input_tokens_seen": 159120025,
      "step": 7355,
      "time_per_iteration": 2.6503794193267822
    },
    {
      "auxiliary_loss_clip": 0.01180541,
      "auxiliary_loss_mlp": 0.01022302,
      "balance_loss_clip": 1.04411626,
      "balance_loss_mlp": 1.01537883,
      "epoch": 0.8845067035411531,
      "flos": 28183915388160.0,
      "grad_norm": 2.318320502163159,
      "language_loss": 0.81007028,
      "learning_rate": 1.3818475112181193e-07,
      "loss": 0.83209872,
      "num_input_tokens_seen": 159138820,
      "step": 7356,
      "time_per_iteration": 2.704979658126831
    },
    {
      "auxiliary_loss_clip": 0.01170831,
      "auxiliary_loss_mlp": 0.01021544,
      "balance_loss_clip": 1.04486156,
      "balance_loss_mlp": 1.01489162,
      "epoch": 0.8846269464317922,
      "flos": 12853219311360.0,
      "grad_norm": 2.194318837350976,
      "language_loss": 0.80249357,
      "learning_rate": 1.3790036929279091e-07,
      "loss": 0.82441735,
      "num_input_tokens_seen": 159155975,
      "step": 7357,
      "time_per_iteration": 2.6640758514404297
    },
    {
      "auxiliary_loss_clip": 0.01171879,
      "auxiliary_loss_mlp": 0.0120799,
      "balance_loss_clip": 1.04746664,
      "balance_loss_mlp": 1.0006963,
      "epoch": 0.8847471893224313,
      "flos": 18624351628800.0,
      "grad_norm": 3.5225252359790193,
      "language_loss": 0.59025717,
      "learning_rate": 1.3761626994280363e-07,
      "loss": 0.61405587,
      "num_input_tokens_seen": 159173445,
      "step": 7358,
      "time_per_iteration": 2.576545000076294
    },
    {
      "auxiliary_loss_clip": 0.01177608,
      "auxiliary_loss_mlp": 0.01022327,
      "balance_loss_clip": 1.04309297,
      "balance_loss_mlp": 1.01521587,
      "epoch": 0.8848674322130704,
      "flos": 35769433449600.0,
      "grad_norm": 2.3910092402582293,
      "language_loss": 0.7379232,
      "learning_rate": 1.3733245311494735e-07,
      "loss": 0.75992256,
      "num_input_tokens_seen": 159196100,
      "step": 7359,
      "time_per_iteration": 2.779780149459839
    },
    {
      "auxiliary_loss_clip": 0.01172029,
      "auxiliary_loss_mlp": 0.01024937,
      "balance_loss_clip": 1.04867613,
      "balance_loss_mlp": 1.01740599,
      "epoch": 0.8849876751037095,
      "flos": 24243760897920.0,
      "grad_norm": 2.1547545019725476,
      "language_loss": 0.70696783,
      "learning_rate": 1.3704891885227676e-07,
      "loss": 0.72893751,
      "num_input_tokens_seen": 159216145,
      "step": 7360,
      "time_per_iteration": 2.641695737838745
    },
    {
      "auxiliary_loss_clip": 0.01175111,
      "auxiliary_loss_mlp": 0.01029142,
      "balance_loss_clip": 1.04160678,
      "balance_loss_mlp": 1.02081561,
      "epoch": 0.8851079179943486,
      "flos": 21500580251520.0,
      "grad_norm": 2.1044632228491698,
      "language_loss": 0.78091377,
      "learning_rate": 1.367656671978037e-07,
      "loss": 0.80295628,
      "num_input_tokens_seen": 159233610,
      "step": 7361,
      "time_per_iteration": 2.6720175743103027
    },
    {
      "auxiliary_loss_clip": 0.0117713,
      "auxiliary_loss_mlp": 0.01024342,
      "balance_loss_clip": 1.04367959,
      "balance_loss_mlp": 1.01730239,
      "epoch": 0.8852281608849877,
      "flos": 15300711198720.0,
      "grad_norm": 2.154575533920338,
      "language_loss": 0.73308665,
      "learning_rate": 1.36482698194498e-07,
      "loss": 0.75510132,
      "num_input_tokens_seen": 159250155,
      "step": 7362,
      "time_per_iteration": 2.6271302700042725
    },
    {
      "auxiliary_loss_clip": 0.01171281,
      "auxiliary_loss_mlp": 0.01024269,
      "balance_loss_clip": 1.04291296,
      "balance_loss_mlp": 1.0164547,
      "epoch": 0.8853484037756267,
      "flos": 23295719283840.0,
      "grad_norm": 1.8169369812520362,
      "language_loss": 0.72137749,
      "learning_rate": 1.3620001188528506e-07,
      "loss": 0.74333298,
      "num_input_tokens_seen": 159270875,
      "step": 7363,
      "time_per_iteration": 2.657651662826538
    },
    {
      "auxiliary_loss_clip": 0.01170293,
      "auxiliary_loss_mlp": 0.01023336,
      "balance_loss_clip": 1.04323041,
      "balance_loss_mlp": 1.0152005,
      "epoch": 0.8854686466662659,
      "flos": 25114773795840.0,
      "grad_norm": 2.5280254400197273,
      "language_loss": 0.73906511,
      "learning_rate": 1.3591760831304865e-07,
      "loss": 0.76100141,
      "num_input_tokens_seen": 159288565,
      "step": 7364,
      "time_per_iteration": 2.6492490768432617
    },
    {
      "auxiliary_loss_clip": 0.01165867,
      "auxiliary_loss_mlp": 0.01023599,
      "balance_loss_clip": 1.04676318,
      "balance_loss_mlp": 1.01599979,
      "epoch": 0.885588889556905,
      "flos": 21390873137280.0,
      "grad_norm": 2.153891279989592,
      "language_loss": 0.79469925,
      "learning_rate": 1.356354875206287e-07,
      "loss": 0.81659395,
      "num_input_tokens_seen": 159306400,
      "step": 7365,
      "time_per_iteration": 2.621051788330078
    },
    {
      "auxiliary_loss_clip": 0.01172756,
      "auxiliary_loss_mlp": 0.01019766,
      "balance_loss_clip": 1.04515982,
      "balance_loss_mlp": 1.01269722,
      "epoch": 0.885709132447544,
      "flos": 26906752431360.0,
      "grad_norm": 2.261895209362663,
      "language_loss": 0.70289654,
      "learning_rate": 1.3535364955082296e-07,
      "loss": 0.72482175,
      "num_input_tokens_seen": 159326250,
      "step": 7366,
      "time_per_iteration": 3.6843135356903076
    },
    {
      "auxiliary_loss_clip": 0.0116321,
      "auxiliary_loss_mlp": 0.01024591,
      "balance_loss_clip": 1.04695535,
      "balance_loss_mlp": 1.01718843,
      "epoch": 0.8858293753381832,
      "flos": 26103394800000.0,
      "grad_norm": 1.940355081367704,
      "language_loss": 0.64837432,
      "learning_rate": 1.3507209444638613e-07,
      "loss": 0.67025232,
      "num_input_tokens_seen": 159348250,
      "step": 7367,
      "time_per_iteration": 2.6214890480041504
    },
    {
      "auxiliary_loss_clip": 0.01168942,
      "auxiliary_loss_mlp": 0.01025531,
      "balance_loss_clip": 1.04615819,
      "balance_loss_mlp": 1.0181756,
      "epoch": 0.8859496182288222,
      "flos": 23292810282240.0,
      "grad_norm": 2.0390606959383764,
      "language_loss": 0.7439636,
      "learning_rate": 1.347908222500298e-07,
      "loss": 0.76590836,
      "num_input_tokens_seen": 159368325,
      "step": 7368,
      "time_per_iteration": 2.665827989578247
    },
    {
      "auxiliary_loss_clip": 0.01159691,
      "auxiliary_loss_mlp": 0.01023642,
      "balance_loss_clip": 1.04222894,
      "balance_loss_mlp": 1.01679635,
      "epoch": 0.8860698611194613,
      "flos": 16872916469760.0,
      "grad_norm": 1.9707197315906635,
      "language_loss": 0.70041871,
      "learning_rate": 1.3450983300442276e-07,
      "loss": 0.72225207,
      "num_input_tokens_seen": 159387555,
      "step": 7369,
      "time_per_iteration": 2.564523458480835
    },
    {
      "auxiliary_loss_clip": 0.01168606,
      "auxiliary_loss_mlp": 0.0102534,
      "balance_loss_clip": 1.04530632,
      "balance_loss_mlp": 1.01822925,
      "epoch": 0.8861901040101005,
      "flos": 24681404206080.0,
      "grad_norm": 2.333164577532482,
      "language_loss": 0.73714656,
      "learning_rate": 1.3422912675219068e-07,
      "loss": 0.75908601,
      "num_input_tokens_seen": 159407310,
      "step": 7370,
      "time_per_iteration": 2.629620313644409
    },
    {
      "auxiliary_loss_clip": 0.01163866,
      "auxiliary_loss_mlp": 0.01021463,
      "balance_loss_clip": 1.04814243,
      "balance_loss_mlp": 1.01489496,
      "epoch": 0.8863103469007395,
      "flos": 24423026699520.0,
      "grad_norm": 1.6098845370361674,
      "language_loss": 0.78951395,
      "learning_rate": 1.339487035359166e-07,
      "loss": 0.81136727,
      "num_input_tokens_seen": 159427680,
      "step": 7371,
      "time_per_iteration": 2.6073880195617676
    },
    {
      "auxiliary_loss_clip": 0.0117184,
      "auxiliary_loss_mlp": 0.01207143,
      "balance_loss_clip": 1.04597735,
      "balance_loss_mlp": 1.00076056,
      "epoch": 0.8864305897913786,
      "flos": 22053964158720.0,
      "grad_norm": 1.7989830297715947,
      "language_loss": 0.85023057,
      "learning_rate": 1.336685633981409e-07,
      "loss": 0.87402046,
      "num_input_tokens_seen": 159448765,
      "step": 7372,
      "time_per_iteration": 2.646817207336426
    },
    {
      "auxiliary_loss_clip": 0.01170628,
      "auxiliary_loss_mlp": 0.01021434,
      "balance_loss_clip": 1.04460955,
      "balance_loss_mlp": 1.01398695,
      "epoch": 0.8865508326820177,
      "flos": 19099449843840.0,
      "grad_norm": 1.7251054656928504,
      "language_loss": 0.74959636,
      "learning_rate": 1.333887063813597e-07,
      "loss": 0.77151704,
      "num_input_tokens_seen": 159466870,
      "step": 7373,
      "time_per_iteration": 2.578850507736206
    },
    {
      "auxiliary_loss_clip": 0.01171081,
      "auxiliary_loss_mlp": 0.01020127,
      "balance_loss_clip": 1.04176259,
      "balance_loss_mlp": 1.01323342,
      "epoch": 0.8866710755726568,
      "flos": 15414189240960.0,
      "grad_norm": 20.36815767781891,
      "language_loss": 0.6659649,
      "learning_rate": 1.331091325280278e-07,
      "loss": 0.687877,
      "num_input_tokens_seen": 159485840,
      "step": 7374,
      "time_per_iteration": 4.5077245235443115
    },
    {
      "auxiliary_loss_clip": 0.01168481,
      "auxiliary_loss_mlp": 0.01023195,
      "balance_loss_clip": 1.03975832,
      "balance_loss_mlp": 1.01572025,
      "epoch": 0.8867913184632958,
      "flos": 20083689388800.0,
      "grad_norm": 1.768276116142462,
      "language_loss": 0.78597271,
      "learning_rate": 1.3282984188055625e-07,
      "loss": 0.80788946,
      "num_input_tokens_seen": 159505630,
      "step": 7375,
      "time_per_iteration": 2.711461305618286
    },
    {
      "auxiliary_loss_clip": 0.01163982,
      "auxiliary_loss_mlp": 0.01028159,
      "balance_loss_clip": 1.0454495,
      "balance_loss_mlp": 1.02125931,
      "epoch": 0.8869115613539349,
      "flos": 23365852588800.0,
      "grad_norm": 1.857109855967553,
      "language_loss": 0.7936641,
      "learning_rate": 1.3255083448131288e-07,
      "loss": 0.81558549,
      "num_input_tokens_seen": 159524675,
      "step": 7376,
      "time_per_iteration": 2.57372784614563
    },
    {
      "auxiliary_loss_clip": 0.01170918,
      "auxiliary_loss_mlp": 0.01026222,
      "balance_loss_clip": 1.04420757,
      "balance_loss_mlp": 1.0189203,
      "epoch": 0.8870318042445741,
      "flos": 21286840371840.0,
      "grad_norm": 3.1579708131819793,
      "language_loss": 0.79014337,
      "learning_rate": 1.3227211037262365e-07,
      "loss": 0.81211472,
      "num_input_tokens_seen": 159541915,
      "step": 7377,
      "time_per_iteration": 2.674503803253174
    },
    {
      "auxiliary_loss_clip": 0.01178676,
      "auxiliary_loss_mlp": 0.01029243,
      "balance_loss_clip": 1.04025054,
      "balance_loss_mlp": 1.02141058,
      "epoch": 0.8871520471352131,
      "flos": 20010862563840.0,
      "grad_norm": 2.15567626241105,
      "language_loss": 0.85579395,
      "learning_rate": 1.319936695967696e-07,
      "loss": 0.87787312,
      "num_input_tokens_seen": 159559740,
      "step": 7378,
      "time_per_iteration": 2.68538236618042
    },
    {
      "auxiliary_loss_clip": 0.0117351,
      "auxiliary_loss_mlp": 0.01024077,
      "balance_loss_clip": 1.04841399,
      "balance_loss_mlp": 1.01586342,
      "epoch": 0.8872722900258522,
      "flos": 22601422321920.0,
      "grad_norm": 2.5303244585400706,
      "language_loss": 0.82321322,
      "learning_rate": 1.3171551219599097e-07,
      "loss": 0.84518909,
      "num_input_tokens_seen": 159578265,
      "step": 7379,
      "time_per_iteration": 3.6000149250030518
    },
    {
      "auxiliary_loss_clip": 0.01167819,
      "auxiliary_loss_mlp": 0.01022183,
      "balance_loss_clip": 1.04974651,
      "balance_loss_mlp": 1.01465476,
      "epoch": 0.8873925329164913,
      "flos": 22163276223360.0,
      "grad_norm": 2.2071594230348017,
      "language_loss": 0.7858094,
      "learning_rate": 1.3143763821248377e-07,
      "loss": 0.8077094,
      "num_input_tokens_seen": 159595350,
      "step": 7380,
      "time_per_iteration": 2.6030430793762207
    },
    {
      "auxiliary_loss_clip": 0.01163401,
      "auxiliary_loss_mlp": 0.01026152,
      "balance_loss_clip": 1.04621649,
      "balance_loss_mlp": 1.01916671,
      "epoch": 0.8875127758071304,
      "flos": 19208223204480.0,
      "grad_norm": 1.9219077266834703,
      "language_loss": 0.72216076,
      "learning_rate": 1.3116004768840118e-07,
      "loss": 0.74405628,
      "num_input_tokens_seen": 159613725,
      "step": 7381,
      "time_per_iteration": 2.591684579849243
    },
    {
      "auxiliary_loss_clip": 0.01167068,
      "auxiliary_loss_mlp": 0.01028327,
      "balance_loss_clip": 1.04610479,
      "balance_loss_mlp": 1.02076292,
      "epoch": 0.8876330186977694,
      "flos": 18110900666880.0,
      "grad_norm": 1.978574864348996,
      "language_loss": 0.74188095,
      "learning_rate": 1.3088274066585348e-07,
      "loss": 0.76383495,
      "num_input_tokens_seen": 159631335,
      "step": 7382,
      "time_per_iteration": 2.5981953144073486
    },
    {
      "auxiliary_loss_clip": 0.01181391,
      "auxiliary_loss_mlp": 0.0102367,
      "balance_loss_clip": 1.04319453,
      "balance_loss_mlp": 1.0168035,
      "epoch": 0.8877532615884086,
      "flos": 22009434272640.0,
      "grad_norm": 2.3136742329152287,
      "language_loss": 0.90766114,
      "learning_rate": 1.3060571718690749e-07,
      "loss": 0.9297117,
      "num_input_tokens_seen": 159648830,
      "step": 7383,
      "time_per_iteration": 2.655799627304077
    },
    {
      "auxiliary_loss_clip": 0.01083319,
      "auxiliary_loss_mlp": 0.01198696,
      "balance_loss_clip": 1.01089275,
      "balance_loss_mlp": 0.99995768,
      "epoch": 0.8878735044790477,
      "flos": 72136924346880.0,
      "grad_norm": 0.7460035792613503,
      "language_loss": 0.56928551,
      "learning_rate": 1.3032897729358805e-07,
      "loss": 0.59210569,
      "num_input_tokens_seen": 159709785,
      "step": 7384,
      "time_per_iteration": 3.2416446208953857
    },
    {
      "auxiliary_loss_clip": 0.01168657,
      "auxiliary_loss_mlp": 0.01208026,
      "balance_loss_clip": 1.03617835,
      "balance_loss_mlp": 1.00079024,
      "epoch": 0.8879937473696867,
      "flos": 27526355061120.0,
      "grad_norm": 1.936727556505264,
      "language_loss": 0.80192161,
      "learning_rate": 1.3005252102787645e-07,
      "loss": 0.82568848,
      "num_input_tokens_seen": 159728725,
      "step": 7385,
      "time_per_iteration": 2.6845157146453857
    },
    {
      "auxiliary_loss_clip": 0.01171823,
      "auxiliary_loss_mlp": 0.01022474,
      "balance_loss_clip": 1.04637372,
      "balance_loss_mlp": 1.0151124,
      "epoch": 0.8881139902603259,
      "flos": 22234091886720.0,
      "grad_norm": 3.0676458711632972,
      "language_loss": 0.73521215,
      "learning_rate": 1.297763484317105e-07,
      "loss": 0.75715506,
      "num_input_tokens_seen": 159747020,
      "step": 7386,
      "time_per_iteration": 2.604424238204956
    },
    {
      "auxiliary_loss_clip": 0.01175548,
      "auxiliary_loss_mlp": 0.0120814,
      "balance_loss_clip": 1.04011691,
      "balance_loss_mlp": 1.00062251,
      "epoch": 0.888234233150965,
      "flos": 20299548170880.0,
      "grad_norm": 2.4070242332620437,
      "language_loss": 0.70655465,
      "learning_rate": 1.2950045954698551e-07,
      "loss": 0.7303915,
      "num_input_tokens_seen": 159764855,
      "step": 7387,
      "time_per_iteration": 2.65161395072937
    },
    {
      "auxiliary_loss_clip": 0.0116531,
      "auxiliary_loss_mlp": 0.01024095,
      "balance_loss_clip": 1.04245269,
      "balance_loss_mlp": 1.017097,
      "epoch": 0.888354476041604,
      "flos": 18147996437760.0,
      "grad_norm": 1.7582052676138544,
      "language_loss": 0.75374031,
      "learning_rate": 1.2922485441555343e-07,
      "loss": 0.77563435,
      "num_input_tokens_seen": 159783935,
      "step": 7388,
      "time_per_iteration": 2.637263774871826
    },
    {
      "auxiliary_loss_clip": 0.01163778,
      "auxiliary_loss_mlp": 0.01022087,
      "balance_loss_clip": 1.04519606,
      "balance_loss_mlp": 1.01460075,
      "epoch": 0.8884747189322432,
      "flos": 22014282608640.0,
      "grad_norm": 2.630831415251122,
      "language_loss": 0.81823254,
      "learning_rate": 1.2894953307922363e-07,
      "loss": 0.84009123,
      "num_input_tokens_seen": 159802895,
      "step": 7389,
      "time_per_iteration": 2.6671249866485596
    },
    {
      "auxiliary_loss_clip": 0.01169687,
      "auxiliary_loss_mlp": 0.01024377,
      "balance_loss_clip": 1.0424422,
      "balance_loss_mlp": 1.01728737,
      "epoch": 0.8885949618228822,
      "flos": 19786779567360.0,
      "grad_norm": 2.2541921316057434,
      "language_loss": 0.84415406,
      "learning_rate": 1.2867449557976208e-07,
      "loss": 0.86609471,
      "num_input_tokens_seen": 159820995,
      "step": 7390,
      "time_per_iteration": 2.6515355110168457
    },
    {
      "auxiliary_loss_clip": 0.01165327,
      "auxiliary_loss_mlp": 0.01025042,
      "balance_loss_clip": 1.04550493,
      "balance_loss_mlp": 1.01786268,
      "epoch": 0.8887152047135213,
      "flos": 20047599198720.0,
      "grad_norm": 3.5679645537694626,
      "language_loss": 0.75995183,
      "learning_rate": 1.283997419588916e-07,
      "loss": 0.78185552,
      "num_input_tokens_seen": 159840465,
      "step": 7391,
      "time_per_iteration": 2.548868179321289
    },
    {
      "auxiliary_loss_clip": 0.01173182,
      "auxiliary_loss_mlp": 0.0102436,
      "balance_loss_clip": 1.04675794,
      "balance_loss_mlp": 1.0169276,
      "epoch": 0.8888354476041604,
      "flos": 18588117784320.0,
      "grad_norm": 2.1665276908357116,
      "language_loss": 0.62021887,
      "learning_rate": 1.2812527225829216e-07,
      "loss": 0.64219427,
      "num_input_tokens_seen": 159858690,
      "step": 7392,
      "time_per_iteration": 2.6773898601531982
    },
    {
      "auxiliary_loss_clip": 0.01177162,
      "auxiliary_loss_mlp": 0.01021792,
      "balance_loss_clip": 1.04968476,
      "balance_loss_mlp": 1.01357269,
      "epoch": 0.8889556904947995,
      "flos": 21689794120320.0,
      "grad_norm": 1.9230060171055963,
      "language_loss": 0.76525021,
      "learning_rate": 1.2785108651960052e-07,
      "loss": 0.78723979,
      "num_input_tokens_seen": 159880325,
      "step": 7393,
      "time_per_iteration": 3.625084161758423
    },
    {
      "auxiliary_loss_clip": 0.0117201,
      "auxiliary_loss_mlp": 0.01022121,
      "balance_loss_clip": 1.04600251,
      "balance_loss_mlp": 1.01477194,
      "epoch": 0.8890759333854386,
      "flos": 27381204201600.0,
      "grad_norm": 1.9237649837404416,
      "language_loss": 0.80742639,
      "learning_rate": 1.2757718478441094e-07,
      "loss": 0.82936764,
      "num_input_tokens_seen": 159901070,
      "step": 7394,
      "time_per_iteration": 2.5990774631500244
    },
    {
      "auxiliary_loss_clip": 0.01170272,
      "auxiliary_loss_mlp": 0.01021448,
      "balance_loss_clip": 1.04349291,
      "balance_loss_mlp": 1.01440895,
      "epoch": 0.8891961762760777,
      "flos": 24498834353280.0,
      "grad_norm": 1.7141063185592589,
      "language_loss": 0.77377307,
      "learning_rate": 1.2730356709427302e-07,
      "loss": 0.79569024,
      "num_input_tokens_seen": 159919750,
      "step": 7395,
      "time_per_iteration": 2.639239549636841
    },
    {
      "auxiliary_loss_clip": 0.01166045,
      "auxiliary_loss_mlp": 0.01028747,
      "balance_loss_clip": 1.046386,
      "balance_loss_mlp": 1.02080202,
      "epoch": 0.8893164191667168,
      "flos": 41499770895360.0,
      "grad_norm": 1.5947440219123867,
      "language_loss": 0.60161114,
      "learning_rate": 1.2703023349069542e-07,
      "loss": 0.62355906,
      "num_input_tokens_seen": 159944600,
      "step": 7396,
      "time_per_iteration": 2.7764244079589844
    },
    {
      "auxiliary_loss_clip": 0.01163915,
      "auxiliary_loss_mlp": 0.01021412,
      "balance_loss_clip": 1.04534125,
      "balance_loss_mlp": 1.01431942,
      "epoch": 0.8894366620573558,
      "flos": 33583623120000.0,
      "grad_norm": 2.001298090963954,
      "language_loss": 0.62092769,
      "learning_rate": 1.2675718401514223e-07,
      "loss": 0.64278096,
      "num_input_tokens_seen": 159968780,
      "step": 7397,
      "time_per_iteration": 2.703049659729004
    },
    {
      "auxiliary_loss_clip": 0.01170181,
      "auxiliary_loss_mlp": 0.01025281,
      "balance_loss_clip": 1.04277635,
      "balance_loss_mlp": 1.01786637,
      "epoch": 0.889556904947995,
      "flos": 16909832672640.0,
      "grad_norm": 2.351198050810489,
      "language_loss": 0.74265069,
      "learning_rate": 1.264844187090346e-07,
      "loss": 0.76460528,
      "num_input_tokens_seen": 159985905,
      "step": 7398,
      "time_per_iteration": 2.586897850036621
    },
    {
      "auxiliary_loss_clip": 0.01165841,
      "auxiliary_loss_mlp": 0.01019682,
      "balance_loss_clip": 1.0424118,
      "balance_loss_mlp": 1.01262498,
      "epoch": 0.889677147838634,
      "flos": 26030855283840.0,
      "grad_norm": 1.7764976668712331,
      "language_loss": 0.75264937,
      "learning_rate": 1.262119376137516e-07,
      "loss": 0.77450466,
      "num_input_tokens_seen": 160006965,
      "step": 7399,
      "time_per_iteration": 2.657841205596924
    },
    {
      "auxiliary_loss_clip": 0.01158588,
      "auxiliary_loss_mlp": 0.0102174,
      "balance_loss_clip": 1.04241562,
      "balance_loss_mlp": 1.014853,
      "epoch": 0.8897973907292731,
      "flos": 26468283110400.0,
      "grad_norm": 1.5702684324580642,
      "language_loss": 0.85091138,
      "learning_rate": 1.2593974077062707e-07,
      "loss": 0.87271464,
      "num_input_tokens_seen": 160028585,
      "step": 7400,
      "time_per_iteration": 3.5288846492767334
    },
    {
      "auxiliary_loss_clip": 0.01164264,
      "auxiliary_loss_mlp": 0.01024978,
      "balance_loss_clip": 1.041255,
      "balance_loss_mlp": 1.01787329,
      "epoch": 0.8899176336199123,
      "flos": 26249694894720.0,
      "grad_norm": 1.7281397067569622,
      "language_loss": 0.63728929,
      "learning_rate": 1.2566782822095423e-07,
      "loss": 0.65918171,
      "num_input_tokens_seen": 160048840,
      "step": 7401,
      "time_per_iteration": 3.3936657905578613
    },
    {
      "auxiliary_loss_clip": 0.01181705,
      "auxiliary_loss_mlp": 0.01025665,
      "balance_loss_clip": 1.04668212,
      "balance_loss_mlp": 1.01822603,
      "epoch": 0.8900378765105513,
      "flos": 20811742156800.0,
      "grad_norm": 1.7431299456975264,
      "language_loss": 0.71317291,
      "learning_rate": 1.2539620000598162e-07,
      "loss": 0.7352466,
      "num_input_tokens_seen": 160068175,
      "step": 7402,
      "time_per_iteration": 2.6337649822235107
    },
    {
      "auxiliary_loss_clip": 0.01166875,
      "auxiliary_loss_mlp": 0.01025141,
      "balance_loss_clip": 1.04822612,
      "balance_loss_mlp": 1.01772058,
      "epoch": 0.8901581194011904,
      "flos": 16472333018880.0,
      "grad_norm": 1.9521117839034716,
      "language_loss": 0.7987147,
      "learning_rate": 1.2512485616691492e-07,
      "loss": 0.82063478,
      "num_input_tokens_seen": 160085230,
      "step": 7403,
      "time_per_iteration": 2.5793533325195312
    },
    {
      "auxiliary_loss_clip": 0.01174305,
      "auxiliary_loss_mlp": 0.01028444,
      "balance_loss_clip": 1.04151869,
      "balance_loss_mlp": 1.02074921,
      "epoch": 0.8902783622918296,
      "flos": 35155253773440.0,
      "grad_norm": 1.4392366526344142,
      "language_loss": 0.81005257,
      "learning_rate": 1.2485379674491681e-07,
      "loss": 0.83208007,
      "num_input_tokens_seen": 160111425,
      "step": 7404,
      "time_per_iteration": 2.8184056282043457
    },
    {
      "auxiliary_loss_clip": 0.01172588,
      "auxiliary_loss_mlp": 0.01026089,
      "balance_loss_clip": 1.04645228,
      "balance_loss_mlp": 1.01862073,
      "epoch": 0.8903986051824686,
      "flos": 17201068145280.0,
      "grad_norm": 2.231268457941699,
      "language_loss": 0.79677093,
      "learning_rate": 1.2458302178110657e-07,
      "loss": 0.81875765,
      "num_input_tokens_seen": 160129790,
      "step": 7405,
      "time_per_iteration": 2.662571668624878
    },
    {
      "auxiliary_loss_clip": 0.0116315,
      "auxiliary_loss_mlp": 0.01019153,
      "balance_loss_clip": 1.04102254,
      "balance_loss_mlp": 1.01233423,
      "epoch": 0.8905188480731077,
      "flos": 25483863997440.0,
      "grad_norm": 2.1627058164025157,
      "language_loss": 0.82522976,
      "learning_rate": 1.2431253131656118e-07,
      "loss": 0.84705281,
      "num_input_tokens_seen": 160149265,
      "step": 7406,
      "time_per_iteration": 3.6270575523376465
    },
    {
      "auxiliary_loss_clip": 0.01161451,
      "auxiliary_loss_mlp": 0.01024337,
      "balance_loss_clip": 1.04247832,
      "balance_loss_mlp": 1.01658797,
      "epoch": 0.8906390909637467,
      "flos": 23365888502400.0,
      "grad_norm": 2.020752701628418,
      "language_loss": 0.76505339,
      "learning_rate": 1.240423253923133e-07,
      "loss": 0.78691125,
      "num_input_tokens_seen": 160168870,
      "step": 7407,
      "time_per_iteration": 2.6155261993408203
    },
    {
      "auxiliary_loss_clip": 0.01169928,
      "auxiliary_loss_mlp": 0.01027346,
      "balance_loss_clip": 1.04588747,
      "balance_loss_mlp": 1.01948452,
      "epoch": 0.8907593338543859,
      "flos": 21068790860160.0,
      "grad_norm": 1.915882870231437,
      "language_loss": 0.6965214,
      "learning_rate": 1.237724040493533e-07,
      "loss": 0.71849418,
      "num_input_tokens_seen": 160187495,
      "step": 7408,
      "time_per_iteration": 2.582782745361328
    },
    {
      "auxiliary_loss_clip": 0.01173931,
      "auxiliary_loss_mlp": 0.0102957,
      "balance_loss_clip": 1.05071294,
      "balance_loss_mlp": 1.02163088,
      "epoch": 0.8908795767450249,
      "flos": 21869562712320.0,
      "grad_norm": 2.271034514781697,
      "language_loss": 0.72877824,
      "learning_rate": 1.2350276732862773e-07,
      "loss": 0.75081325,
      "num_input_tokens_seen": 160208520,
      "step": 7409,
      "time_per_iteration": 2.647495985031128
    },
    {
      "auxiliary_loss_clip": 0.01069383,
      "auxiliary_loss_mlp": 0.01001419,
      "balance_loss_clip": 1.00972414,
      "balance_loss_mlp": 1.00070667,
      "epoch": 0.890999819635664,
      "flos": 66307869348480.0,
      "grad_norm": 0.8356132093216447,
      "language_loss": 0.56664741,
      "learning_rate": 1.2323341527103993e-07,
      "loss": 0.58735538,
      "num_input_tokens_seen": 160263720,
      "step": 7410,
      "time_per_iteration": 3.0684938430786133
    },
    {
      "auxiliary_loss_clip": 0.01165343,
      "auxiliary_loss_mlp": 0.0102442,
      "balance_loss_clip": 1.04734814,
      "balance_loss_mlp": 1.01708293,
      "epoch": 0.8911200625263032,
      "flos": 26869908055680.0,
      "grad_norm": 2.0095835096495622,
      "language_loss": 0.85367793,
      "learning_rate": 1.2296434791745135e-07,
      "loss": 0.87557554,
      "num_input_tokens_seen": 160282170,
      "step": 7411,
      "time_per_iteration": 2.7367186546325684
    },
    {
      "auxiliary_loss_clip": 0.01171979,
      "auxiliary_loss_mlp": 0.01022573,
      "balance_loss_clip": 1.04618979,
      "balance_loss_mlp": 1.01509011,
      "epoch": 0.8912403054169422,
      "flos": 20885825957760.0,
      "grad_norm": 1.7194732046628594,
      "language_loss": 0.76563394,
      "learning_rate": 1.2269556530867875e-07,
      "loss": 0.78757942,
      "num_input_tokens_seen": 160300725,
      "step": 7412,
      "time_per_iteration": 2.6004233360290527
    },
    {
      "auxiliary_loss_clip": 0.01172703,
      "auxiliary_loss_mlp": 0.01026151,
      "balance_loss_clip": 1.04952049,
      "balance_loss_mlp": 1.01781869,
      "epoch": 0.8913605483075813,
      "flos": 27016567286400.0,
      "grad_norm": 2.0163564912234437,
      "language_loss": 0.81894505,
      "learning_rate": 1.2242706748549614e-07,
      "loss": 0.84093356,
      "num_input_tokens_seen": 160318720,
      "step": 7413,
      "time_per_iteration": 2.6173806190490723
    },
    {
      "auxiliary_loss_clip": 0.01170128,
      "auxiliary_loss_mlp": 0.01022057,
      "balance_loss_clip": 1.0401516,
      "balance_loss_mlp": 1.01469278,
      "epoch": 0.8914807911982204,
      "flos": 23621500661760.0,
      "grad_norm": 2.0440567256317914,
      "language_loss": 0.8213985,
      "learning_rate": 1.2215885448863473e-07,
      "loss": 0.84332037,
      "num_input_tokens_seen": 160339595,
      "step": 7414,
      "time_per_iteration": 2.737140655517578
    },
    {
      "auxiliary_loss_clip": 0.01170726,
      "auxiliary_loss_mlp": 0.01025092,
      "balance_loss_clip": 1.04459143,
      "balance_loss_mlp": 1.01833928,
      "epoch": 0.8916010340888595,
      "flos": 24462277286400.0,
      "grad_norm": 2.0528759682003437,
      "language_loss": 0.80213201,
      "learning_rate": 1.2189092635878152e-07,
      "loss": 0.82409012,
      "num_input_tokens_seen": 160361045,
      "step": 7415,
      "time_per_iteration": 2.656893253326416
    },
    {
      "auxiliary_loss_clip": 0.01163829,
      "auxiliary_loss_mlp": 0.01025554,
      "balance_loss_clip": 1.04080629,
      "balance_loss_mlp": 1.01815677,
      "epoch": 0.8917212769794985,
      "flos": 21215773313280.0,
      "grad_norm": 2.1138162901283026,
      "language_loss": 0.77731955,
      "learning_rate": 1.216232831365822e-07,
      "loss": 0.79921341,
      "num_input_tokens_seen": 160379990,
      "step": 7416,
      "time_per_iteration": 2.7272732257843018
    },
    {
      "auxiliary_loss_clip": 0.01178338,
      "auxiliary_loss_mlp": 0.01024658,
      "balance_loss_clip": 1.04655993,
      "balance_loss_mlp": 1.01684356,
      "epoch": 0.8918415198701377,
      "flos": 25513992529920.0,
      "grad_norm": 1.8215691156575864,
      "language_loss": 0.80787027,
      "learning_rate": 1.2135592486263678e-07,
      "loss": 0.82990015,
      "num_input_tokens_seen": 160399240,
      "step": 7417,
      "time_per_iteration": 2.6753923892974854
    },
    {
      "auxiliary_loss_clip": 0.01172806,
      "auxiliary_loss_mlp": 0.01024542,
      "balance_loss_clip": 1.04506803,
      "balance_loss_mlp": 1.01726437,
      "epoch": 0.8919617627607768,
      "flos": 37853006693760.0,
      "grad_norm": 1.6040195950390852,
      "language_loss": 0.6129899,
      "learning_rate": 1.2108885157750415e-07,
      "loss": 0.63496339,
      "num_input_tokens_seen": 160421600,
      "step": 7418,
      "time_per_iteration": 2.7405428886413574
    },
    {
      "auxiliary_loss_clip": 0.01171811,
      "auxiliary_loss_mlp": 0.01207403,
      "balance_loss_clip": 1.0452553,
      "balance_loss_mlp": 1.00091946,
      "epoch": 0.8920820056514158,
      "flos": 26213676531840.0,
      "grad_norm": 1.8016838415552012,
      "language_loss": 0.80050969,
      "learning_rate": 1.2082206332169897e-07,
      "loss": 0.82430184,
      "num_input_tokens_seen": 160441695,
      "step": 7419,
      "time_per_iteration": 3.532318115234375
    },
    {
      "auxiliary_loss_clip": 0.01165204,
      "auxiliary_loss_mlp": 0.01026143,
      "balance_loss_clip": 1.04448867,
      "balance_loss_mlp": 1.01828718,
      "epoch": 0.892202248542055,
      "flos": 17383135207680.0,
      "grad_norm": 2.5997445267751265,
      "language_loss": 0.73280466,
      "learning_rate": 1.2055556013569225e-07,
      "loss": 0.75471812,
      "num_input_tokens_seen": 160457205,
      "step": 7420,
      "time_per_iteration": 2.4951980113983154
    },
    {
      "auxiliary_loss_clip": 0.01174493,
      "auxiliary_loss_mlp": 0.01022796,
      "balance_loss_clip": 1.04453945,
      "balance_loss_mlp": 1.01556563,
      "epoch": 0.892322491432694,
      "flos": 21324223451520.0,
      "grad_norm": 1.6428318904389831,
      "language_loss": 0.82343596,
      "learning_rate": 1.2028934205991315e-07,
      "loss": 0.8454088,
      "num_input_tokens_seen": 160476525,
      "step": 7421,
      "time_per_iteration": 2.511608123779297
    },
    {
      "auxiliary_loss_clip": 0.01167791,
      "auxiliary_loss_mlp": 0.01021564,
      "balance_loss_clip": 1.04378819,
      "balance_loss_mlp": 1.01385379,
      "epoch": 0.8924427343233331,
      "flos": 24029374573440.0,
      "grad_norm": 1.7672825793811666,
      "language_loss": 0.76734126,
      "learning_rate": 1.2002340913474607e-07,
      "loss": 0.78923482,
      "num_input_tokens_seen": 160500160,
      "step": 7422,
      "time_per_iteration": 2.6446821689605713
    },
    {
      "auxiliary_loss_clip": 0.01167103,
      "auxiliary_loss_mlp": 0.01025896,
      "balance_loss_clip": 1.04678011,
      "balance_loss_mlp": 1.01794517,
      "epoch": 0.8925629772139723,
      "flos": 30008069631360.0,
      "grad_norm": 2.629114935358391,
      "language_loss": 0.73989356,
      "learning_rate": 1.1975776140053317e-07,
      "loss": 0.76182353,
      "num_input_tokens_seen": 160520130,
      "step": 7423,
      "time_per_iteration": 2.6727705001831055
    },
    {
      "auxiliary_loss_clip": 0.01177891,
      "auxiliary_loss_mlp": 0.01025259,
      "balance_loss_clip": 1.0413208,
      "balance_loss_mlp": 1.01733112,
      "epoch": 0.8926832201046113,
      "flos": 22601709630720.0,
      "grad_norm": 2.3478276300127736,
      "language_loss": 0.73477232,
      "learning_rate": 1.194923988975729e-07,
      "loss": 0.75680387,
      "num_input_tokens_seen": 160539730,
      "step": 7424,
      "time_per_iteration": 2.701313018798828
    },
    {
      "auxiliary_loss_clip": 0.01169083,
      "auxiliary_loss_mlp": 0.01020856,
      "balance_loss_clip": 1.04193091,
      "balance_loss_mlp": 1.01255357,
      "epoch": 0.8928034629952504,
      "flos": 13297722117120.0,
      "grad_norm": 2.6436089690500593,
      "language_loss": 0.7359165,
      "learning_rate": 1.192273216661206e-07,
      "loss": 0.7578159,
      "num_input_tokens_seen": 160557820,
      "step": 7425,
      "time_per_iteration": 2.9179515838623047
    },
    {
      "auxiliary_loss_clip": 0.01097157,
      "auxiliary_loss_mlp": 0.01001199,
      "balance_loss_clip": 1.01146567,
      "balance_loss_mlp": 1.00045431,
      "epoch": 0.8929237058858895,
      "flos": 54854556744960.0,
      "grad_norm": 0.761967886383919,
      "language_loss": 0.57499856,
      "learning_rate": 1.189625297463881e-07,
      "loss": 0.59598207,
      "num_input_tokens_seen": 160619510,
      "step": 7426,
      "time_per_iteration": 3.136173725128174
    },
    {
      "auxiliary_loss_clip": 0.01172187,
      "auxiliary_loss_mlp": 0.01024979,
      "balance_loss_clip": 1.03717685,
      "balance_loss_mlp": 1.01790714,
      "epoch": 0.8930439487765286,
      "flos": 28883850785280.0,
      "grad_norm": 1.8003950954916434,
      "language_loss": 0.79560828,
      "learning_rate": 1.1869802317854394e-07,
      "loss": 0.81757993,
      "num_input_tokens_seen": 160643295,
      "step": 7427,
      "time_per_iteration": 3.651620864868164
    },
    {
      "auxiliary_loss_clip": 0.01182213,
      "auxiliary_loss_mlp": 0.01023776,
      "balance_loss_clip": 1.04258537,
      "balance_loss_mlp": 1.01624799,
      "epoch": 0.8931641916671677,
      "flos": 22419283432320.0,
      "grad_norm": 1.8430296147971676,
      "language_loss": 0.72254026,
      "learning_rate": 1.1843380200271425e-07,
      "loss": 0.74460012,
      "num_input_tokens_seen": 160662495,
      "step": 7428,
      "time_per_iteration": 3.607053279876709
    },
    {
      "auxiliary_loss_clip": 0.01168643,
      "auxiliary_loss_mlp": 0.01018673,
      "balance_loss_clip": 1.04266012,
      "balance_loss_mlp": 1.01090074,
      "epoch": 0.8932844345578068,
      "flos": 25843149786240.0,
      "grad_norm": 1.7722054693747857,
      "language_loss": 0.80551904,
      "learning_rate": 1.181698662589805e-07,
      "loss": 0.82739222,
      "num_input_tokens_seen": 160682080,
      "step": 7429,
      "time_per_iteration": 2.667931079864502
    },
    {
      "auxiliary_loss_clip": 0.0116778,
      "auxiliary_loss_mlp": 0.01024275,
      "balance_loss_clip": 1.04454041,
      "balance_loss_mlp": 1.01666379,
      "epoch": 0.8934046774484459,
      "flos": 22925803069440.0,
      "grad_norm": 1.8170126612845814,
      "language_loss": 0.76281559,
      "learning_rate": 1.1790621598738249e-07,
      "loss": 0.78473616,
      "num_input_tokens_seen": 160700395,
      "step": 7430,
      "time_per_iteration": 2.7128567695617676
    },
    {
      "auxiliary_loss_clip": 0.01164204,
      "auxiliary_loss_mlp": 0.01024841,
      "balance_loss_clip": 1.04827416,
      "balance_loss_mlp": 1.01810586,
      "epoch": 0.8935249203390849,
      "flos": 24462097718400.0,
      "grad_norm": 1.9607705982775447,
      "language_loss": 0.7483809,
      "learning_rate": 1.1764285122791461e-07,
      "loss": 0.7702713,
      "num_input_tokens_seen": 160721115,
      "step": 7431,
      "time_per_iteration": 2.6285574436187744
    },
    {
      "auxiliary_loss_clip": 0.01167468,
      "auxiliary_loss_mlp": 0.01022188,
      "balance_loss_clip": 1.04269028,
      "balance_loss_mlp": 1.01494861,
      "epoch": 0.8936451632297241,
      "flos": 15742735966080.0,
      "grad_norm": 2.080405097713294,
      "language_loss": 0.77122134,
      "learning_rate": 1.173797720205294e-07,
      "loss": 0.79311788,
      "num_input_tokens_seen": 160739150,
      "step": 7432,
      "time_per_iteration": 2.6495954990386963
    },
    {
      "auxiliary_loss_clip": 0.01172636,
      "auxiliary_loss_mlp": 0.01028555,
      "balance_loss_clip": 1.04771662,
      "balance_loss_mlp": 1.02040982,
      "epoch": 0.8937654061203631,
      "flos": 35115500396160.0,
      "grad_norm": 3.6326086049222335,
      "language_loss": 0.71885121,
      "learning_rate": 1.1711697840513602e-07,
      "loss": 0.74086308,
      "num_input_tokens_seen": 160758585,
      "step": 7433,
      "time_per_iteration": 3.6837127208709717
    },
    {
      "auxiliary_loss_clip": 0.01160712,
      "auxiliary_loss_mlp": 0.01021877,
      "balance_loss_clip": 1.04290617,
      "balance_loss_mlp": 1.01444435,
      "epoch": 0.8938856490110022,
      "flos": 16107444708480.0,
      "grad_norm": 2.5673947578043568,
      "language_loss": 0.70951796,
      "learning_rate": 1.1685447042160012e-07,
      "loss": 0.73134387,
      "num_input_tokens_seen": 160776620,
      "step": 7434,
      "time_per_iteration": 2.6353495121002197
    },
    {
      "auxiliary_loss_clip": 0.01169292,
      "auxiliary_loss_mlp": 0.01027521,
      "balance_loss_clip": 1.04801929,
      "balance_loss_mlp": 1.01983166,
      "epoch": 0.8940058919016414,
      "flos": 20704189858560.0,
      "grad_norm": 1.6108116925759108,
      "language_loss": 0.7166732,
      "learning_rate": 1.1659224810974367e-07,
      "loss": 0.73864138,
      "num_input_tokens_seen": 160796580,
      "step": 7435,
      "time_per_iteration": 2.618809223175049
    },
    {
      "auxiliary_loss_clip": 0.01168713,
      "auxiliary_loss_mlp": 0.01026698,
      "balance_loss_clip": 1.04415488,
      "balance_loss_mlp": 1.01941466,
      "epoch": 0.8941261347922804,
      "flos": 25229041937280.0,
      "grad_norm": 1.8326346821953996,
      "language_loss": 0.68385243,
      "learning_rate": 1.1633031150934591e-07,
      "loss": 0.70580661,
      "num_input_tokens_seen": 160819610,
      "step": 7436,
      "time_per_iteration": 2.7563636302948
    },
    {
      "auxiliary_loss_clip": 0.01172287,
      "auxiliary_loss_mlp": 0.01029722,
      "balance_loss_clip": 1.04800534,
      "balance_loss_mlp": 1.02192616,
      "epoch": 0.8942463776829195,
      "flos": 19537236806400.0,
      "grad_norm": 1.8358592760982377,
      "language_loss": 0.79951864,
      "learning_rate": 1.1606866066014176e-07,
      "loss": 0.82153875,
      "num_input_tokens_seen": 160838660,
      "step": 7437,
      "time_per_iteration": 2.6870245933532715
    },
    {
      "auxiliary_loss_clip": 0.0117327,
      "auxiliary_loss_mlp": 0.01022047,
      "balance_loss_clip": 1.04423189,
      "balance_loss_mlp": 1.01442957,
      "epoch": 0.8943666205735585,
      "flos": 22301567585280.0,
      "grad_norm": 2.7780729411615117,
      "language_loss": 0.75492382,
      "learning_rate": 1.1580729560182434e-07,
      "loss": 0.77687699,
      "num_input_tokens_seen": 160854515,
      "step": 7438,
      "time_per_iteration": 2.636228561401367
    },
    {
      "auxiliary_loss_clip": 0.01167049,
      "auxiliary_loss_mlp": 0.01207673,
      "balance_loss_clip": 1.04778767,
      "balance_loss_mlp": 1.00061035,
      "epoch": 0.8944868634641977,
      "flos": 18912893581440.0,
      "grad_norm": 1.908795804608955,
      "language_loss": 0.71079558,
      "learning_rate": 1.1554621637404171e-07,
      "loss": 0.73454285,
      "num_input_tokens_seen": 160872605,
      "step": 7439,
      "time_per_iteration": 2.5909440517425537
    },
    {
      "auxiliary_loss_clip": 0.01172005,
      "auxiliary_loss_mlp": 0.01019715,
      "balance_loss_clip": 1.04631376,
      "balance_loss_mlp": 1.01239538,
      "epoch": 0.8946071063548368,
      "flos": 14460904241280.0,
      "grad_norm": 2.209409343917481,
      "language_loss": 0.61326671,
      "learning_rate": 1.1528542301639999e-07,
      "loss": 0.63518393,
      "num_input_tokens_seen": 160889395,
      "step": 7440,
      "time_per_iteration": 2.545606851577759
    },
    {
      "auxiliary_loss_clip": 0.01176715,
      "auxiliary_loss_mlp": 0.01019662,
      "balance_loss_clip": 1.04166889,
      "balance_loss_mlp": 1.01258349,
      "epoch": 0.8947273492454758,
      "flos": 20084084438400.0,
      "grad_norm": 3.6827930081893694,
      "language_loss": 0.82449967,
      "learning_rate": 1.1502491556846105e-07,
      "loss": 0.84646344,
      "num_input_tokens_seen": 160907890,
      "step": 7441,
      "time_per_iteration": 2.6784024238586426
    },
    {
      "auxiliary_loss_clip": 0.0117134,
      "auxiliary_loss_mlp": 0.01023217,
      "balance_loss_clip": 1.04401064,
      "balance_loss_mlp": 1.01597798,
      "epoch": 0.894847592136115,
      "flos": 18550555136640.0,
      "grad_norm": 2.2859685112335733,
      "language_loss": 0.81748396,
      "learning_rate": 1.1476469406974331e-07,
      "loss": 0.8394295,
      "num_input_tokens_seen": 160923490,
      "step": 7442,
      "time_per_iteration": 2.604135036468506
    },
    {
      "auxiliary_loss_clip": 0.01164421,
      "auxiliary_loss_mlp": 0.01023911,
      "balance_loss_clip": 1.04750288,
      "balance_loss_mlp": 1.01699424,
      "epoch": 0.894967835026754,
      "flos": 23478468704640.0,
      "grad_norm": 1.7753170209288212,
      "language_loss": 0.77159423,
      "learning_rate": 1.1450475855972341e-07,
      "loss": 0.79347754,
      "num_input_tokens_seen": 160944280,
      "step": 7443,
      "time_per_iteration": 2.603395462036133
    },
    {
      "auxiliary_loss_clip": 0.0116971,
      "auxiliary_loss_mlp": 0.01207951,
      "balance_loss_clip": 1.04150271,
      "balance_loss_mlp": 1.00075829,
      "epoch": 0.8950880779173931,
      "flos": 15188310564480.0,
      "grad_norm": 2.0217400989219296,
      "language_loss": 0.7080512,
      "learning_rate": 1.1424510907783158e-07,
      "loss": 0.7318278,
      "num_input_tokens_seen": 160961560,
      "step": 7444,
      "time_per_iteration": 2.622746706008911
    },
    {
      "auxiliary_loss_clip": 0.01174886,
      "auxiliary_loss_mlp": 0.01027234,
      "balance_loss_clip": 1.04255795,
      "balance_loss_mlp": 1.02017701,
      "epoch": 0.8952083208080323,
      "flos": 22091957769600.0,
      "grad_norm": 1.618797311786408,
      "language_loss": 0.82993871,
      "learning_rate": 1.1398574566345787e-07,
      "loss": 0.85195994,
      "num_input_tokens_seen": 160982195,
      "step": 7445,
      "time_per_iteration": 3.6144320964813232
    },
    {
      "auxiliary_loss_clip": 0.01174981,
      "auxiliary_loss_mlp": 0.0102224,
      "balance_loss_clip": 1.04205239,
      "balance_loss_mlp": 1.01460481,
      "epoch": 0.8953285636986713,
      "flos": 23254026572160.0,
      "grad_norm": 2.526866789451798,
      "language_loss": 0.82301903,
      "learning_rate": 1.1372666835594702e-07,
      "loss": 0.84499121,
      "num_input_tokens_seen": 161000520,
      "step": 7446,
      "time_per_iteration": 2.6851954460144043
    },
    {
      "auxiliary_loss_clip": 0.0117128,
      "auxiliary_loss_mlp": 0.01022528,
      "balance_loss_clip": 1.04551363,
      "balance_loss_mlp": 1.01570344,
      "epoch": 0.8954488065893104,
      "flos": 16362661818240.0,
      "grad_norm": 2.044582358443426,
      "language_loss": 0.72255397,
      "learning_rate": 1.1346787719460071e-07,
      "loss": 0.74449211,
      "num_input_tokens_seen": 161019405,
      "step": 7447,
      "time_per_iteration": 2.633760690689087
    },
    {
      "auxiliary_loss_clip": 0.01170245,
      "auxiliary_loss_mlp": 0.01025815,
      "balance_loss_clip": 1.04478812,
      "balance_loss_mlp": 1.01848364,
      "epoch": 0.8955690494799495,
      "flos": 18257883120000.0,
      "grad_norm": 1.7414737243933718,
      "language_loss": 0.72149837,
      "learning_rate": 1.1320937221867732e-07,
      "loss": 0.74345899,
      "num_input_tokens_seen": 161036985,
      "step": 7448,
      "time_per_iteration": 2.650190830230713
    },
    {
      "auxiliary_loss_clip": 0.01168393,
      "auxiliary_loss_mlp": 0.01021332,
      "balance_loss_clip": 1.04202318,
      "balance_loss_mlp": 1.01479983,
      "epoch": 0.8956892923705886,
      "flos": 25447486498560.0,
      "grad_norm": 2.716377699139976,
      "language_loss": 0.80025017,
      "learning_rate": 1.1295115346739192e-07,
      "loss": 0.82214749,
      "num_input_tokens_seen": 161056985,
      "step": 7449,
      "time_per_iteration": 2.6513867378234863
    },
    {
      "auxiliary_loss_clip": 0.0117468,
      "auxiliary_loss_mlp": 0.01026011,
      "balance_loss_clip": 1.04502535,
      "balance_loss_mlp": 1.01846504,
      "epoch": 0.8958095352612276,
      "flos": 52661883939840.0,
      "grad_norm": 2.2374262815440558,
      "language_loss": 0.73344803,
      "learning_rate": 1.1269322097991629e-07,
      "loss": 0.7554549,
      "num_input_tokens_seen": 161080270,
      "step": 7450,
      "time_per_iteration": 2.920642614364624
    },
    {
      "auxiliary_loss_clip": 0.01175162,
      "auxiliary_loss_mlp": 0.01027991,
      "balance_loss_clip": 1.04933167,
      "balance_loss_mlp": 1.02013206,
      "epoch": 0.8959297781518668,
      "flos": 23186335392000.0,
      "grad_norm": 2.1097555549886406,
      "language_loss": 0.67891848,
      "learning_rate": 1.1243557479537846e-07,
      "loss": 0.70095003,
      "num_input_tokens_seen": 161100160,
      "step": 7451,
      "time_per_iteration": 2.605870008468628
    },
    {
      "auxiliary_loss_clip": 0.01163981,
      "auxiliary_loss_mlp": 0.01020286,
      "balance_loss_clip": 1.04443502,
      "balance_loss_mlp": 1.01277924,
      "epoch": 0.8960500210425059,
      "flos": 20334309557760.0,
      "grad_norm": 2.1149137099040125,
      "language_loss": 0.68676347,
      "learning_rate": 1.121782149528634e-07,
      "loss": 0.70860612,
      "num_input_tokens_seen": 161117260,
      "step": 7452,
      "time_per_iteration": 2.6157166957855225
    },
    {
      "auxiliary_loss_clip": 0.01173583,
      "auxiliary_loss_mlp": 0.01019484,
      "balance_loss_clip": 1.04517651,
      "balance_loss_mlp": 1.01238823,
      "epoch": 0.8961702639331449,
      "flos": 19901694153600.0,
      "grad_norm": 2.712646309810521,
      "language_loss": 0.79129654,
      "learning_rate": 1.1192114149141208e-07,
      "loss": 0.81322718,
      "num_input_tokens_seen": 161136895,
      "step": 7453,
      "time_per_iteration": 2.577495574951172
    },
    {
      "auxiliary_loss_clip": 0.01176432,
      "auxiliary_loss_mlp": 0.01028814,
      "balance_loss_clip": 1.04411364,
      "balance_loss_mlp": 1.02050555,
      "epoch": 0.8962905068237841,
      "flos": 12896348567040.0,
      "grad_norm": 3.5024989291799495,
      "language_loss": 0.66050839,
      "learning_rate": 1.1166435445002197e-07,
      "loss": 0.6825608,
      "num_input_tokens_seen": 161154565,
      "step": 7454,
      "time_per_iteration": 3.5235517024993896
    },
    {
      "auxiliary_loss_clip": 0.01173827,
      "auxiliary_loss_mlp": 0.010258,
      "balance_loss_clip": 1.04801118,
      "balance_loss_mlp": 1.01785445,
      "epoch": 0.8964107497144231,
      "flos": 23440331439360.0,
      "grad_norm": 2.044764508235782,
      "language_loss": 0.68766546,
      "learning_rate": 1.1140785386764818e-07,
      "loss": 0.70966172,
      "num_input_tokens_seen": 161173265,
      "step": 7455,
      "time_per_iteration": 3.5141327381134033
    },
    {
      "auxiliary_loss_clip": 0.01164901,
      "auxiliary_loss_mlp": 0.01027232,
      "balance_loss_clip": 1.04540753,
      "balance_loss_mlp": 1.01975179,
      "epoch": 0.8965309926050622,
      "flos": 19500176949120.0,
      "grad_norm": 2.3635851237803798,
      "language_loss": 0.69825876,
      "learning_rate": 1.1115163978320153e-07,
      "loss": 0.72018009,
      "num_input_tokens_seen": 161191995,
      "step": 7456,
      "time_per_iteration": 2.6120355129241943
    },
    {
      "auxiliary_loss_clip": 0.01174926,
      "auxiliary_loss_mlp": 0.01207963,
      "balance_loss_clip": 1.04690146,
      "balance_loss_mlp": 1.00070643,
      "epoch": 0.8966512354957014,
      "flos": 28658008022400.0,
      "grad_norm": 2.0032691268838403,
      "language_loss": 0.82198209,
      "learning_rate": 1.1089571223554917e-07,
      "loss": 0.84581101,
      "num_input_tokens_seen": 161212880,
      "step": 7457,
      "time_per_iteration": 2.65979266166687
    },
    {
      "auxiliary_loss_clip": 0.01170183,
      "auxiliary_loss_mlp": 0.01023627,
      "balance_loss_clip": 1.04475307,
      "balance_loss_mlp": 1.0161947,
      "epoch": 0.8967714783863404,
      "flos": 23370916406400.0,
      "grad_norm": 2.5468714575023506,
      "language_loss": 0.85420942,
      "learning_rate": 1.1064007126351537e-07,
      "loss": 0.87614751,
      "num_input_tokens_seen": 161233595,
      "step": 7458,
      "time_per_iteration": 2.573920488357544
    },
    {
      "auxiliary_loss_clip": 0.01167949,
      "auxiliary_loss_mlp": 0.01022312,
      "balance_loss_clip": 1.0447371,
      "balance_loss_mlp": 1.01492655,
      "epoch": 0.8968917212769795,
      "flos": 24535175938560.0,
      "grad_norm": 2.4234838825150407,
      "language_loss": 0.76715088,
      "learning_rate": 1.1038471690588003e-07,
      "loss": 0.78905356,
      "num_input_tokens_seen": 161252740,
      "step": 7459,
      "time_per_iteration": 3.5697951316833496
    },
    {
      "auxiliary_loss_clip": 0.01174903,
      "auxiliary_loss_mlp": 0.01021878,
      "balance_loss_clip": 1.04343295,
      "balance_loss_mlp": 1.0146482,
      "epoch": 0.8970119641676186,
      "flos": 23475416048640.0,
      "grad_norm": 2.711159925307401,
      "language_loss": 0.80172116,
      "learning_rate": 1.1012964920138145e-07,
      "loss": 0.82368898,
      "num_input_tokens_seen": 161272325,
      "step": 7460,
      "time_per_iteration": 2.6920254230499268
    },
    {
      "auxiliary_loss_clip": 0.01164934,
      "auxiliary_loss_mlp": 0.0102569,
      "balance_loss_clip": 1.04014993,
      "balance_loss_mlp": 1.01864779,
      "epoch": 0.8971322070582577,
      "flos": 24538192680960.0,
      "grad_norm": 1.7057632235705358,
      "language_loss": 0.75753427,
      "learning_rate": 1.0987486818871205e-07,
      "loss": 0.77944052,
      "num_input_tokens_seen": 161295915,
      "step": 7461,
      "time_per_iteration": 2.722466230392456
    },
    {
      "auxiliary_loss_clip": 0.01169736,
      "auxiliary_loss_mlp": 0.01207842,
      "balance_loss_clip": 1.04572284,
      "balance_loss_mlp": 1.00073826,
      "epoch": 0.8972524499488967,
      "flos": 21797454159360.0,
      "grad_norm": 2.711687762682318,
      "language_loss": 0.73468494,
      "learning_rate": 1.0962037390652245e-07,
      "loss": 0.75846076,
      "num_input_tokens_seen": 161314935,
      "step": 7462,
      "time_per_iteration": 2.606426239013672
    },
    {
      "auxiliary_loss_clip": 0.01170705,
      "auxiliary_loss_mlp": 0.01024331,
      "balance_loss_clip": 1.04415655,
      "balance_loss_mlp": 1.01679385,
      "epoch": 0.8973726928395359,
      "flos": 21726243446400.0,
      "grad_norm": 1.7073436877396313,
      "language_loss": 0.71836638,
      "learning_rate": 1.0936616639341911e-07,
      "loss": 0.74031675,
      "num_input_tokens_seen": 161335225,
      "step": 7463,
      "time_per_iteration": 2.6107795238494873
    },
    {
      "auxiliary_loss_clip": 0.01063977,
      "auxiliary_loss_mlp": 0.01000836,
      "balance_loss_clip": 1.01250339,
      "balance_loss_mlp": 1.00016868,
      "epoch": 0.897492935730175,
      "flos": 53837100097920.0,
      "grad_norm": 0.7306323482953813,
      "language_loss": 0.54673994,
      "learning_rate": 1.0911224568796473e-07,
      "loss": 0.56738806,
      "num_input_tokens_seen": 161393420,
      "step": 7464,
      "time_per_iteration": 3.2003986835479736
    },
    {
      "auxiliary_loss_clip": 0.01168207,
      "auxiliary_loss_mlp": 0.01028292,
      "balance_loss_clip": 1.04699588,
      "balance_loss_mlp": 1.02127016,
      "epoch": 0.897613178620814,
      "flos": 18290346036480.0,
      "grad_norm": 2.3802661890580876,
      "language_loss": 0.71453696,
      "learning_rate": 1.0885861182867984e-07,
      "loss": 0.73650193,
      "num_input_tokens_seen": 161411525,
      "step": 7465,
      "time_per_iteration": 2.5679233074188232
    },
    {
      "auxiliary_loss_clip": 0.01175062,
      "auxiliary_loss_mlp": 0.01021141,
      "balance_loss_clip": 1.04414701,
      "balance_loss_mlp": 1.01365197,
      "epoch": 0.8977334215114532,
      "flos": 32993718059520.0,
      "grad_norm": 2.083568548522288,
      "language_loss": 0.70620036,
      "learning_rate": 1.0860526485403942e-07,
      "loss": 0.72816241,
      "num_input_tokens_seen": 161432800,
      "step": 7466,
      "time_per_iteration": 2.7840943336486816
    },
    {
      "auxiliary_loss_clip": 0.0116718,
      "auxiliary_loss_mlp": 0.01023401,
      "balance_loss_clip": 1.04772115,
      "balance_loss_mlp": 1.01639175,
      "epoch": 0.8978536644020922,
      "flos": 15195636938880.0,
      "grad_norm": 1.5931966347671485,
      "language_loss": 0.77229041,
      "learning_rate": 1.0835220480247675e-07,
      "loss": 0.79419619,
      "num_input_tokens_seen": 161451295,
      "step": 7467,
      "time_per_iteration": 2.558580160140991
    },
    {
      "auxiliary_loss_clip": 0.01169012,
      "auxiliary_loss_mlp": 0.01025308,
      "balance_loss_clip": 1.04456055,
      "balance_loss_mlp": 1.01785696,
      "epoch": 0.8979739072927313,
      "flos": 18004389863040.0,
      "grad_norm": 2.2083478875884373,
      "language_loss": 0.8349883,
      "learning_rate": 1.0809943171238067e-07,
      "loss": 0.85693145,
      "num_input_tokens_seen": 161469220,
      "step": 7468,
      "time_per_iteration": 2.6540517807006836
    },
    {
      "auxiliary_loss_clip": 0.01179454,
      "auxiliary_loss_mlp": 0.0102795,
      "balance_loss_clip": 1.04550552,
      "balance_loss_mlp": 1.01920557,
      "epoch": 0.8980941501833704,
      "flos": 22271546793600.0,
      "grad_norm": 3.5458906441386224,
      "language_loss": 0.63002199,
      "learning_rate": 1.078469456220965e-07,
      "loss": 0.65209603,
      "num_input_tokens_seen": 161489375,
      "step": 7469,
      "time_per_iteration": 2.6129045486450195
    },
    {
      "auxiliary_loss_clip": 0.01168467,
      "auxiliary_loss_mlp": 0.01022635,
      "balance_loss_clip": 1.04404402,
      "balance_loss_mlp": 1.01536632,
      "epoch": 0.8982143930740095,
      "flos": 37560729726720.0,
      "grad_norm": 1.7522653419608054,
      "language_loss": 0.69603693,
      "learning_rate": 1.0759474656992606e-07,
      "loss": 0.71794796,
      "num_input_tokens_seen": 161512145,
      "step": 7470,
      "time_per_iteration": 2.748039484024048
    },
    {
      "auxiliary_loss_clip": 0.01176499,
      "auxiliary_loss_mlp": 0.01025532,
      "balance_loss_clip": 1.04403222,
      "balance_loss_mlp": 1.0176909,
      "epoch": 0.8983346359646486,
      "flos": 18076893465600.0,
      "grad_norm": 2.2526068667370867,
      "language_loss": 0.77999973,
      "learning_rate": 1.0734283459412785e-07,
      "loss": 0.80202007,
      "num_input_tokens_seen": 161528995,
      "step": 7471,
      "time_per_iteration": 3.5187230110168457
    },
    {
      "auxiliary_loss_clip": 0.01182371,
      "auxiliary_loss_mlp": 0.01026838,
      "balance_loss_clip": 1.04159653,
      "balance_loss_mlp": 1.01845169,
      "epoch": 0.8984548788552876,
      "flos": 20558895344640.0,
      "grad_norm": 1.7564153567135563,
      "language_loss": 0.80700123,
      "learning_rate": 1.0709120973291707e-07,
      "loss": 0.82909328,
      "num_input_tokens_seen": 161548775,
      "step": 7472,
      "time_per_iteration": 2.634603500366211
    },
    {
      "auxiliary_loss_clip": 0.01169657,
      "auxiliary_loss_mlp": 0.01027157,
      "balance_loss_clip": 1.04859626,
      "balance_loss_mlp": 1.01931334,
      "epoch": 0.8985751217459268,
      "flos": 17785442511360.0,
      "grad_norm": 2.027511020708571,
      "language_loss": 0.77554178,
      "learning_rate": 1.0683987202446475e-07,
      "loss": 0.79750991,
      "num_input_tokens_seen": 161566960,
      "step": 7473,
      "time_per_iteration": 2.627959966659546
    },
    {
      "auxiliary_loss_clip": 0.01173144,
      "auxiliary_loss_mlp": 0.0102192,
      "balance_loss_clip": 1.04634547,
      "balance_loss_mlp": 1.01435888,
      "epoch": 0.8986953646365659,
      "flos": 21617003208960.0,
      "grad_norm": 2.019512202088956,
      "language_loss": 0.69799751,
      "learning_rate": 1.0658882150689862e-07,
      "loss": 0.71994817,
      "num_input_tokens_seen": 161585820,
      "step": 7474,
      "time_per_iteration": 2.595921516418457
    },
    {
      "auxiliary_loss_clip": 0.01177936,
      "auxiliary_loss_mlp": 0.0102091,
      "balance_loss_clip": 1.04322529,
      "balance_loss_mlp": 1.01325631,
      "epoch": 0.8988156075272049,
      "flos": 14027355083520.0,
      "grad_norm": 2.7268207954957218,
      "language_loss": 0.78048217,
      "learning_rate": 1.0633805821830288e-07,
      "loss": 0.80247056,
      "num_input_tokens_seen": 161602505,
      "step": 7475,
      "time_per_iteration": 2.711996078491211
    },
    {
      "auxiliary_loss_clip": 0.01171766,
      "auxiliary_loss_mlp": 0.01026388,
      "balance_loss_clip": 1.04443908,
      "balance_loss_mlp": 1.01866889,
      "epoch": 0.8989358504178441,
      "flos": 29059202004480.0,
      "grad_norm": 2.7946048449468437,
      "language_loss": 0.83128107,
      "learning_rate": 1.0608758219671753e-07,
      "loss": 0.85326254,
      "num_input_tokens_seen": 161621545,
      "step": 7476,
      "time_per_iteration": 2.766005754470825
    },
    {
      "auxiliary_loss_clip": 0.01176702,
      "auxiliary_loss_mlp": 0.0102187,
      "balance_loss_clip": 1.04470277,
      "balance_loss_mlp": 1.01496828,
      "epoch": 0.8990560933084831,
      "flos": 20230420446720.0,
      "grad_norm": 1.8331037886677637,
      "language_loss": 0.70545989,
      "learning_rate": 1.0583739348014065e-07,
      "loss": 0.7274456,
      "num_input_tokens_seen": 161642630,
      "step": 7477,
      "time_per_iteration": 2.722236156463623
    },
    {
      "auxiliary_loss_clip": 0.01168973,
      "auxiliary_loss_mlp": 0.01022255,
      "balance_loss_clip": 1.05048704,
      "balance_loss_mlp": 1.01514995,
      "epoch": 0.8991763361991222,
      "flos": 25520672459520.0,
      "grad_norm": 2.2219806967781714,
      "language_loss": 0.84459931,
      "learning_rate": 1.0558749210652518e-07,
      "loss": 0.86651158,
      "num_input_tokens_seen": 161662560,
      "step": 7478,
      "time_per_iteration": 2.583010196685791
    },
    {
      "auxiliary_loss_clip": 0.01178252,
      "auxiliary_loss_mlp": 0.01020354,
      "balance_loss_clip": 1.04326057,
      "balance_loss_mlp": 1.01313925,
      "epoch": 0.8992965790897613,
      "flos": 25119191168640.0,
      "grad_norm": 3.187330205901446,
      "language_loss": 0.85671133,
      "learning_rate": 1.053378781137808e-07,
      "loss": 0.8786974,
      "num_input_tokens_seen": 161683480,
      "step": 7479,
      "time_per_iteration": 2.9205949306488037
    },
    {
      "auxiliary_loss_clip": 0.01175427,
      "auxiliary_loss_mlp": 0.01026673,
      "balance_loss_clip": 1.04502821,
      "balance_loss_mlp": 1.01915085,
      "epoch": 0.8994168219804004,
      "flos": 16070815814400.0,
      "grad_norm": 4.509395787672912,
      "language_loss": 0.78082216,
      "learning_rate": 1.0508855153977392e-07,
      "loss": 0.80284321,
      "num_input_tokens_seen": 161699945,
      "step": 7480,
      "time_per_iteration": 3.74355149269104
    },
    {
      "auxiliary_loss_clip": 0.01170485,
      "auxiliary_loss_mlp": 0.01028508,
      "balance_loss_clip": 1.04399037,
      "balance_loss_mlp": 1.02069402,
      "epoch": 0.8995370648710395,
      "flos": 24825764966400.0,
      "grad_norm": 3.537071057476907,
      "language_loss": 0.67141509,
      "learning_rate": 1.0483951242232669e-07,
      "loss": 0.69340497,
      "num_input_tokens_seen": 161720420,
      "step": 7481,
      "time_per_iteration": 2.7137033939361572
    },
    {
      "auxiliary_loss_clip": 0.01063214,
      "auxiliary_loss_mlp": 0.01000844,
      "balance_loss_clip": 1.01053524,
      "balance_loss_mlp": 1.00009012,
      "epoch": 0.8996573077616786,
      "flos": 63116238378240.0,
      "grad_norm": 0.9736555217092626,
      "language_loss": 0.57676125,
      "learning_rate": 1.0459076079921936e-07,
      "loss": 0.59740186,
      "num_input_tokens_seen": 161773080,
      "step": 7482,
      "time_per_iteration": 4.014577388763428
    },
    {
      "auxiliary_loss_clip": 0.01166674,
      "auxiliary_loss_mlp": 0.0102959,
      "balance_loss_clip": 1.04412794,
      "balance_loss_mlp": 1.02168345,
      "epoch": 0.8997775506523177,
      "flos": 18219674027520.0,
      "grad_norm": 2.5427980741647067,
      "language_loss": 0.85520148,
      "learning_rate": 1.0434229670818618e-07,
      "loss": 0.87716413,
      "num_input_tokens_seen": 161789755,
      "step": 7483,
      "time_per_iteration": 2.8038957118988037
    },
    {
      "auxiliary_loss_clip": 0.01162226,
      "auxiliary_loss_mlp": 0.01023696,
      "balance_loss_clip": 1.04151595,
      "balance_loss_mlp": 1.0159502,
      "epoch": 0.8998977935429567,
      "flos": 24166768095360.0,
      "grad_norm": 1.6671388849367084,
      "language_loss": 0.7986871,
      "learning_rate": 1.0409412018691944e-07,
      "loss": 0.82054627,
      "num_input_tokens_seen": 161810220,
      "step": 7484,
      "time_per_iteration": 2.623904228210449
    },
    {
      "auxiliary_loss_clip": 0.01169402,
      "auxiliary_loss_mlp": 0.01023824,
      "balance_loss_clip": 1.04569864,
      "balance_loss_mlp": 1.01652813,
      "epoch": 0.9000180364335959,
      "flos": 20773030273920.0,
      "grad_norm": 2.6903753014117893,
      "language_loss": 0.75233209,
      "learning_rate": 1.0384623127306724e-07,
      "loss": 0.77426434,
      "num_input_tokens_seen": 161827565,
      "step": 7485,
      "time_per_iteration": 2.6328420639038086
    },
    {
      "auxiliary_loss_clip": 0.01172064,
      "auxiliary_loss_mlp": 0.01022272,
      "balance_loss_clip": 1.04198933,
      "balance_loss_mlp": 1.0154618,
      "epoch": 0.900138279324235,
      "flos": 19205745166080.0,
      "grad_norm": 1.9801551588122195,
      "language_loss": 0.79252857,
      "learning_rate": 1.0359863000423397e-07,
      "loss": 0.81447196,
      "num_input_tokens_seen": 161845700,
      "step": 7486,
      "time_per_iteration": 3.6092565059661865
    },
    {
      "auxiliary_loss_clip": 0.01167884,
      "auxiliary_loss_mlp": 0.01024649,
      "balance_loss_clip": 1.04808629,
      "balance_loss_mlp": 1.01750505,
      "epoch": 0.900258522214874,
      "flos": 28731158069760.0,
      "grad_norm": 1.70573558297886,
      "language_loss": 0.71986806,
      "learning_rate": 1.0335131641798112e-07,
      "loss": 0.74179345,
      "num_input_tokens_seen": 161867660,
      "step": 7487,
      "time_per_iteration": 2.6388676166534424
    },
    {
      "auxiliary_loss_clip": 0.01076495,
      "auxiliary_loss_mlp": 0.01001546,
      "balance_loss_clip": 1.01027179,
      "balance_loss_mlp": 1.00081921,
      "epoch": 0.9003787651055132,
      "flos": 58280685655680.0,
      "grad_norm": 0.8285284761338851,
      "language_loss": 0.55583221,
      "learning_rate": 1.0310429055182512e-07,
      "loss": 0.57661265,
      "num_input_tokens_seen": 161921980,
      "step": 7488,
      "time_per_iteration": 3.0554585456848145
    },
    {
      "auxiliary_loss_clip": 0.0117646,
      "auxiliary_loss_mlp": 0.01029117,
      "balance_loss_clip": 1.0432142,
      "balance_loss_mlp": 1.02146375,
      "epoch": 0.9004990079961522,
      "flos": 25556475340800.0,
      "grad_norm": 1.667035656960349,
      "language_loss": 0.74434406,
      "learning_rate": 1.0285755244324024e-07,
      "loss": 0.76639986,
      "num_input_tokens_seen": 161942725,
      "step": 7489,
      "time_per_iteration": 2.77152419090271
    },
    {
      "auxiliary_loss_clip": 0.01173634,
      "auxiliary_loss_mlp": 0.01207299,
      "balance_loss_clip": 1.04219389,
      "balance_loss_mlp": 1.00068855,
      "epoch": 0.9006192508867913,
      "flos": 23335185352320.0,
      "grad_norm": 1.4651490968353578,
      "language_loss": 0.68872923,
      "learning_rate": 1.0261110212965629e-07,
      "loss": 0.7125386,
      "num_input_tokens_seen": 161964520,
      "step": 7490,
      "time_per_iteration": 2.7855536937713623
    },
    {
      "auxiliary_loss_clip": 0.01173523,
      "auxiliary_loss_mlp": 0.01023626,
      "balance_loss_clip": 1.04549921,
      "balance_loss_mlp": 1.01649988,
      "epoch": 0.9007394937774305,
      "flos": 18040300485120.0,
      "grad_norm": 2.165260028037673,
      "language_loss": 0.78851622,
      "learning_rate": 1.023649396484596e-07,
      "loss": 0.81048775,
      "num_input_tokens_seen": 161983575,
      "step": 7491,
      "time_per_iteration": 2.6501357555389404
    },
    {
      "auxiliary_loss_clip": 0.0116658,
      "auxiliary_loss_mlp": 0.01026591,
      "balance_loss_clip": 1.04668689,
      "balance_loss_mlp": 1.01946497,
      "epoch": 0.9008597366680695,
      "flos": 43068456633600.0,
      "grad_norm": 2.7089905625550923,
      "language_loss": 0.67751974,
      "learning_rate": 1.0211906503699275e-07,
      "loss": 0.69945145,
      "num_input_tokens_seen": 162006550,
      "step": 7492,
      "time_per_iteration": 2.7971572875976562
    },
    {
      "auxiliary_loss_clip": 0.01174452,
      "auxiliary_loss_mlp": 0.01025572,
      "balance_loss_clip": 1.04903698,
      "balance_loss_mlp": 1.01733446,
      "epoch": 0.9009799795587086,
      "flos": 14939055112320.0,
      "grad_norm": 3.0186664834687735,
      "language_loss": 0.82360184,
      "learning_rate": 1.0187347833255455e-07,
      "loss": 0.84560215,
      "num_input_tokens_seen": 162022455,
      "step": 7493,
      "time_per_iteration": 2.5819478034973145
    },
    {
      "auxiliary_loss_clip": 0.0116544,
      "auxiliary_loss_mlp": 0.01025204,
      "balance_loss_clip": 1.04799771,
      "balance_loss_mlp": 1.0179441,
      "epoch": 0.9011002224493477,
      "flos": 21579584215680.0,
      "grad_norm": 1.7621898476280782,
      "language_loss": 0.79028749,
      "learning_rate": 1.0162817957240056e-07,
      "loss": 0.81219393,
      "num_input_tokens_seen": 162042350,
      "step": 7494,
      "time_per_iteration": 2.6382532119750977
    },
    {
      "auxiliary_loss_clip": 0.01070511,
      "auxiliary_loss_mlp": 0.01001491,
      "balance_loss_clip": 1.01102114,
      "balance_loss_mlp": 1.00072241,
      "epoch": 0.9012204653399868,
      "flos": 71166367883520.0,
      "grad_norm": 0.8860795774101554,
      "language_loss": 0.62992561,
      "learning_rate": 1.0138316879374253e-07,
      "loss": 0.65064561,
      "num_input_tokens_seen": 162111640,
      "step": 7495,
      "time_per_iteration": 3.3165504932403564
    },
    {
      "auxiliary_loss_clip": 0.01176109,
      "auxiliary_loss_mlp": 0.01021387,
      "balance_loss_clip": 1.04736066,
      "balance_loss_mlp": 1.01376915,
      "epoch": 0.9013407082306258,
      "flos": 15594963413760.0,
      "grad_norm": 5.164302403161177,
      "language_loss": 0.74431694,
      "learning_rate": 1.0113844603374833e-07,
      "loss": 0.76629186,
      "num_input_tokens_seen": 162128165,
      "step": 7496,
      "time_per_iteration": 2.65054988861084
    },
    {
      "auxiliary_loss_clip": 0.01172379,
      "auxiliary_loss_mlp": 0.01024974,
      "balance_loss_clip": 1.04319942,
      "balance_loss_mlp": 1.0166707,
      "epoch": 0.901460951121265,
      "flos": 15049157276160.0,
      "grad_norm": 2.7193396178298785,
      "language_loss": 0.72297049,
      "learning_rate": 1.0089401132954178e-07,
      "loss": 0.74494398,
      "num_input_tokens_seen": 162146145,
      "step": 7497,
      "time_per_iteration": 2.7320382595062256
    },
    {
      "auxiliary_loss_clip": 0.0117074,
      "auxiliary_loss_mlp": 0.01021592,
      "balance_loss_clip": 1.04502881,
      "balance_loss_mlp": 1.01468968,
      "epoch": 0.9015811940119041,
      "flos": 22236857233920.0,
      "grad_norm": 1.866922830445223,
      "language_loss": 0.72699189,
      "learning_rate": 1.006498647182037e-07,
      "loss": 0.7489152,
      "num_input_tokens_seen": 162164800,
      "step": 7498,
      "time_per_iteration": 3.5768487453460693
    },
    {
      "auxiliary_loss_clip": 0.01177262,
      "auxiliary_loss_mlp": 0.0102792,
      "balance_loss_clip": 1.03958678,
      "balance_loss_mlp": 1.02050245,
      "epoch": 0.9017014369025431,
      "flos": 24973824827520.0,
      "grad_norm": 2.4455714331390492,
      "language_loss": 0.72078967,
      "learning_rate": 1.004060062367713e-07,
      "loss": 0.74284142,
      "num_input_tokens_seen": 162185895,
      "step": 7499,
      "time_per_iteration": 2.8354079723358154
    },
    {
      "auxiliary_loss_clip": 0.01172419,
      "auxiliary_loss_mlp": 0.01024629,
      "balance_loss_clip": 1.04668617,
      "balance_loss_mlp": 1.01683831,
      "epoch": 0.9018216797931822,
      "flos": 18114168804480.0,
      "grad_norm": 1.9182233783649598,
      "language_loss": 0.6955421,
      "learning_rate": 1.0016243592223728e-07,
      "loss": 0.71751261,
      "num_input_tokens_seen": 162206295,
      "step": 7500,
      "time_per_iteration": 2.729393243789673
    },
    {
      "auxiliary_loss_clip": 0.01177621,
      "auxiliary_loss_mlp": 0.01022787,
      "balance_loss_clip": 1.04007566,
      "balance_loss_mlp": 1.01524365,
      "epoch": 0.9019419226838213,
      "flos": 37268452759680.0,
      "grad_norm": 2.0349030376475055,
      "language_loss": 0.65319705,
      "learning_rate": 9.991915381155114e-08,
      "loss": 0.67520106,
      "num_input_tokens_seen": 162229275,
      "step": 7501,
      "time_per_iteration": 2.831906318664551
    },
    {
      "auxiliary_loss_clip": 0.01173219,
      "auxiliary_loss_mlp": 0.01025082,
      "balance_loss_clip": 1.04641604,
      "balance_loss_mlp": 1.01757812,
      "epoch": 0.9020621655744604,
      "flos": 23441121538560.0,
      "grad_norm": 2.899332084596431,
      "language_loss": 0.75032163,
      "learning_rate": 9.967615994161871e-08,
      "loss": 0.77230465,
      "num_input_tokens_seen": 162248935,
      "step": 7502,
      "time_per_iteration": 2.6831977367401123
    },
    {
      "auxiliary_loss_clip": 0.01166292,
      "auxiliary_loss_mlp": 0.01022736,
      "balance_loss_clip": 1.04685807,
      "balance_loss_mlp": 1.01554203,
      "epoch": 0.9021824084650995,
      "flos": 22857465444480.0,
      "grad_norm": 1.9099252451617006,
      "language_loss": 0.78273004,
      "learning_rate": 9.943345434930161e-08,
      "loss": 0.80462039,
      "num_input_tokens_seen": 162269185,
      "step": 7503,
      "time_per_iteration": 2.5894927978515625
    },
    {
      "auxiliary_loss_clip": 0.0117448,
      "auxiliary_loss_mlp": 0.01027079,
      "balance_loss_clip": 1.0449996,
      "balance_loss_mlp": 1.01978898,
      "epoch": 0.9023026513557386,
      "flos": 22127581082880.0,
      "grad_norm": 2.0893844328364977,
      "language_loss": 0.69327694,
      "learning_rate": 9.919103707141885e-08,
      "loss": 0.71529251,
      "num_input_tokens_seen": 162288065,
      "step": 7504,
      "time_per_iteration": 2.696180820465088
    },
    {
      "auxiliary_loss_clip": 0.01166772,
      "auxiliary_loss_mlp": 0.0102441,
      "balance_loss_clip": 1.04467201,
      "balance_loss_mlp": 1.01618409,
      "epoch": 0.9024228942463777,
      "flos": 24199087357440.0,
      "grad_norm": 1.9470161279166627,
      "language_loss": 0.7645247,
      "learning_rate": 9.89489081447441e-08,
      "loss": 0.78643656,
      "num_input_tokens_seen": 162305265,
      "step": 7505,
      "time_per_iteration": 2.5932440757751465
    },
    {
      "auxiliary_loss_clip": 0.01172245,
      "auxiliary_loss_mlp": 0.01023054,
      "balance_loss_clip": 1.04373419,
      "balance_loss_mlp": 1.01497757,
      "epoch": 0.9025431371370167,
      "flos": 25008262992000.0,
      "grad_norm": 1.7801431019492984,
      "language_loss": 0.83124208,
      "learning_rate": 9.870706760600844e-08,
      "loss": 0.85319507,
      "num_input_tokens_seen": 162325215,
      "step": 7506,
      "time_per_iteration": 2.7050654888153076
    },
    {
      "auxiliary_loss_clip": 0.01186113,
      "auxiliary_loss_mlp": 0.01024504,
      "balance_loss_clip": 1.04681611,
      "balance_loss_mlp": 1.01664245,
      "epoch": 0.9026633800276559,
      "flos": 18952862440320.0,
      "grad_norm": 1.853391480923401,
      "language_loss": 0.73161972,
      "learning_rate": 9.846551549189918e-08,
      "loss": 0.75372589,
      "num_input_tokens_seen": 162344820,
      "step": 7507,
      "time_per_iteration": 3.660769462585449
    },
    {
      "auxiliary_loss_clip": 0.01172401,
      "auxiliary_loss_mlp": 0.01024853,
      "balance_loss_clip": 1.04609179,
      "balance_loss_mlp": 1.01736426,
      "epoch": 0.902783622918295,
      "flos": 32416059536640.0,
      "grad_norm": 3.148451867535307,
      "language_loss": 0.68672085,
      "learning_rate": 9.822425183905902e-08,
      "loss": 0.70869339,
      "num_input_tokens_seen": 162365345,
      "step": 7508,
      "time_per_iteration": 2.768596649169922
    },
    {
      "auxiliary_loss_clip": 0.01084577,
      "auxiliary_loss_mlp": 0.00999752,
      "balance_loss_clip": 1.01158559,
      "balance_loss_mlp": 0.99901599,
      "epoch": 0.902903865808934,
      "flos": 63717453244800.0,
      "grad_norm": 0.9144730342937663,
      "language_loss": 0.75108314,
      "learning_rate": 9.798327668408823e-08,
      "loss": 0.77192652,
      "num_input_tokens_seen": 162426980,
      "step": 7509,
      "time_per_iteration": 4.187403202056885
    },
    {
      "auxiliary_loss_clip": 0.01171084,
      "auxiliary_loss_mlp": 0.01025941,
      "balance_loss_clip": 1.04807091,
      "balance_loss_mlp": 1.018121,
      "epoch": 0.9030241086995732,
      "flos": 23804034600960.0,
      "grad_norm": 2.6670864291222305,
      "language_loss": 0.6886158,
      "learning_rate": 9.774259006354158e-08,
      "loss": 0.71058607,
      "num_input_tokens_seen": 162447050,
      "step": 7510,
      "time_per_iteration": 2.6720244884490967
    },
    {
      "auxiliary_loss_clip": 0.01175913,
      "auxiliary_loss_mlp": 0.01028068,
      "balance_loss_clip": 1.04393482,
      "balance_loss_mlp": 1.02047443,
      "epoch": 0.9031443515902122,
      "flos": 26395887248640.0,
      "grad_norm": 2.573991012155795,
      "language_loss": 0.76409268,
      "learning_rate": 9.750219201393184e-08,
      "loss": 0.78613245,
      "num_input_tokens_seen": 162467015,
      "step": 7511,
      "time_per_iteration": 2.6865479946136475
    },
    {
      "auxiliary_loss_clip": 0.01167172,
      "auxiliary_loss_mlp": 0.01017586,
      "balance_loss_clip": 1.04440713,
      "balance_loss_mlp": 1.01035559,
      "epoch": 0.9032645944808513,
      "flos": 24939350749440.0,
      "grad_norm": 1.784130434677874,
      "language_loss": 0.77568841,
      "learning_rate": 9.726208257172697e-08,
      "loss": 0.79753602,
      "num_input_tokens_seen": 162488710,
      "step": 7512,
      "time_per_iteration": 2.714064598083496
    },
    {
      "auxiliary_loss_clip": 0.01165161,
      "auxiliary_loss_mlp": 0.01019618,
      "balance_loss_clip": 1.04633486,
      "balance_loss_mlp": 1.01209569,
      "epoch": 0.9033848373714904,
      "flos": 21178821196800.0,
      "grad_norm": 2.1382696525631752,
      "language_loss": 0.75088274,
      "learning_rate": 9.702226177335115e-08,
      "loss": 0.77273047,
      "num_input_tokens_seen": 162507205,
      "step": 7513,
      "time_per_iteration": 3.530181407928467
    },
    {
      "auxiliary_loss_clip": 0.01171523,
      "auxiliary_loss_mlp": 0.01033754,
      "balance_loss_clip": 1.04543996,
      "balance_loss_mlp": 1.02558875,
      "epoch": 0.9035050802621295,
      "flos": 26286359702400.0,
      "grad_norm": 1.6473849812879753,
      "language_loss": 0.724226,
      "learning_rate": 9.67827296551853e-08,
      "loss": 0.74627876,
      "num_input_tokens_seen": 162528490,
      "step": 7514,
      "time_per_iteration": 2.7220022678375244
    },
    {
      "auxiliary_loss_clip": 0.01163225,
      "auxiliary_loss_mlp": 0.01207731,
      "balance_loss_clip": 1.04075432,
      "balance_loss_mlp": 1.00075364,
      "epoch": 0.9036253231527686,
      "flos": 24204546224640.0,
      "grad_norm": 2.2379724012566644,
      "language_loss": 0.68344605,
      "learning_rate": 9.65434862535659e-08,
      "loss": 0.70715559,
      "num_input_tokens_seen": 162547860,
      "step": 7515,
      "time_per_iteration": 2.6099584102630615
    },
    {
      "auxiliary_loss_clip": 0.01174173,
      "auxiliary_loss_mlp": 0.01028669,
      "balance_loss_clip": 1.04306221,
      "balance_loss_mlp": 1.02088451,
      "epoch": 0.9037455660434077,
      "flos": 18072655660800.0,
      "grad_norm": 2.8718236216700728,
      "language_loss": 0.64944565,
      "learning_rate": 9.630453160478635e-08,
      "loss": 0.6714741,
      "num_input_tokens_seen": 162563215,
      "step": 7516,
      "time_per_iteration": 2.598499059677124
    },
    {
      "auxiliary_loss_clip": 0.01178991,
      "auxiliary_loss_mlp": 0.01022949,
      "balance_loss_clip": 1.04185903,
      "balance_loss_mlp": 1.01586795,
      "epoch": 0.9038658089340468,
      "flos": 24060795995520.0,
      "grad_norm": 2.0075233877290803,
      "language_loss": 0.8253932,
      "learning_rate": 9.60658657450959e-08,
      "loss": 0.84741259,
      "num_input_tokens_seen": 162583515,
      "step": 7517,
      "time_per_iteration": 2.6791393756866455
    },
    {
      "auxiliary_loss_clip": 0.0115652,
      "auxiliary_loss_mlp": 0.01025235,
      "balance_loss_clip": 1.03978062,
      "balance_loss_mlp": 1.01775467,
      "epoch": 0.9039860518246858,
      "flos": 21834298535040.0,
      "grad_norm": 1.6366785481355637,
      "language_loss": 0.79389179,
      "learning_rate": 9.582748871069979e-08,
      "loss": 0.81570935,
      "num_input_tokens_seen": 162602955,
      "step": 7518,
      "time_per_iteration": 2.690481662750244
    },
    {
      "auxiliary_loss_clip": 0.01173964,
      "auxiliary_loss_mlp": 0.0120704,
      "balance_loss_clip": 1.04387724,
      "balance_loss_mlp": 1.00067782,
      "epoch": 0.904106294715325,
      "flos": 26614870513920.0,
      "grad_norm": 1.9621688434369966,
      "language_loss": 0.83435071,
      "learning_rate": 9.558940053775954e-08,
      "loss": 0.85816073,
      "num_input_tokens_seen": 162621595,
      "step": 7519,
      "time_per_iteration": 2.6352386474609375
    },
    {
      "auxiliary_loss_clip": 0.01166997,
      "auxiliary_loss_mlp": 0.01027053,
      "balance_loss_clip": 1.04517794,
      "balance_loss_mlp": 1.01965034,
      "epoch": 0.904226537605964,
      "flos": 17785693906560.0,
      "grad_norm": 1.832562020773942,
      "language_loss": 0.67912626,
      "learning_rate": 9.535160126239294e-08,
      "loss": 0.70106673,
      "num_input_tokens_seen": 162638220,
      "step": 7520,
      "time_per_iteration": 2.6525471210479736
    },
    {
      "auxiliary_loss_clip": 0.01166295,
      "auxiliary_loss_mlp": 0.01024313,
      "balance_loss_clip": 1.04533625,
      "balance_loss_mlp": 1.01694298,
      "epoch": 0.9043467804966031,
      "flos": 24790428961920.0,
      "grad_norm": 1.536887169736648,
      "language_loss": 0.70650238,
      "learning_rate": 9.511409092067424e-08,
      "loss": 0.72840846,
      "num_input_tokens_seen": 162658575,
      "step": 7521,
      "time_per_iteration": 2.701843738555908
    },
    {
      "auxiliary_loss_clip": 0.0117399,
      "auxiliary_loss_mlp": 0.01027216,
      "balance_loss_clip": 1.04627752,
      "balance_loss_mlp": 1.01977694,
      "epoch": 0.9044670233872423,
      "flos": 22632125472000.0,
      "grad_norm": 1.7286373154452428,
      "language_loss": 0.67462873,
      "learning_rate": 9.487686954863327e-08,
      "loss": 0.69664085,
      "num_input_tokens_seen": 162678295,
      "step": 7522,
      "time_per_iteration": 2.8279435634613037
    },
    {
      "auxiliary_loss_clip": 0.0116902,
      "auxiliary_loss_mlp": 0.01021709,
      "balance_loss_clip": 1.04651141,
      "balance_loss_mlp": 1.01448178,
      "epoch": 0.9045872662778813,
      "flos": 23771320289280.0,
      "grad_norm": 2.3367253446256178,
      "language_loss": 0.77662438,
      "learning_rate": 9.46399371822566e-08,
      "loss": 0.79853165,
      "num_input_tokens_seen": 162698070,
      "step": 7523,
      "time_per_iteration": 2.6570072174072266
    },
    {
      "auxiliary_loss_clip": 0.01168632,
      "auxiliary_loss_mlp": 0.01023108,
      "balance_loss_clip": 1.04864371,
      "balance_loss_mlp": 1.01524019,
      "epoch": 0.9047075091685204,
      "flos": 15191039998080.0,
      "grad_norm": 1.9644707173800788,
      "language_loss": 0.72381341,
      "learning_rate": 9.440329385748657e-08,
      "loss": 0.74573082,
      "num_input_tokens_seen": 162715140,
      "step": 7524,
      "time_per_iteration": 2.608849287033081
    },
    {
      "auxiliary_loss_clip": 0.01174439,
      "auxiliary_loss_mlp": 0.01017232,
      "balance_loss_clip": 1.04439914,
      "balance_loss_mlp": 1.01066661,
      "epoch": 0.9048277520591596,
      "flos": 18003707504640.0,
      "grad_norm": 1.925527136835355,
      "language_loss": 0.70737487,
      "learning_rate": 9.416693961022137e-08,
      "loss": 0.72929162,
      "num_input_tokens_seen": 162733390,
      "step": 7525,
      "time_per_iteration": 3.5723772048950195
    },
    {
      "auxiliary_loss_clip": 0.01166161,
      "auxiliary_loss_mlp": 0.01022642,
      "balance_loss_clip": 1.03758931,
      "balance_loss_mlp": 1.0152514,
      "epoch": 0.9049479949497986,
      "flos": 21872471713920.0,
      "grad_norm": 1.8674053165416384,
      "language_loss": 0.77327961,
      "learning_rate": 9.393087447631654e-08,
      "loss": 0.79516768,
      "num_input_tokens_seen": 162751670,
      "step": 7526,
      "time_per_iteration": 2.7062225341796875
    },
    {
      "auxiliary_loss_clip": 0.01173007,
      "auxiliary_loss_mlp": 0.01021806,
      "balance_loss_clip": 1.04395628,
      "balance_loss_mlp": 1.01485598,
      "epoch": 0.9050682378404377,
      "flos": 20773928113920.0,
      "grad_norm": 2.361663486822037,
      "language_loss": 0.73045379,
      "learning_rate": 9.36950984915823e-08,
      "loss": 0.75240189,
      "num_input_tokens_seen": 162770025,
      "step": 7527,
      "time_per_iteration": 2.65461802482605
    },
    {
      "auxiliary_loss_clip": 0.01169168,
      "auxiliary_loss_mlp": 0.01024929,
      "balance_loss_clip": 1.0496788,
      "balance_loss_mlp": 1.01751423,
      "epoch": 0.9051884807310768,
      "flos": 21580015178880.0,
      "grad_norm": 1.9764835323686156,
      "language_loss": 0.69374877,
      "learning_rate": 9.345961169178607e-08,
      "loss": 0.71568972,
      "num_input_tokens_seen": 162789710,
      "step": 7528,
      "time_per_iteration": 2.5788702964782715
    },
    {
      "auxiliary_loss_clip": 0.01160425,
      "auxiliary_loss_mlp": 0.0102233,
      "balance_loss_clip": 1.04497457,
      "balance_loss_mlp": 1.01531136,
      "epoch": 0.9053087236217159,
      "flos": 21908059113600.0,
      "grad_norm": 1.4857179730960597,
      "language_loss": 0.728306,
      "learning_rate": 9.322441411265081e-08,
      "loss": 0.75013357,
      "num_input_tokens_seen": 162810695,
      "step": 7529,
      "time_per_iteration": 2.859023332595825
    },
    {
      "auxiliary_loss_clip": 0.01168425,
      "auxiliary_loss_mlp": 0.01025828,
      "balance_loss_clip": 1.044842,
      "balance_loss_mlp": 1.01854467,
      "epoch": 0.9054289665123549,
      "flos": 17055809544960.0,
      "grad_norm": 3.2038760350308197,
      "language_loss": 0.73478037,
      "learning_rate": 9.298950578985554e-08,
      "loss": 0.75672293,
      "num_input_tokens_seen": 162827770,
      "step": 7530,
      "time_per_iteration": 2.565423011779785
    },
    {
      "auxiliary_loss_clip": 0.0116528,
      "auxiliary_loss_mlp": 0.01208286,
      "balance_loss_clip": 1.04605091,
      "balance_loss_mlp": 1.00058484,
      "epoch": 0.905549209402994,
      "flos": 20777268078720.0,
      "grad_norm": 2.5029878899155116,
      "language_loss": 0.70901203,
      "learning_rate": 9.275488675903665e-08,
      "loss": 0.73274761,
      "num_input_tokens_seen": 162846715,
      "step": 7531,
      "time_per_iteration": 2.6314098834991455
    },
    {
      "auxiliary_loss_clip": 0.01174635,
      "auxiliary_loss_mlp": 0.01022645,
      "balance_loss_clip": 1.04168284,
      "balance_loss_mlp": 1.01513505,
      "epoch": 0.9056694522936332,
      "flos": 21686813291520.0,
      "grad_norm": 2.2201180776635576,
      "language_loss": 0.73762715,
      "learning_rate": 9.252055705578454e-08,
      "loss": 0.75959992,
      "num_input_tokens_seen": 162866215,
      "step": 7532,
      "time_per_iteration": 2.6453731060028076
    },
    {
      "auxiliary_loss_clip": 0.01166939,
      "auxiliary_loss_mlp": 0.01023615,
      "balance_loss_clip": 1.04371858,
      "balance_loss_mlp": 1.01680541,
      "epoch": 0.9057896951842722,
      "flos": 29569133433600.0,
      "grad_norm": 1.6738554442694897,
      "language_loss": 0.72236478,
      "learning_rate": 9.228651671564747e-08,
      "loss": 0.74427032,
      "num_input_tokens_seen": 162888245,
      "step": 7533,
      "time_per_iteration": 2.853027105331421
    },
    {
      "auxiliary_loss_clip": 0.01173147,
      "auxiliary_loss_mlp": 0.0102273,
      "balance_loss_clip": 1.0438211,
      "balance_loss_mlp": 1.01550913,
      "epoch": 0.9059099380749113,
      "flos": 27892248952320.0,
      "grad_norm": 1.7250645635631106,
      "language_loss": 0.77964509,
      "learning_rate": 9.205276577412901e-08,
      "loss": 0.80160379,
      "num_input_tokens_seen": 162911025,
      "step": 7534,
      "time_per_iteration": 3.769404649734497
    },
    {
      "auxiliary_loss_clip": 0.01179254,
      "auxiliary_loss_mlp": 0.0120731,
      "balance_loss_clip": 1.04431129,
      "balance_loss_mlp": 1.00065112,
      "epoch": 0.9060301809655504,
      "flos": 17748993185280.0,
      "grad_norm": 2.7245463301626223,
      "language_loss": 0.76521194,
      "learning_rate": 9.181930426668905e-08,
      "loss": 0.78907764,
      "num_input_tokens_seen": 162927820,
      "step": 7535,
      "time_per_iteration": 2.6613810062408447
    },
    {
      "auxiliary_loss_clip": 0.01171684,
      "auxiliary_loss_mlp": 0.01024048,
      "balance_loss_clip": 1.04121447,
      "balance_loss_mlp": 1.01728892,
      "epoch": 0.9061504238561895,
      "flos": 31759432963200.0,
      "grad_norm": 2.430531106681681,
      "language_loss": 0.68098676,
      "learning_rate": 9.158613222874346e-08,
      "loss": 0.70294404,
      "num_input_tokens_seen": 162949445,
      "step": 7536,
      "time_per_iteration": 3.610461711883545
    },
    {
      "auxiliary_loss_clip": 0.01170858,
      "auxiliary_loss_mlp": 0.01019792,
      "balance_loss_clip": 1.04386878,
      "balance_loss_mlp": 1.01251411,
      "epoch": 0.9062706667468285,
      "flos": 20048066075520.0,
      "grad_norm": 1.6968712287357,
      "language_loss": 0.81687558,
      "learning_rate": 9.135324969566394e-08,
      "loss": 0.83878207,
      "num_input_tokens_seen": 162968945,
      "step": 7537,
      "time_per_iteration": 2.7160940170288086
    },
    {
      "auxiliary_loss_clip": 0.0117364,
      "auxiliary_loss_mlp": 0.01022176,
      "balance_loss_clip": 1.04701948,
      "balance_loss_mlp": 1.01477635,
      "epoch": 0.9063909096374677,
      "flos": 18437292576000.0,
      "grad_norm": 3.169286490471717,
      "language_loss": 0.75815177,
      "learning_rate": 9.112065670277913e-08,
      "loss": 0.78010994,
      "num_input_tokens_seen": 162985310,
      "step": 7538,
      "time_per_iteration": 2.5929789543151855
    },
    {
      "auxiliary_loss_clip": 0.01167775,
      "auxiliary_loss_mlp": 0.01022477,
      "balance_loss_clip": 1.04244149,
      "balance_loss_mlp": 1.01546192,
      "epoch": 0.9065111525281068,
      "flos": 33547353361920.0,
      "grad_norm": 2.0312354761914793,
      "language_loss": 0.72842836,
      "learning_rate": 9.088835328537303e-08,
      "loss": 0.75033087,
      "num_input_tokens_seen": 163006900,
      "step": 7539,
      "time_per_iteration": 3.6572606563568115
    },
    {
      "auxiliary_loss_clip": 0.01176526,
      "auxiliary_loss_mlp": 0.01022812,
      "balance_loss_clip": 1.0462153,
      "balance_loss_mlp": 1.01527238,
      "epoch": 0.9066313954187458,
      "flos": 23367863750400.0,
      "grad_norm": 2.324730675794416,
      "language_loss": 0.7232517,
      "learning_rate": 9.065633947868568e-08,
      "loss": 0.74524504,
      "num_input_tokens_seen": 163026505,
      "step": 7540,
      "time_per_iteration": 2.6340034008026123
    },
    {
      "auxiliary_loss_clip": 0.01175605,
      "auxiliary_loss_mlp": 0.01207574,
      "balance_loss_clip": 1.04553938,
      "balance_loss_mlp": 1.00065112,
      "epoch": 0.906751638309385,
      "flos": 26249623067520.0,
      "grad_norm": 2.3815917079503817,
      "language_loss": 0.80078995,
      "learning_rate": 9.042461531791379e-08,
      "loss": 0.8246218,
      "num_input_tokens_seen": 163044925,
      "step": 7541,
      "time_per_iteration": 2.71146559715271
    },
    {
      "auxiliary_loss_clip": 0.01163277,
      "auxiliary_loss_mlp": 0.01024,
      "balance_loss_clip": 1.04632521,
      "balance_loss_mlp": 1.01700842,
      "epoch": 0.906871881200024,
      "flos": 16544477485440.0,
      "grad_norm": 1.8216563895696756,
      "language_loss": 0.78028858,
      "learning_rate": 9.019318083820903e-08,
      "loss": 0.80216134,
      "num_input_tokens_seen": 163063505,
      "step": 7542,
      "time_per_iteration": 2.543381690979004
    },
    {
      "auxiliary_loss_clip": 0.01170293,
      "auxiliary_loss_mlp": 0.01026273,
      "balance_loss_clip": 1.04842377,
      "balance_loss_mlp": 1.01875687,
      "epoch": 0.9069921240906631,
      "flos": 24605129675520.0,
      "grad_norm": 1.8452483430456657,
      "language_loss": 0.85286683,
      "learning_rate": 8.996203607468045e-08,
      "loss": 0.87483245,
      "num_input_tokens_seen": 163082505,
      "step": 7543,
      "time_per_iteration": 2.664212226867676
    },
    {
      "auxiliary_loss_clip": 0.0116482,
      "auxiliary_loss_mlp": 0.01025759,
      "balance_loss_clip": 1.0431478,
      "balance_loss_mlp": 1.01835656,
      "epoch": 0.9071123669813023,
      "flos": 25374731500800.0,
      "grad_norm": 1.5439516965504454,
      "language_loss": 0.75575292,
      "learning_rate": 8.973118106239241e-08,
      "loss": 0.7776587,
      "num_input_tokens_seen": 163105110,
      "step": 7544,
      "time_per_iteration": 2.595405340194702
    },
    {
      "auxiliary_loss_clip": 0.01177491,
      "auxiliary_loss_mlp": 0.01026908,
      "balance_loss_clip": 1.03781843,
      "balance_loss_mlp": 1.01926696,
      "epoch": 0.9072326098719413,
      "flos": 26725798690560.0,
      "grad_norm": 4.351248504918502,
      "language_loss": 0.94659835,
      "learning_rate": 8.95006158363656e-08,
      "loss": 0.96864235,
      "num_input_tokens_seen": 163125295,
      "step": 7545,
      "time_per_iteration": 2.8508501052856445
    },
    {
      "auxiliary_loss_clip": 0.01167929,
      "auxiliary_loss_mlp": 0.01027668,
      "balance_loss_clip": 1.04688656,
      "balance_loss_mlp": 1.01955581,
      "epoch": 0.9073528527625804,
      "flos": 23878800760320.0,
      "grad_norm": 2.6417399429597137,
      "language_loss": 0.77521378,
      "learning_rate": 8.9270340431576e-08,
      "loss": 0.79716974,
      "num_input_tokens_seen": 163144385,
      "step": 7546,
      "time_per_iteration": 2.611631155014038
    },
    {
      "auxiliary_loss_clip": 0.01171099,
      "auxiliary_loss_mlp": 0.01025735,
      "balance_loss_clip": 1.04508758,
      "balance_loss_mlp": 1.01853156,
      "epoch": 0.9074730956532195,
      "flos": 37852144767360.0,
      "grad_norm": 2.0717240313744596,
      "language_loss": 0.73647422,
      "learning_rate": 8.904035488295658e-08,
      "loss": 0.75844258,
      "num_input_tokens_seen": 163163885,
      "step": 7547,
      "time_per_iteration": 2.7661118507385254
    },
    {
      "auxiliary_loss_clip": 0.01070193,
      "auxiliary_loss_mlp": 0.01198668,
      "balance_loss_clip": 1.01083934,
      "balance_loss_mlp": 0.99988872,
      "epoch": 0.9075933385438586,
      "flos": 65173307385600.0,
      "grad_norm": 0.6594226621411328,
      "language_loss": 0.53228688,
      "learning_rate": 8.881065922539632e-08,
      "loss": 0.55497551,
      "num_input_tokens_seen": 163224325,
      "step": 7548,
      "time_per_iteration": 3.057614803314209
    },
    {
      "auxiliary_loss_clip": 0.01166724,
      "auxiliary_loss_mlp": 0.01020334,
      "balance_loss_clip": 1.04190278,
      "balance_loss_mlp": 1.01344347,
      "epoch": 0.9077135814344977,
      "flos": 19931571290880.0,
      "grad_norm": 1.6504461821476177,
      "language_loss": 0.73274899,
      "learning_rate": 8.85812534937389e-08,
      "loss": 0.75461948,
      "num_input_tokens_seen": 163242425,
      "step": 7549,
      "time_per_iteration": 2.6218302249908447
    },
    {
      "auxiliary_loss_clip": 0.01176889,
      "auxiliary_loss_mlp": 0.0102557,
      "balance_loss_clip": 1.04779339,
      "balance_loss_mlp": 1.01789308,
      "epoch": 0.9078338243251368,
      "flos": 17529650784000.0,
      "grad_norm": 2.6548373780932906,
      "language_loss": 0.67647624,
      "learning_rate": 8.835213772278583e-08,
      "loss": 0.69850087,
      "num_input_tokens_seen": 163259280,
      "step": 7550,
      "time_per_iteration": 2.5812828540802
    },
    {
      "auxiliary_loss_clip": 0.01163558,
      "auxiliary_loss_mlp": 0.01020759,
      "balance_loss_clip": 1.04186618,
      "balance_loss_mlp": 1.01351142,
      "epoch": 0.9079540672157759,
      "flos": 28803410277120.0,
      "grad_norm": 2.0666101980759466,
      "language_loss": 0.79236501,
      "learning_rate": 8.812331194729373e-08,
      "loss": 0.81420821,
      "num_input_tokens_seen": 163278925,
      "step": 7551,
      "time_per_iteration": 3.694525957107544
    },
    {
      "auxiliary_loss_clip": 0.01174973,
      "auxiliary_loss_mlp": 0.01027053,
      "balance_loss_clip": 1.05368137,
      "balance_loss_mlp": 1.019436,
      "epoch": 0.9080743101064149,
      "flos": 23513840622720.0,
      "grad_norm": 1.8789103549350963,
      "language_loss": 0.72200519,
      "learning_rate": 8.789477620197461e-08,
      "loss": 0.74402541,
      "num_input_tokens_seen": 163298450,
      "step": 7552,
      "time_per_iteration": 2.5979816913604736
    },
    {
      "auxiliary_loss_clip": 0.01171909,
      "auxiliary_loss_mlp": 0.01024767,
      "balance_loss_clip": 1.04428363,
      "balance_loss_mlp": 1.01726532,
      "epoch": 0.9081945529970541,
      "flos": 22778102344320.0,
      "grad_norm": 2.424824352628486,
      "language_loss": 0.79363763,
      "learning_rate": 8.766653052149831e-08,
      "loss": 0.81560445,
      "num_input_tokens_seen": 163313635,
      "step": 7553,
      "time_per_iteration": 2.610443592071533
    },
    {
      "auxiliary_loss_clip": 0.01168571,
      "auxiliary_loss_mlp": 0.01026417,
      "balance_loss_clip": 1.04343426,
      "balance_loss_mlp": 1.01823926,
      "epoch": 0.9083147958876931,
      "flos": 18873714821760.0,
      "grad_norm": 2.2741428813270375,
      "language_loss": 0.7444042,
      "learning_rate": 8.743857494048823e-08,
      "loss": 0.76635408,
      "num_input_tokens_seen": 163330450,
      "step": 7554,
      "time_per_iteration": 2.579148054122925
    },
    {
      "auxiliary_loss_clip": 0.0117141,
      "auxiliary_loss_mlp": 0.01025133,
      "balance_loss_clip": 1.04218018,
      "balance_loss_mlp": 1.01730692,
      "epoch": 0.9084350387783322,
      "flos": 18909374048640.0,
      "grad_norm": 2.076302699353639,
      "language_loss": 0.63339716,
      "learning_rate": 8.721090949352605e-08,
      "loss": 0.65536261,
      "num_input_tokens_seen": 163346690,
      "step": 7555,
      "time_per_iteration": 2.6873362064361572
    },
    {
      "auxiliary_loss_clip": 0.01178179,
      "auxiliary_loss_mlp": 0.01024207,
      "balance_loss_clip": 1.04825425,
      "balance_loss_mlp": 1.0163331,
      "epoch": 0.9085552816689714,
      "flos": 20595488325120.0,
      "grad_norm": 1.9522229852811042,
      "language_loss": 0.7324748,
      "learning_rate": 8.698353421514793e-08,
      "loss": 0.7544986,
      "num_input_tokens_seen": 163365065,
      "step": 7556,
      "time_per_iteration": 2.563694953918457
    },
    {
      "auxiliary_loss_clip": 0.01168862,
      "auxiliary_loss_mlp": 0.01024082,
      "balance_loss_clip": 1.04647136,
      "balance_loss_mlp": 1.01714444,
      "epoch": 0.9086755245596104,
      "flos": 18113163223680.0,
      "grad_norm": 2.1580288846506384,
      "language_loss": 0.80127978,
      "learning_rate": 8.67564491398467e-08,
      "loss": 0.82320917,
      "num_input_tokens_seen": 163382070,
      "step": 7557,
      "time_per_iteration": 2.665195941925049
    },
    {
      "auxiliary_loss_clip": 0.01171049,
      "auxiliary_loss_mlp": 0.01026558,
      "balance_loss_clip": 1.04510808,
      "balance_loss_mlp": 1.01859772,
      "epoch": 0.9087957674502495,
      "flos": 19129793857920.0,
      "grad_norm": 1.8154302150998167,
      "language_loss": 0.73789579,
      "learning_rate": 8.652965430207104e-08,
      "loss": 0.75987184,
      "num_input_tokens_seen": 163399975,
      "step": 7558,
      "time_per_iteration": 2.5848469734191895
    },
    {
      "auxiliary_loss_clip": 0.0117207,
      "auxiliary_loss_mlp": 0.01027007,
      "balance_loss_clip": 1.04597366,
      "balance_loss_mlp": 1.01927924,
      "epoch": 0.9089160103408886,
      "flos": 18109930999680.0,
      "grad_norm": 2.6123709942845017,
      "language_loss": 0.65625942,
      "learning_rate": 8.630314973622521e-08,
      "loss": 0.67825019,
      "num_input_tokens_seen": 163417520,
      "step": 7559,
      "time_per_iteration": 2.5797250270843506
    },
    {
      "auxiliary_loss_clip": 0.01167476,
      "auxiliary_loss_mlp": 0.0102442,
      "balance_loss_clip": 1.04767632,
      "balance_loss_mlp": 1.01732707,
      "epoch": 0.9090362532315277,
      "flos": 33364855336320.0,
      "grad_norm": 2.103662201488486,
      "language_loss": 0.70966506,
      "learning_rate": 8.607693547666995e-08,
      "loss": 0.73158407,
      "num_input_tokens_seen": 163440060,
      "step": 7560,
      "time_per_iteration": 2.6565842628479004
    },
    {
      "auxiliary_loss_clip": 0.01083259,
      "auxiliary_loss_mlp": 0.0100182,
      "balance_loss_clip": 1.00950193,
      "balance_loss_mlp": 1.00109529,
      "epoch": 0.9091564961221668,
      "flos": 71480585082240.0,
      "grad_norm": 0.895498878481397,
      "language_loss": 0.57906973,
      "learning_rate": 8.585101155772201e-08,
      "loss": 0.59992051,
      "num_input_tokens_seen": 163502180,
      "step": 7561,
      "time_per_iteration": 3.302762985229492
    },
    {
      "auxiliary_loss_clip": 0.01164052,
      "auxiliary_loss_mlp": 0.01026236,
      "balance_loss_clip": 1.04024208,
      "balance_loss_mlp": 1.01861835,
      "epoch": 0.9092767390128058,
      "flos": 24712574232960.0,
      "grad_norm": 1.8258295081730156,
      "language_loss": 0.68740249,
      "learning_rate": 8.562537801365377e-08,
      "loss": 0.70930541,
      "num_input_tokens_seen": 163521915,
      "step": 7562,
      "time_per_iteration": 4.523227214813232
    },
    {
      "auxiliary_loss_clip": 0.01167785,
      "auxiliary_loss_mlp": 0.0102477,
      "balance_loss_clip": 1.04752564,
      "balance_loss_mlp": 1.01738501,
      "epoch": 0.909396981903445,
      "flos": 23586487879680.0,
      "grad_norm": 1.7621769040680322,
      "language_loss": 0.70200408,
      "learning_rate": 8.540003487869362e-08,
      "loss": 0.72392964,
      "num_input_tokens_seen": 163543585,
      "step": 7563,
      "time_per_iteration": 2.591097116470337
    },
    {
      "auxiliary_loss_clip": 0.01160834,
      "auxiliary_loss_mlp": 0.01024828,
      "balance_loss_clip": 1.0386765,
      "balance_loss_mlp": 1.01732349,
      "epoch": 0.909517224794084,
      "flos": 23404169422080.0,
      "grad_norm": 1.8410492068321473,
      "language_loss": 0.79606175,
      "learning_rate": 8.517498218702557e-08,
      "loss": 0.8179183,
      "num_input_tokens_seen": 163561515,
      "step": 7564,
      "time_per_iteration": 2.716359853744507
    },
    {
      "auxiliary_loss_clip": 0.01170414,
      "auxiliary_loss_mlp": 0.01020857,
      "balance_loss_clip": 1.04321766,
      "balance_loss_mlp": 1.01382041,
      "epoch": 0.9096374676847231,
      "flos": 19208618254080.0,
      "grad_norm": 1.9216425434898448,
      "language_loss": 0.69448614,
      "learning_rate": 8.49502199727905e-08,
      "loss": 0.71639884,
      "num_input_tokens_seen": 163579540,
      "step": 7565,
      "time_per_iteration": 2.6589553356170654
    },
    {
      "auxiliary_loss_clip": 0.01164777,
      "auxiliary_loss_mlp": 0.0102472,
      "balance_loss_clip": 1.0426414,
      "balance_loss_mlp": 1.01707268,
      "epoch": 0.9097577105753623,
      "flos": 33292495388160.0,
      "grad_norm": 2.6642191475795305,
      "language_loss": 0.66020286,
      "learning_rate": 8.472574827008428e-08,
      "loss": 0.68209785,
      "num_input_tokens_seen": 163600425,
      "step": 7566,
      "time_per_iteration": 3.5973129272460938
    },
    {
      "auxiliary_loss_clip": 0.01166071,
      "auxiliary_loss_mlp": 0.01023863,
      "balance_loss_clip": 1.04294825,
      "balance_loss_mlp": 1.01635909,
      "epoch": 0.9098779534660013,
      "flos": 21906443001600.0,
      "grad_norm": 2.0692376039218523,
      "language_loss": 0.8369655,
      "learning_rate": 8.450156711295942e-08,
      "loss": 0.8588649,
      "num_input_tokens_seen": 163620595,
      "step": 7567,
      "time_per_iteration": 2.72125244140625
    },
    {
      "auxiliary_loss_clip": 0.01173156,
      "auxiliary_loss_mlp": 0.01026429,
      "balance_loss_clip": 1.04773211,
      "balance_loss_mlp": 1.01916301,
      "epoch": 0.9099981963566404,
      "flos": 25730354102400.0,
      "grad_norm": 2.3494985136427116,
      "language_loss": 0.86997426,
      "learning_rate": 8.427767653542383e-08,
      "loss": 0.89197016,
      "num_input_tokens_seen": 163635765,
      "step": 7568,
      "time_per_iteration": 2.6799888610839844
    },
    {
      "auxiliary_loss_clip": 0.0117113,
      "auxiliary_loss_mlp": 0.01024863,
      "balance_loss_clip": 1.0394901,
      "balance_loss_mlp": 1.01822937,
      "epoch": 0.9101184392472795,
      "flos": 21069437304960.0,
      "grad_norm": 1.9601230797296978,
      "language_loss": 0.70307434,
      "learning_rate": 8.405407657144125e-08,
      "loss": 0.72503424,
      "num_input_tokens_seen": 163654925,
      "step": 7569,
      "time_per_iteration": 2.678158760070801
    },
    {
      "auxiliary_loss_clip": 0.01165683,
      "auxiliary_loss_mlp": 0.01023462,
      "balance_loss_clip": 1.0429672,
      "balance_loss_mlp": 1.01626146,
      "epoch": 0.9102386821379186,
      "flos": 24752614919040.0,
      "grad_norm": 2.4517432358789635,
      "language_loss": 0.72550559,
      "learning_rate": 8.383076725493232e-08,
      "loss": 0.74739707,
      "num_input_tokens_seen": 163672245,
      "step": 7570,
      "time_per_iteration": 2.657353401184082
    },
    {
      "auxiliary_loss_clip": 0.01169533,
      "auxiliary_loss_mlp": 0.01020912,
      "balance_loss_clip": 1.04571176,
      "balance_loss_mlp": 1.01398039,
      "epoch": 0.9103589250285576,
      "flos": 22562818179840.0,
      "grad_norm": 2.365387156146332,
      "language_loss": 0.68164837,
      "learning_rate": 8.360774861977216e-08,
      "loss": 0.70355284,
      "num_input_tokens_seen": 163691365,
      "step": 7571,
      "time_per_iteration": 2.6436970233917236
    },
    {
      "auxiliary_loss_clip": 0.01169804,
      "auxiliary_loss_mlp": 0.01022072,
      "balance_loss_clip": 1.04042113,
      "balance_loss_mlp": 1.01481509,
      "epoch": 0.9104791679191968,
      "flos": 25373474524800.0,
      "grad_norm": 1.931761613061375,
      "language_loss": 0.74807227,
      "learning_rate": 8.338502069979281e-08,
      "loss": 0.76999104,
      "num_input_tokens_seen": 163711675,
      "step": 7572,
      "time_per_iteration": 2.6566622257232666
    },
    {
      "auxiliary_loss_clip": 0.01171125,
      "auxiliary_loss_mlp": 0.01024719,
      "balance_loss_clip": 1.04486156,
      "balance_loss_mlp": 1.01704764,
      "epoch": 0.9105994108098359,
      "flos": 14426681558400.0,
      "grad_norm": 2.7552062586160058,
      "language_loss": 0.79886568,
      "learning_rate": 8.316258352878214e-08,
      "loss": 0.82082415,
      "num_input_tokens_seen": 163728095,
      "step": 7573,
      "time_per_iteration": 2.6738874912261963
    },
    {
      "auxiliary_loss_clip": 0.01174321,
      "auxiliary_loss_mlp": 0.01025352,
      "balance_loss_clip": 1.04533231,
      "balance_loss_mlp": 1.01781452,
      "epoch": 0.9107196537004749,
      "flos": 26718292748160.0,
      "grad_norm": 2.028939889994267,
      "language_loss": 0.71282256,
      "learning_rate": 8.294043714048338e-08,
      "loss": 0.73481929,
      "num_input_tokens_seen": 163747175,
      "step": 7574,
      "time_per_iteration": 2.6552734375
    },
    {
      "auxiliary_loss_clip": 0.01077989,
      "auxiliary_loss_mlp": 0.01000016,
      "balance_loss_clip": 1.01097012,
      "balance_loss_mlp": 0.99923515,
      "epoch": 0.9108398965911141,
      "flos": 66532634703360.0,
      "grad_norm": 0.7491289314721288,
      "language_loss": 0.60443294,
      "learning_rate": 8.271858156859624e-08,
      "loss": 0.62521303,
      "num_input_tokens_seen": 163812545,
      "step": 7575,
      "time_per_iteration": 3.25056791305542
    },
    {
      "auxiliary_loss_clip": 0.01164372,
      "auxiliary_loss_mlp": 0.01018703,
      "balance_loss_clip": 1.04748559,
      "balance_loss_mlp": 1.01141357,
      "epoch": 0.9109601394817531,
      "flos": 25411073086080.0,
      "grad_norm": 1.6755268337330562,
      "language_loss": 0.74061537,
      "learning_rate": 8.249701684677557e-08,
      "loss": 0.76244611,
      "num_input_tokens_seen": 163833870,
      "step": 7576,
      "time_per_iteration": 2.6216940879821777
    },
    {
      "auxiliary_loss_clip": 0.01170536,
      "auxiliary_loss_mlp": 0.01022125,
      "balance_loss_clip": 1.04857743,
      "balance_loss_mlp": 1.01479912,
      "epoch": 0.9110803823723922,
      "flos": 22747794243840.0,
      "grad_norm": 29.36317637358638,
      "language_loss": 0.81307304,
      "learning_rate": 8.227574300863294e-08,
      "loss": 0.83499962,
      "num_input_tokens_seen": 163854040,
      "step": 7577,
      "time_per_iteration": 2.6425552368164062
    },
    {
      "auxiliary_loss_clip": 0.01178329,
      "auxiliary_loss_mlp": 0.01025106,
      "balance_loss_clip": 1.04807138,
      "balance_loss_mlp": 1.01739633,
      "epoch": 0.9112006252630314,
      "flos": 48469924131840.0,
      "grad_norm": 1.744840687490264,
      "language_loss": 0.69843251,
      "learning_rate": 8.205476008773548e-08,
      "loss": 0.72046679,
      "num_input_tokens_seen": 163878040,
      "step": 7578,
      "time_per_iteration": 3.8788654804229736
    },
    {
      "auxiliary_loss_clip": 0.01164057,
      "auxiliary_loss_mlp": 0.01023235,
      "balance_loss_clip": 1.04194283,
      "balance_loss_mlp": 1.01585913,
      "epoch": 0.9113208681536704,
      "flos": 30009649829760.0,
      "grad_norm": 2.2290570630172066,
      "language_loss": 0.82300186,
      "learning_rate": 8.183406811760596e-08,
      "loss": 0.84487474,
      "num_input_tokens_seen": 163897770,
      "step": 7579,
      "time_per_iteration": 2.725571632385254
    },
    {
      "auxiliary_loss_clip": 0.01159645,
      "auxiliary_loss_mlp": 0.01020826,
      "balance_loss_clip": 1.03942144,
      "balance_loss_mlp": 1.01364982,
      "epoch": 0.9114411110443095,
      "flos": 25594971742080.0,
      "grad_norm": 1.606596696076947,
      "language_loss": 0.74219954,
      "learning_rate": 8.161366713172313e-08,
      "loss": 0.76400423,
      "num_input_tokens_seen": 163920160,
      "step": 7580,
      "time_per_iteration": 2.7301855087280273
    },
    {
      "auxiliary_loss_clip": 0.01181352,
      "auxiliary_loss_mlp": 0.01026564,
      "balance_loss_clip": 1.04416347,
      "balance_loss_mlp": 1.01870847,
      "epoch": 0.9115613539349486,
      "flos": 18399729928320.0,
      "grad_norm": 2.7608114011693154,
      "language_loss": 0.8478204,
      "learning_rate": 8.139355716352137e-08,
      "loss": 0.86989951,
      "num_input_tokens_seen": 163935000,
      "step": 7581,
      "time_per_iteration": 2.660775899887085
    },
    {
      "auxiliary_loss_clip": 0.01174756,
      "auxiliary_loss_mlp": 0.01028786,
      "balance_loss_clip": 1.04376256,
      "balance_loss_mlp": 1.02085876,
      "epoch": 0.9116815968255877,
      "flos": 21726171619200.0,
      "grad_norm": 1.5853435369478597,
      "language_loss": 0.70062631,
      "learning_rate": 8.117373824639196e-08,
      "loss": 0.72266173,
      "num_input_tokens_seen": 163955265,
      "step": 7582,
      "time_per_iteration": 2.687943458557129
    },
    {
      "auxiliary_loss_clip": 0.01062484,
      "auxiliary_loss_mlp": 0.01001908,
      "balance_loss_clip": 1.01003504,
      "balance_loss_mlp": 1.00118399,
      "epoch": 0.9118018397162267,
      "flos": 65363526835200.0,
      "grad_norm": 0.7199559344431642,
      "language_loss": 0.59202737,
      "learning_rate": 8.095421041368067e-08,
      "loss": 0.61267126,
      "num_input_tokens_seen": 164014680,
      "step": 7583,
      "time_per_iteration": 3.0178003311157227
    },
    {
      "auxiliary_loss_clip": 0.01172039,
      "auxiliary_loss_mlp": 0.01207764,
      "balance_loss_clip": 1.04702592,
      "balance_loss_mlp": 1.00085568,
      "epoch": 0.9119220826068659,
      "flos": 20922885815040.0,
      "grad_norm": 2.064202096150199,
      "language_loss": 0.70504475,
      "learning_rate": 8.073497369868999e-08,
      "loss": 0.72884274,
      "num_input_tokens_seen": 164033140,
      "step": 7584,
      "time_per_iteration": 2.6433165073394775
    },
    {
      "auxiliary_loss_clip": 0.01181159,
      "auxiliary_loss_mlp": 0.01024733,
      "balance_loss_clip": 1.04623652,
      "balance_loss_mlp": 1.0172075,
      "epoch": 0.912042325497505,
      "flos": 28366449327360.0,
      "grad_norm": 1.8756481822265025,
      "language_loss": 0.7561335,
      "learning_rate": 8.051602813467772e-08,
      "loss": 0.77819246,
      "num_input_tokens_seen": 164054995,
      "step": 7585,
      "time_per_iteration": 2.734229326248169
    },
    {
      "auxiliary_loss_clip": 0.01171669,
      "auxiliary_loss_mlp": 0.01022095,
      "balance_loss_clip": 1.04602504,
      "balance_loss_mlp": 1.01519251,
      "epoch": 0.912162568388144,
      "flos": 17566782468480.0,
      "grad_norm": 3.5761150911676114,
      "language_loss": 0.71259302,
      "learning_rate": 8.029737375485756e-08,
      "loss": 0.73453063,
      "num_input_tokens_seen": 164074225,
      "step": 7586,
      "time_per_iteration": 2.5769081115722656
    },
    {
      "auxiliary_loss_clip": 0.01166486,
      "auxiliary_loss_mlp": 0.01023617,
      "balance_loss_clip": 1.04769027,
      "balance_loss_mlp": 1.0165447,
      "epoch": 0.9122828112787832,
      "flos": 19827897661440.0,
      "grad_norm": 1.7364913153510368,
      "language_loss": 0.72392046,
      "learning_rate": 8.007901059239986e-08,
      "loss": 0.74582154,
      "num_input_tokens_seen": 164093505,
      "step": 7587,
      "time_per_iteration": 2.737466812133789
    },
    {
      "auxiliary_loss_clip": 0.01171138,
      "auxiliary_loss_mlp": 0.01020194,
      "balance_loss_clip": 1.04144514,
      "balance_loss_mlp": 1.0129106,
      "epoch": 0.9124030541694222,
      "flos": 20813789232000.0,
      "grad_norm": 1.8749377145203372,
      "language_loss": 0.80346465,
      "learning_rate": 7.986093868042964e-08,
      "loss": 0.825378,
      "num_input_tokens_seen": 164113750,
      "step": 7588,
      "time_per_iteration": 3.5685698986053467
    },
    {
      "auxiliary_loss_clip": 0.01167119,
      "auxiliary_loss_mlp": 0.01027061,
      "balance_loss_clip": 1.04589081,
      "balance_loss_mlp": 1.01988482,
      "epoch": 0.9125232970600613,
      "flos": 25192305302400.0,
      "grad_norm": 1.6613234106508685,
      "language_loss": 0.68059051,
      "learning_rate": 7.964315805202826e-08,
      "loss": 0.70253229,
      "num_input_tokens_seen": 164134330,
      "step": 7589,
      "time_per_iteration": 3.676095485687256
    },
    {
      "auxiliary_loss_clip": 0.01169384,
      "auxiliary_loss_mlp": 0.01021885,
      "balance_loss_clip": 1.04401958,
      "balance_loss_mlp": 1.01403821,
      "epoch": 0.9126435399507005,
      "flos": 19719591177600.0,
      "grad_norm": 1.6577379096704472,
      "language_loss": 0.7342571,
      "learning_rate": 7.942566874023304e-08,
      "loss": 0.7561698,
      "num_input_tokens_seen": 164153515,
      "step": 7590,
      "time_per_iteration": 2.6278817653656006
    },
    {
      "auxiliary_loss_clip": 0.01167866,
      "auxiliary_loss_mlp": 0.01024726,
      "balance_loss_clip": 1.04148734,
      "balance_loss_mlp": 1.01712644,
      "epoch": 0.9127637828413395,
      "flos": 19573614305280.0,
      "grad_norm": 3.3513711499004204,
      "language_loss": 0.70029211,
      "learning_rate": 7.920847077803649e-08,
      "loss": 0.7222181,
      "num_input_tokens_seen": 164171305,
      "step": 7591,
      "time_per_iteration": 2.648372173309326
    },
    {
      "auxiliary_loss_clip": 0.01164822,
      "auxiliary_loss_mlp": 0.01028206,
      "balance_loss_clip": 1.03644085,
      "balance_loss_mlp": 1.02074337,
      "epoch": 0.9128840257319786,
      "flos": 20230635928320.0,
      "grad_norm": 2.3517561903538566,
      "language_loss": 0.82709444,
      "learning_rate": 7.899156419838826e-08,
      "loss": 0.84902477,
      "num_input_tokens_seen": 164190275,
      "step": 7592,
      "time_per_iteration": 2.684344530105591
    },
    {
      "auxiliary_loss_clip": 0.01170413,
      "auxiliary_loss_mlp": 0.01021117,
      "balance_loss_clip": 1.04172707,
      "balance_loss_mlp": 1.01421213,
      "epoch": 0.9130042686226177,
      "flos": 24858658846080.0,
      "grad_norm": 1.9495236185807097,
      "language_loss": 0.65587103,
      "learning_rate": 7.87749490341918e-08,
      "loss": 0.67778635,
      "num_input_tokens_seen": 164210550,
      "step": 7593,
      "time_per_iteration": 3.596607208251953
    },
    {
      "auxiliary_loss_clip": 0.01170724,
      "auxiliary_loss_mlp": 0.01024233,
      "balance_loss_clip": 1.04934454,
      "balance_loss_mlp": 1.01662731,
      "epoch": 0.9131245115132568,
      "flos": 23581747284480.0,
      "grad_norm": 2.2563957271029342,
      "language_loss": 0.83251202,
      "learning_rate": 7.855862531830836e-08,
      "loss": 0.85446167,
      "num_input_tokens_seen": 164226660,
      "step": 7594,
      "time_per_iteration": 2.5420446395874023
    },
    {
      "auxiliary_loss_clip": 0.0116615,
      "auxiliary_loss_mlp": 0.01025529,
      "balance_loss_clip": 1.04364765,
      "balance_loss_mlp": 1.01833773,
      "epoch": 0.9132447544038959,
      "flos": 19931607204480.0,
      "grad_norm": 1.5486436539966189,
      "language_loss": 0.72859371,
      "learning_rate": 7.834259308355373e-08,
      "loss": 0.75051045,
      "num_input_tokens_seen": 164245425,
      "step": 7595,
      "time_per_iteration": 2.6303927898406982
    },
    {
      "auxiliary_loss_clip": 0.01162969,
      "auxiliary_loss_mlp": 0.0102577,
      "balance_loss_clip": 1.03711247,
      "balance_loss_mlp": 1.01842976,
      "epoch": 0.9133649972945349,
      "flos": 21981747864960.0,
      "grad_norm": 2.098899376521632,
      "language_loss": 0.75047565,
      "learning_rate": 7.812685236269989e-08,
      "loss": 0.77236301,
      "num_input_tokens_seen": 164264085,
      "step": 7596,
      "time_per_iteration": 2.709765911102295
    },
    {
      "auxiliary_loss_clip": 0.01078728,
      "auxiliary_loss_mlp": 0.0100126,
      "balance_loss_clip": 1.01345754,
      "balance_loss_mlp": 1.00058389,
      "epoch": 0.9134852401851741,
      "flos": 71240523511680.0,
      "grad_norm": 1.6781814507233235,
      "language_loss": 0.58601296,
      "learning_rate": 7.791140318847445e-08,
      "loss": 0.60681289,
      "num_input_tokens_seen": 164322220,
      "step": 7597,
      "time_per_iteration": 3.191145896911621
    },
    {
      "auxiliary_loss_clip": 0.01166432,
      "auxiliary_loss_mlp": 0.01018094,
      "balance_loss_clip": 1.04566789,
      "balance_loss_mlp": 1.01118612,
      "epoch": 0.9136054830758131,
      "flos": 23626923615360.0,
      "grad_norm": 1.440079595930018,
      "language_loss": 0.80522931,
      "learning_rate": 7.769624559356081e-08,
      "loss": 0.82707453,
      "num_input_tokens_seen": 164345615,
      "step": 7598,
      "time_per_iteration": 2.676682710647583
    },
    {
      "auxiliary_loss_clip": 0.01167603,
      "auxiliary_loss_mlp": 0.0102644,
      "balance_loss_clip": 1.04392815,
      "balance_loss_mlp": 1.01786876,
      "epoch": 0.9137257259664522,
      "flos": 23438858981760.0,
      "grad_norm": 2.358868051043412,
      "language_loss": 0.75527918,
      "learning_rate": 7.748137961059842e-08,
      "loss": 0.77721959,
      "num_input_tokens_seen": 164359595,
      "step": 7599,
      "time_per_iteration": 2.634692668914795
    },
    {
      "auxiliary_loss_clip": 0.01163794,
      "auxiliary_loss_mlp": 0.01022608,
      "balance_loss_clip": 1.04766774,
      "balance_loss_mlp": 1.01516294,
      "epoch": 0.9138459688570914,
      "flos": 19127854523520.0,
      "grad_norm": 3.1973962434599716,
      "language_loss": 0.65767962,
      "learning_rate": 7.726680527218211e-08,
      "loss": 0.67954361,
      "num_input_tokens_seen": 164376635,
      "step": 7600,
      "time_per_iteration": 2.5478780269622803
    },
    {
      "auxiliary_loss_clip": 0.01165084,
      "auxiliary_loss_mlp": 0.01020928,
      "balance_loss_clip": 1.04461372,
      "balance_loss_mlp": 1.01350737,
      "epoch": 0.9139662117477304,
      "flos": 46281240714240.0,
      "grad_norm": 1.8066686258415725,
      "language_loss": 0.75840771,
      "learning_rate": 7.70525226108627e-08,
      "loss": 0.78026783,
      "num_input_tokens_seen": 164400305,
      "step": 7601,
      "time_per_iteration": 2.8439176082611084
    },
    {
      "auxiliary_loss_clip": 0.01171755,
      "auxiliary_loss_mlp": 0.01026369,
      "balance_loss_clip": 1.04927051,
      "balance_loss_mlp": 1.01904917,
      "epoch": 0.9140864546383695,
      "flos": 22273198819200.0,
      "grad_norm": 2.6430602704319917,
      "language_loss": 0.79918915,
      "learning_rate": 7.683853165914666e-08,
      "loss": 0.82117033,
      "num_input_tokens_seen": 164418075,
      "step": 7602,
      "time_per_iteration": 2.6340911388397217
    },
    {
      "auxiliary_loss_clip": 0.01176114,
      "auxiliary_loss_mlp": 0.01027197,
      "balance_loss_clip": 1.04227138,
      "balance_loss_mlp": 1.02018738,
      "epoch": 0.9142066975290086,
      "flos": 17530009920000.0,
      "grad_norm": 13.20137630130776,
      "language_loss": 0.77084661,
      "learning_rate": 7.662483244949602e-08,
      "loss": 0.79287976,
      "num_input_tokens_seen": 164435335,
      "step": 7603,
      "time_per_iteration": 2.714785575866699
    },
    {
      "auxiliary_loss_clip": 0.01165822,
      "auxiliary_loss_mlp": 0.0102002,
      "balance_loss_clip": 1.04207742,
      "balance_loss_mlp": 1.01286101,
      "epoch": 0.9143269404196477,
      "flos": 17712148809600.0,
      "grad_norm": 6.332940262886805,
      "language_loss": 0.80769682,
      "learning_rate": 7.641142501432951e-08,
      "loss": 0.82955527,
      "num_input_tokens_seen": 164451530,
      "step": 7604,
      "time_per_iteration": 3.6247425079345703
    },
    {
      "auxiliary_loss_clip": 0.0116612,
      "auxiliary_loss_mlp": 0.01022786,
      "balance_loss_clip": 1.04312181,
      "balance_loss_mlp": 1.01554978,
      "epoch": 0.9144471833102867,
      "flos": 33323414019840.0,
      "grad_norm": 1.6547219781787705,
      "language_loss": 0.73740554,
      "learning_rate": 7.619830938602013e-08,
      "loss": 0.75929457,
      "num_input_tokens_seen": 164472755,
      "step": 7605,
      "time_per_iteration": 2.7486422061920166
    },
    {
      "auxiliary_loss_clip": 0.01164418,
      "auxiliary_loss_mlp": 0.0102171,
      "balance_loss_clip": 1.04540253,
      "balance_loss_mlp": 1.0143491,
      "epoch": 0.9145674262009259,
      "flos": 21068970428160.0,
      "grad_norm": 1.8405749186429308,
      "language_loss": 0.82456726,
      "learning_rate": 7.598548559689777e-08,
      "loss": 0.84642851,
      "num_input_tokens_seen": 164491155,
      "step": 7606,
      "time_per_iteration": 2.674974203109741
    },
    {
      "auxiliary_loss_clip": 0.01169431,
      "auxiliary_loss_mlp": 0.01019769,
      "balance_loss_clip": 1.04167867,
      "balance_loss_mlp": 1.01262879,
      "epoch": 0.914687669091565,
      "flos": 16800269212800.0,
      "grad_norm": 5.557460884335564,
      "language_loss": 0.81696123,
      "learning_rate": 7.577295367924751e-08,
      "loss": 0.83885324,
      "num_input_tokens_seen": 164507555,
      "step": 7607,
      "time_per_iteration": 2.633223056793213
    },
    {
      "auxiliary_loss_clip": 0.01176339,
      "auxiliary_loss_mlp": 0.01019395,
      "balance_loss_clip": 1.04586625,
      "balance_loss_mlp": 1.01163411,
      "epoch": 0.914807911982204,
      "flos": 25773627012480.0,
      "grad_norm": 1.7692708218614435,
      "language_loss": 0.82082665,
      "learning_rate": 7.556071366531002e-08,
      "loss": 0.84278393,
      "num_input_tokens_seen": 164528525,
      "step": 7608,
      "time_per_iteration": 2.688936710357666
    },
    {
      "auxiliary_loss_clip": 0.01171198,
      "auxiliary_loss_mlp": 0.01022747,
      "balance_loss_clip": 1.04826307,
      "balance_loss_mlp": 1.01505828,
      "epoch": 0.9149281548728432,
      "flos": 19208043636480.0,
      "grad_norm": 2.2091439413298994,
      "language_loss": 0.79450184,
      "learning_rate": 7.53487655872822e-08,
      "loss": 0.8164413,
      "num_input_tokens_seen": 164547695,
      "step": 7609,
      "time_per_iteration": 2.5734565258026123
    },
    {
      "auxiliary_loss_clip": 0.01179906,
      "auxiliary_loss_mlp": 0.01022453,
      "balance_loss_clip": 1.04009819,
      "balance_loss_mlp": 1.0145731,
      "epoch": 0.9150483977634822,
      "flos": 26870554500480.0,
      "grad_norm": 1.7568677659236716,
      "language_loss": 0.74431324,
      "learning_rate": 7.513710947731656e-08,
      "loss": 0.76633686,
      "num_input_tokens_seen": 164568905,
      "step": 7610,
      "time_per_iteration": 2.7057995796203613
    },
    {
      "auxiliary_loss_clip": 0.01162862,
      "auxiliary_loss_mlp": 0.01023356,
      "balance_loss_clip": 1.04286885,
      "balance_loss_mlp": 1.01578879,
      "epoch": 0.9151686406541213,
      "flos": 21908956953600.0,
      "grad_norm": 1.635479401782676,
      "language_loss": 0.8542974,
      "learning_rate": 7.492574536752095e-08,
      "loss": 0.87615955,
      "num_input_tokens_seen": 164588895,
      "step": 7611,
      "time_per_iteration": 2.6245288848876953
    },
    {
      "auxiliary_loss_clip": 0.0116448,
      "auxiliary_loss_mlp": 0.01026233,
      "balance_loss_clip": 1.04585886,
      "balance_loss_mlp": 1.01938426,
      "epoch": 0.9152888835447605,
      "flos": 27308556944640.0,
      "grad_norm": 2.295236700746716,
      "language_loss": 0.7792154,
      "learning_rate": 7.471467328995907e-08,
      "loss": 0.80112255,
      "num_input_tokens_seen": 164607705,
      "step": 7612,
      "time_per_iteration": 2.6862542629241943
    },
    {
      "auxiliary_loss_clip": 0.01193528,
      "auxiliary_loss_mlp": 0.01021932,
      "balance_loss_clip": 1.03753185,
      "balance_loss_mlp": 1.01411188,
      "epoch": 0.9154091264353995,
      "flos": 13370728510080.0,
      "grad_norm": 2.8077934864374656,
      "language_loss": 0.60809553,
      "learning_rate": 7.450389327665018e-08,
      "loss": 0.6302501,
      "num_input_tokens_seen": 164625540,
      "step": 7613,
      "time_per_iteration": 2.673807144165039
    },
    {
      "auxiliary_loss_clip": 0.01174723,
      "auxiliary_loss_mlp": 0.01025626,
      "balance_loss_clip": 1.04636693,
      "balance_loss_mlp": 1.01794326,
      "epoch": 0.9155293693260386,
      "flos": 20193037367040.0,
      "grad_norm": 4.068087755676112,
      "language_loss": 0.68511266,
      "learning_rate": 7.429340535957029e-08,
      "loss": 0.70711607,
      "num_input_tokens_seen": 164640735,
      "step": 7614,
      "time_per_iteration": 3.462736129760742
    },
    {
      "auxiliary_loss_clip": 0.01173446,
      "auxiliary_loss_mlp": 0.01022363,
      "balance_loss_clip": 1.04471064,
      "balance_loss_mlp": 1.01546097,
      "epoch": 0.9156496122166777,
      "flos": 19354990176000.0,
      "grad_norm": 2.5730142828612705,
      "language_loss": 0.70937073,
      "learning_rate": 7.40832095706494e-08,
      "loss": 0.73132879,
      "num_input_tokens_seen": 164657430,
      "step": 7615,
      "time_per_iteration": 2.6232075691223145
    },
    {
      "auxiliary_loss_clip": 0.01177593,
      "auxiliary_loss_mlp": 0.01025838,
      "balance_loss_clip": 1.04361498,
      "balance_loss_mlp": 1.01905489,
      "epoch": 0.9157698551073168,
      "flos": 21107287261440.0,
      "grad_norm": 1.7340479212824975,
      "language_loss": 0.80204165,
      "learning_rate": 7.387330594177443e-08,
      "loss": 0.82407594,
      "num_input_tokens_seen": 164679505,
      "step": 7616,
      "time_per_iteration": 3.589362382888794
    },
    {
      "auxiliary_loss_clip": 0.0116713,
      "auxiliary_loss_mlp": 0.01025912,
      "balance_loss_clip": 1.0420835,
      "balance_loss_mlp": 1.01923919,
      "epoch": 0.9158900979979558,
      "flos": 25193167228800.0,
      "grad_norm": 1.9614463340037844,
      "language_loss": 0.79072642,
      "learning_rate": 7.366369450478749e-08,
      "loss": 0.81265676,
      "num_input_tokens_seen": 164700615,
      "step": 7617,
      "time_per_iteration": 2.6709604263305664
    },
    {
      "auxiliary_loss_clip": 0.01170207,
      "auxiliary_loss_mlp": 0.01026592,
      "balance_loss_clip": 1.04249072,
      "balance_loss_mlp": 1.01910579,
      "epoch": 0.916010340888595,
      "flos": 30146648302080.0,
      "grad_norm": 2.061149106789414,
      "language_loss": 0.66570133,
      "learning_rate": 7.345437529148646e-08,
      "loss": 0.68766928,
      "num_input_tokens_seen": 164719625,
      "step": 7618,
      "time_per_iteration": 2.775324821472168
    },
    {
      "auxiliary_loss_clip": 0.0117111,
      "auxiliary_loss_mlp": 0.0102845,
      "balance_loss_clip": 1.04177141,
      "balance_loss_mlp": 1.02124953,
      "epoch": 0.9161305837792341,
      "flos": 17091827907840.0,
      "grad_norm": 1.9382034694578718,
      "language_loss": 0.72702724,
      "learning_rate": 7.324534833362483e-08,
      "loss": 0.74902284,
      "num_input_tokens_seen": 164737200,
      "step": 7619,
      "time_per_iteration": 2.6990859508514404
    },
    {
      "auxiliary_loss_clip": 0.01171098,
      "auxiliary_loss_mlp": 0.01022889,
      "balance_loss_clip": 1.04487753,
      "balance_loss_mlp": 1.01560545,
      "epoch": 0.9162508266698731,
      "flos": 22893699288960.0,
      "grad_norm": 1.7990584093901543,
      "language_loss": 0.68449175,
      "learning_rate": 7.303661366291192e-08,
      "loss": 0.70643163,
      "num_input_tokens_seen": 164757870,
      "step": 7620,
      "time_per_iteration": 3.7023820877075195
    },
    {
      "auxiliary_loss_clip": 0.01174244,
      "auxiliary_loss_mlp": 0.01024001,
      "balance_loss_clip": 1.04027081,
      "balance_loss_mlp": 1.01677692,
      "epoch": 0.9163710695605123,
      "flos": 19974808287360.0,
      "grad_norm": 2.0419348375182236,
      "language_loss": 0.81577915,
      "learning_rate": 7.28281713110126e-08,
      "loss": 0.83776164,
      "num_input_tokens_seen": 164775945,
      "step": 7621,
      "time_per_iteration": 2.6928863525390625
    },
    {
      "auxiliary_loss_clip": 0.01164616,
      "auxiliary_loss_mlp": 0.01024131,
      "balance_loss_clip": 1.04316807,
      "balance_loss_mlp": 1.01705563,
      "epoch": 0.9164913124511513,
      "flos": 22783812606720.0,
      "grad_norm": 1.8004887472336668,
      "language_loss": 0.77615321,
      "learning_rate": 7.262002130954759e-08,
      "loss": 0.79804075,
      "num_input_tokens_seen": 164794400,
      "step": 7622,
      "time_per_iteration": 2.7242255210876465
    },
    {
      "auxiliary_loss_clip": 0.01180914,
      "auxiliary_loss_mlp": 0.01026518,
      "balance_loss_clip": 1.04257584,
      "balance_loss_mlp": 1.01888251,
      "epoch": 0.9166115553417904,
      "flos": 24900854348160.0,
      "grad_norm": 2.198586987495513,
      "language_loss": 0.79290795,
      "learning_rate": 7.241216369009296e-08,
      "loss": 0.8149823,
      "num_input_tokens_seen": 164814585,
      "step": 7623,
      "time_per_iteration": 2.781240940093994
    },
    {
      "auxiliary_loss_clip": 0.01163932,
      "auxiliary_loss_mlp": 0.01019608,
      "balance_loss_clip": 1.04531384,
      "balance_loss_mlp": 1.01259553,
      "epoch": 0.9167317982324296,
      "flos": 25702919089920.0,
      "grad_norm": 12.9271409061881,
      "language_loss": 0.66380155,
      "learning_rate": 7.220459848418037e-08,
      "loss": 0.68563694,
      "num_input_tokens_seen": 164834660,
      "step": 7624,
      "time_per_iteration": 2.6246519088745117
    },
    {
      "auxiliary_loss_clip": 0.01164613,
      "auxiliary_loss_mlp": 0.0102224,
      "balance_loss_clip": 1.04838896,
      "balance_loss_mlp": 1.01526642,
      "epoch": 0.9168520411230686,
      "flos": 15632813370240.0,
      "grad_norm": 2.119528647473502,
      "language_loss": 0.80051768,
      "learning_rate": 7.199732572329708e-08,
      "loss": 0.82238621,
      "num_input_tokens_seen": 164852560,
      "step": 7625,
      "time_per_iteration": 2.5336244106292725
    },
    {
      "auxiliary_loss_clip": 0.0117798,
      "auxiliary_loss_mlp": 0.01027147,
      "balance_loss_clip": 1.04364944,
      "balance_loss_mlp": 1.01946735,
      "epoch": 0.9169722840137077,
      "flos": 30258151096320.0,
      "grad_norm": 2.2199595524912827,
      "language_loss": 0.76071596,
      "learning_rate": 7.179034543888684e-08,
      "loss": 0.78276718,
      "num_input_tokens_seen": 164872065,
      "step": 7626,
      "time_per_iteration": 2.6578593254089355
    },
    {
      "auxiliary_loss_clip": 0.01170526,
      "auxiliary_loss_mlp": 0.01024854,
      "balance_loss_clip": 1.04434621,
      "balance_loss_mlp": 1.01759458,
      "epoch": 0.9170925269043467,
      "flos": 22491643380480.0,
      "grad_norm": 3.5457509140879253,
      "language_loss": 0.77437449,
      "learning_rate": 7.158365766234808e-08,
      "loss": 0.79632825,
      "num_input_tokens_seen": 164890915,
      "step": 7627,
      "time_per_iteration": 2.6208324432373047
    },
    {
      "auxiliary_loss_clip": 0.01165221,
      "auxiliary_loss_mlp": 0.01023135,
      "balance_loss_clip": 1.03949773,
      "balance_loss_mlp": 1.01499915,
      "epoch": 0.9172127697949859,
      "flos": 22893914770560.0,
      "grad_norm": 2.42404516665021,
      "language_loss": 0.7262426,
      "learning_rate": 7.137726242503527e-08,
      "loss": 0.74812615,
      "num_input_tokens_seen": 164909835,
      "step": 7628,
      "time_per_iteration": 2.676797389984131
    },
    {
      "auxiliary_loss_clip": 0.01167733,
      "auxiliary_loss_mlp": 0.01208495,
      "balance_loss_clip": 1.04638696,
      "balance_loss_mlp": 1.00073719,
      "epoch": 0.917333012685625,
      "flos": 17451867882240.0,
      "grad_norm": 2.33281709437122,
      "language_loss": 0.77997416,
      "learning_rate": 7.11711597582585e-08,
      "loss": 0.80373645,
      "num_input_tokens_seen": 164927195,
      "step": 7629,
      "time_per_iteration": 2.684004545211792
    },
    {
      "auxiliary_loss_clip": 0.01171587,
      "auxiliary_loss_mlp": 0.01021189,
      "balance_loss_clip": 1.03865731,
      "balance_loss_mlp": 1.01445949,
      "epoch": 0.917453255576264,
      "flos": 14318949692160.0,
      "grad_norm": 2.30263233754921,
      "language_loss": 0.8012054,
      "learning_rate": 7.096534969328271e-08,
      "loss": 0.82313317,
      "num_input_tokens_seen": 164944640,
      "step": 7630,
      "time_per_iteration": 2.6391043663024902
    },
    {
      "auxiliary_loss_clip": 0.01173938,
      "auxiliary_loss_mlp": 0.01021668,
      "balance_loss_clip": 1.04240811,
      "balance_loss_mlp": 1.01453292,
      "epoch": 0.9175734984669032,
      "flos": 20741177888640.0,
      "grad_norm": 2.0550252963884406,
      "language_loss": 0.84275025,
      "learning_rate": 7.075983226132987e-08,
      "loss": 0.86470628,
      "num_input_tokens_seen": 164963570,
      "step": 7631,
      "time_per_iteration": 3.6149182319641113
    },
    {
      "auxiliary_loss_clip": 0.011748,
      "auxiliary_loss_mlp": 0.01208214,
      "balance_loss_clip": 1.04273736,
      "balance_loss_mlp": 1.00069344,
      "epoch": 0.9176937413575422,
      "flos": 14830497233280.0,
      "grad_norm": 8.810159924745914,
      "language_loss": 0.79606879,
      "learning_rate": 7.055460749357656e-08,
      "loss": 0.8198989,
      "num_input_tokens_seen": 164979850,
      "step": 7632,
      "time_per_iteration": 2.6218104362487793
    },
    {
      "auxiliary_loss_clip": 0.01173528,
      "auxiliary_loss_mlp": 0.01026348,
      "balance_loss_clip": 1.0465312,
      "balance_loss_mlp": 1.0187906,
      "epoch": 0.9178139842481813,
      "flos": 18474603828480.0,
      "grad_norm": 1.7531612826033782,
      "language_loss": 0.70140839,
      "learning_rate": 7.034967542115521e-08,
      "loss": 0.72340715,
      "num_input_tokens_seen": 164998115,
      "step": 7633,
      "time_per_iteration": 2.63356351852417
    },
    {
      "auxiliary_loss_clip": 0.01158841,
      "auxiliary_loss_mlp": 0.01207329,
      "balance_loss_clip": 1.04353595,
      "balance_loss_mlp": 1.00074279,
      "epoch": 0.9179342271388204,
      "flos": 20047455544320.0,
      "grad_norm": 2.6582487455090282,
      "language_loss": 0.7574752,
      "learning_rate": 7.014503607515388e-08,
      "loss": 0.78113687,
      "num_input_tokens_seen": 165017420,
      "step": 7634,
      "time_per_iteration": 2.626668930053711
    },
    {
      "auxiliary_loss_clip": 0.01172999,
      "auxiliary_loss_mlp": 0.01027864,
      "balance_loss_clip": 1.04852009,
      "balance_loss_mlp": 1.02041304,
      "epoch": 0.9180544700294595,
      "flos": 24676232647680.0,
      "grad_norm": 2.91864546101214,
      "language_loss": 0.68360668,
      "learning_rate": 6.994068948661592e-08,
      "loss": 0.70561534,
      "num_input_tokens_seen": 165035575,
      "step": 7635,
      "time_per_iteration": 2.630631923675537
    },
    {
      "auxiliary_loss_clip": 0.01170189,
      "auxiliary_loss_mlp": 0.01025211,
      "balance_loss_clip": 1.04689062,
      "balance_loss_mlp": 1.01684844,
      "epoch": 0.9181747129200986,
      "flos": 16727478301440.0,
      "grad_norm": 2.0324492574956703,
      "language_loss": 0.77141553,
      "learning_rate": 6.973663568654142e-08,
      "loss": 0.79336953,
      "num_input_tokens_seen": 165053280,
      "step": 7636,
      "time_per_iteration": 2.5618350505828857
    },
    {
      "auxiliary_loss_clip": 0.01165473,
      "auxiliary_loss_mlp": 0.01028549,
      "balance_loss_clip": 1.04818559,
      "balance_loss_mlp": 1.02090466,
      "epoch": 0.9182949558107377,
      "flos": 24271626873600.0,
      "grad_norm": 2.7586596310569664,
      "language_loss": 0.65488875,
      "learning_rate": 6.953287470588386e-08,
      "loss": 0.67682898,
      "num_input_tokens_seen": 165071235,
      "step": 7637,
      "time_per_iteration": 2.6508443355560303
    },
    {
      "auxiliary_loss_clip": 0.01172581,
      "auxiliary_loss_mlp": 0.01024138,
      "balance_loss_clip": 1.0448277,
      "balance_loss_mlp": 1.01663399,
      "epoch": 0.9184151987013768,
      "flos": 22082117443200.0,
      "grad_norm": 2.7997748592234015,
      "language_loss": 0.85730296,
      "learning_rate": 6.932940657555452e-08,
      "loss": 0.8792702,
      "num_input_tokens_seen": 165087365,
      "step": 7638,
      "time_per_iteration": 2.6181089878082275
    },
    {
      "auxiliary_loss_clip": 0.01159716,
      "auxiliary_loss_mlp": 0.01021022,
      "balance_loss_clip": 1.04551625,
      "balance_loss_mlp": 1.01426232,
      "epoch": 0.9185354415920158,
      "flos": 32166732257280.0,
      "grad_norm": 1.5540537129335485,
      "language_loss": 0.76504183,
      "learning_rate": 6.912623132641938e-08,
      "loss": 0.78684926,
      "num_input_tokens_seen": 165112455,
      "step": 7639,
      "time_per_iteration": 2.777719497680664
    },
    {
      "auxiliary_loss_clip": 0.01172863,
      "auxiliary_loss_mlp": 0.01026774,
      "balance_loss_clip": 1.0450151,
      "balance_loss_mlp": 1.01901352,
      "epoch": 0.918655684482655,
      "flos": 20997831542400.0,
      "grad_norm": 2.172921516971412,
      "language_loss": 0.77024859,
      "learning_rate": 6.892334898929952e-08,
      "loss": 0.79224497,
      "num_input_tokens_seen": 165132700,
      "step": 7640,
      "time_per_iteration": 2.6832852363586426
    },
    {
      "auxiliary_loss_clip": 0.01163086,
      "auxiliary_loss_mlp": 0.01023626,
      "balance_loss_clip": 1.04407108,
      "balance_loss_mlp": 1.01650894,
      "epoch": 0.918775927373294,
      "flos": 15560704817280.0,
      "grad_norm": 1.9994384565175574,
      "language_loss": 0.85007381,
      "learning_rate": 6.872075959497236e-08,
      "loss": 0.87194097,
      "num_input_tokens_seen": 165151475,
      "step": 7641,
      "time_per_iteration": 3.60410737991333
    },
    {
      "auxiliary_loss_clip": 0.01169781,
      "auxiliary_loss_mlp": 0.01021553,
      "balance_loss_clip": 1.04378211,
      "balance_loss_mlp": 1.01449633,
      "epoch": 0.9188961702639331,
      "flos": 29934057657600.0,
      "grad_norm": 2.446218087587961,
      "language_loss": 0.83031738,
      "learning_rate": 6.85184631741702e-08,
      "loss": 0.85223073,
      "num_input_tokens_seen": 165172040,
      "step": 7642,
      "time_per_iteration": 2.7468245029449463
    },
    {
      "auxiliary_loss_clip": 0.01168008,
      "auxiliary_loss_mlp": 0.01021543,
      "balance_loss_clip": 1.0445863,
      "balance_loss_mlp": 1.01393735,
      "epoch": 0.9190164131545723,
      "flos": 20701244943360.0,
      "grad_norm": 2.3707301608543943,
      "language_loss": 0.77384937,
      "learning_rate": 6.831645975758161e-08,
      "loss": 0.79574484,
      "num_input_tokens_seen": 165189980,
      "step": 7643,
      "time_per_iteration": 3.6150965690612793
    },
    {
      "auxiliary_loss_clip": 0.01163931,
      "auxiliary_loss_mlp": 0.01022017,
      "balance_loss_clip": 1.04406524,
      "balance_loss_mlp": 1.01484668,
      "epoch": 0.9191366560452113,
      "flos": 25629912696960.0,
      "grad_norm": 3.0373102318551286,
      "language_loss": 0.67415434,
      "learning_rate": 6.811474937585026e-08,
      "loss": 0.69601381,
      "num_input_tokens_seen": 165209770,
      "step": 7644,
      "time_per_iteration": 2.6329030990600586
    },
    {
      "auxiliary_loss_clip": 0.01169678,
      "auxiliary_loss_mlp": 0.01022458,
      "balance_loss_clip": 1.04211879,
      "balance_loss_mlp": 1.01519537,
      "epoch": 0.9192568989358504,
      "flos": 21434325615360.0,
      "grad_norm": 1.7974160931189544,
      "language_loss": 0.78892875,
      "learning_rate": 6.79133320595755e-08,
      "loss": 0.81085014,
      "num_input_tokens_seen": 165229690,
      "step": 7645,
      "time_per_iteration": 2.729130268096924
    },
    {
      "auxiliary_loss_clip": 0.01174549,
      "auxiliary_loss_mlp": 0.01021487,
      "balance_loss_clip": 1.04690814,
      "balance_loss_mlp": 1.01454043,
      "epoch": 0.9193771418264896,
      "flos": 23185078416000.0,
      "grad_norm": 3.4304355356067715,
      "language_loss": 0.75331086,
      "learning_rate": 6.771220783931198e-08,
      "loss": 0.77527118,
      "num_input_tokens_seen": 165249850,
      "step": 7646,
      "time_per_iteration": 3.617279291152954
    },
    {
      "auxiliary_loss_clip": 0.01114115,
      "auxiliary_loss_mlp": 0.01198595,
      "balance_loss_clip": 1.02389908,
      "balance_loss_mlp": 0.99983579,
      "epoch": 0.9194973847171286,
      "flos": 70582963184640.0,
      "grad_norm": 0.8610683709922536,
      "language_loss": 0.64563751,
      "learning_rate": 6.751137674556994e-08,
      "loss": 0.66876459,
      "num_input_tokens_seen": 165310235,
      "step": 7647,
      "time_per_iteration": 3.4931375980377197
    },
    {
      "auxiliary_loss_clip": 0.01170707,
      "auxiliary_loss_mlp": 0.01021157,
      "balance_loss_clip": 1.04416752,
      "balance_loss_mlp": 1.01373601,
      "epoch": 0.9196176276077677,
      "flos": 14720682378240.0,
      "grad_norm": 1.9031675624861015,
      "language_loss": 0.77625,
      "learning_rate": 6.731083880881572e-08,
      "loss": 0.79816866,
      "num_input_tokens_seen": 165326455,
      "step": 7648,
      "time_per_iteration": 2.8336145877838135
    },
    {
      "auxiliary_loss_clip": 0.01171105,
      "auxiliary_loss_mlp": 0.01020389,
      "balance_loss_clip": 1.04350936,
      "balance_loss_mlp": 1.01344776,
      "epoch": 0.9197378704984068,
      "flos": 23294893271040.0,
      "grad_norm": 3.850964257444618,
      "language_loss": 0.81025845,
      "learning_rate": 6.711059405947072e-08,
      "loss": 0.83217335,
      "num_input_tokens_seen": 165344645,
      "step": 7649,
      "time_per_iteration": 2.6459438800811768
    },
    {
      "auxiliary_loss_clip": 0.01171823,
      "auxiliary_loss_mlp": 0.01022663,
      "balance_loss_clip": 1.04390574,
      "balance_loss_mlp": 1.01503372,
      "epoch": 0.9198581133890459,
      "flos": 20302564913280.0,
      "grad_norm": 1.9380847528456928,
      "language_loss": 0.77279186,
      "learning_rate": 6.691064252791156e-08,
      "loss": 0.79473674,
      "num_input_tokens_seen": 165364120,
      "step": 7650,
      "time_per_iteration": 2.642129898071289
    },
    {
      "auxiliary_loss_clip": 0.01168117,
      "auxiliary_loss_mlp": 0.01023873,
      "balance_loss_clip": 1.04095316,
      "balance_loss_mlp": 1.01658344,
      "epoch": 0.9199783562796849,
      "flos": 17675663569920.0,
      "grad_norm": 1.9807812950360215,
      "language_loss": 0.77916896,
      "learning_rate": 6.67109842444713e-08,
      "loss": 0.80108887,
      "num_input_tokens_seen": 165383050,
      "step": 7651,
      "time_per_iteration": 2.628207206726074
    },
    {
      "auxiliary_loss_clip": 0.01167444,
      "auxiliary_loss_mlp": 0.012082,
      "balance_loss_clip": 1.04793298,
      "balance_loss_mlp": 1.00065374,
      "epoch": 0.9200985991703241,
      "flos": 17676022705920.0,
      "grad_norm": 2.212266482128683,
      "language_loss": 0.76876819,
      "learning_rate": 6.651161923943704e-08,
      "loss": 0.79252464,
      "num_input_tokens_seen": 165400955,
      "step": 7652,
      "time_per_iteration": 2.6946492195129395
    },
    {
      "auxiliary_loss_clip": 0.01163783,
      "auxiliary_loss_mlp": 0.01026153,
      "balance_loss_clip": 1.0438149,
      "balance_loss_mlp": 1.01841044,
      "epoch": 0.9202188420609632,
      "flos": 20996574566400.0,
      "grad_norm": 3.827574168568271,
      "language_loss": 0.77395892,
      "learning_rate": 6.631254754305326e-08,
      "loss": 0.79585826,
      "num_input_tokens_seen": 165420415,
      "step": 7653,
      "time_per_iteration": 2.5898001194000244
    },
    {
      "auxiliary_loss_clip": 0.01168278,
      "auxiliary_loss_mlp": 0.01023736,
      "balance_loss_clip": 1.04712963,
      "balance_loss_mlp": 1.01610947,
      "epoch": 0.9203390849516022,
      "flos": 13918222586880.0,
      "grad_norm": 2.4039380346940202,
      "language_loss": 0.78387666,
      "learning_rate": 6.611376918551848e-08,
      "loss": 0.8057968,
      "num_input_tokens_seen": 165439200,
      "step": 7654,
      "time_per_iteration": 2.619682550430298
    },
    {
      "auxiliary_loss_clip": 0.01170249,
      "auxiliary_loss_mlp": 0.01207633,
      "balance_loss_clip": 1.04042578,
      "balance_loss_mlp": 1.00054324,
      "epoch": 0.9204593278422414,
      "flos": 21175912195200.0,
      "grad_norm": 2.367320694055107,
      "language_loss": 0.79655069,
      "learning_rate": 6.591528419698744e-08,
      "loss": 0.82032949,
      "num_input_tokens_seen": 165458985,
      "step": 7655,
      "time_per_iteration": 2.62298846244812
    },
    {
      "auxiliary_loss_clip": 0.01173675,
      "auxiliary_loss_mlp": 0.01025678,
      "balance_loss_clip": 1.04328036,
      "balance_loss_mlp": 1.01892138,
      "epoch": 0.9205795707328804,
      "flos": 14501375890560.0,
      "grad_norm": 2.3638635006686464,
      "language_loss": 0.83599901,
      "learning_rate": 6.571709260756986e-08,
      "loss": 0.85799253,
      "num_input_tokens_seen": 165475630,
      "step": 7656,
      "time_per_iteration": 2.61470103263855
    },
    {
      "auxiliary_loss_clip": 0.0117437,
      "auxiliary_loss_mlp": 0.01027019,
      "balance_loss_clip": 1.0504086,
      "balance_loss_mlp": 1.01924074,
      "epoch": 0.9206998136235195,
      "flos": 22417559579520.0,
      "grad_norm": 2.3862542739408346,
      "language_loss": 0.76526189,
      "learning_rate": 6.551919444733122e-08,
      "loss": 0.78727579,
      "num_input_tokens_seen": 165493445,
      "step": 7657,
      "time_per_iteration": 2.5849051475524902
    },
    {
      "auxiliary_loss_clip": 0.01167389,
      "auxiliary_loss_mlp": 0.01024526,
      "balance_loss_clip": 1.043993,
      "balance_loss_mlp": 1.01674175,
      "epoch": 0.9208200565141585,
      "flos": 53358407544960.0,
      "grad_norm": 3.926258212634378,
      "language_loss": 0.65811163,
      "learning_rate": 6.53215897462931e-08,
      "loss": 0.68003082,
      "num_input_tokens_seen": 165517200,
      "step": 7658,
      "time_per_iteration": 3.8659448623657227
    },
    {
      "auxiliary_loss_clip": 0.01166651,
      "auxiliary_loss_mlp": 0.01027613,
      "balance_loss_clip": 1.04534245,
      "balance_loss_mlp": 1.01991248,
      "epoch": 0.9209402994047977,
      "flos": 30589139946240.0,
      "grad_norm": 2.58500114661039,
      "language_loss": 0.75089842,
      "learning_rate": 6.512427853443103e-08,
      "loss": 0.77284098,
      "num_input_tokens_seen": 165539280,
      "step": 7659,
      "time_per_iteration": 2.708263397216797
    },
    {
      "auxiliary_loss_clip": 0.01170754,
      "auxiliary_loss_mlp": 0.01021105,
      "balance_loss_clip": 1.04558468,
      "balance_loss_mlp": 1.01346374,
      "epoch": 0.9210605422954368,
      "flos": 29132711187840.0,
      "grad_norm": 2.0660031722934513,
      "language_loss": 0.75876892,
      "learning_rate": 6.492726084167799e-08,
      "loss": 0.78068751,
      "num_input_tokens_seen": 165561395,
      "step": 7660,
      "time_per_iteration": 2.674739360809326
    },
    {
      "auxiliary_loss_clip": 0.01063339,
      "auxiliary_loss_mlp": 0.01002979,
      "balance_loss_clip": 1.0106616,
      "balance_loss_mlp": 1.00224936,
      "epoch": 0.9211807851860758,
      "flos": 54853838472960.0,
      "grad_norm": 0.7823247995021935,
      "language_loss": 0.57451296,
      "learning_rate": 6.473053669792072e-08,
      "loss": 0.5951761,
      "num_input_tokens_seen": 165616085,
      "step": 7661,
      "time_per_iteration": 3.0493621826171875
    },
    {
      "auxiliary_loss_clip": 0.01166634,
      "auxiliary_loss_mlp": 0.01023739,
      "balance_loss_clip": 1.04353988,
      "balance_loss_mlp": 1.01559091,
      "epoch": 0.921301028076715,
      "flos": 19201974238080.0,
      "grad_norm": 4.506360400403197,
      "language_loss": 0.72044587,
      "learning_rate": 6.453410613300248e-08,
      "loss": 0.74234962,
      "num_input_tokens_seen": 165634015,
      "step": 7662,
      "time_per_iteration": 2.597381830215454
    },
    {
      "auxiliary_loss_clip": 0.01177253,
      "auxiliary_loss_mlp": 0.01028373,
      "balance_loss_clip": 1.04048872,
      "balance_loss_mlp": 1.02100551,
      "epoch": 0.921421270967354,
      "flos": 27526893765120.0,
      "grad_norm": 1.711524454449915,
      "language_loss": 0.58144093,
      "learning_rate": 6.43379691767214e-08,
      "loss": 0.60349721,
      "num_input_tokens_seen": 165653220,
      "step": 7663,
      "time_per_iteration": 2.757535696029663
    },
    {
      "auxiliary_loss_clip": 0.01092852,
      "auxiliary_loss_mlp": 0.01001329,
      "balance_loss_clip": 1.01346374,
      "balance_loss_mlp": 1.00056922,
      "epoch": 0.9215415138579931,
      "flos": 70209311955840.0,
      "grad_norm": 0.7174366018956767,
      "language_loss": 0.55059785,
      "learning_rate": 6.414212585883105e-08,
      "loss": 0.57153964,
      "num_input_tokens_seen": 165715850,
      "step": 7664,
      "time_per_iteration": 3.308479070663452
    },
    {
      "auxiliary_loss_clip": 0.01174917,
      "auxiliary_loss_mlp": 0.01020651,
      "balance_loss_clip": 1.0453589,
      "balance_loss_mlp": 1.0128547,
      "epoch": 0.9216617567486323,
      "flos": 35553107790720.0,
      "grad_norm": 1.4898548793574362,
      "language_loss": 0.69876719,
      "learning_rate": 6.394657620904143e-08,
      "loss": 0.72072291,
      "num_input_tokens_seen": 165738960,
      "step": 7665,
      "time_per_iteration": 2.716580629348755
    },
    {
      "auxiliary_loss_clip": 0.011715,
      "auxiliary_loss_mlp": 0.01024042,
      "balance_loss_clip": 1.04943776,
      "balance_loss_mlp": 1.01612401,
      "epoch": 0.9217819996392713,
      "flos": 29533330552320.0,
      "grad_norm": 1.9066655873176186,
      "language_loss": 0.71486747,
      "learning_rate": 6.375132025701657e-08,
      "loss": 0.7368229,
      "num_input_tokens_seen": 165761260,
      "step": 7666,
      "time_per_iteration": 2.766972064971924
    },
    {
      "auxiliary_loss_clip": 0.0117144,
      "auxiliary_loss_mlp": 0.01025599,
      "balance_loss_clip": 1.0498898,
      "balance_loss_mlp": 1.01791632,
      "epoch": 0.9219022425299104,
      "flos": 14574669592320.0,
      "grad_norm": 2.347724941616994,
      "language_loss": 0.69154298,
      "learning_rate": 6.355635803237724e-08,
      "loss": 0.71351337,
      "num_input_tokens_seen": 165776960,
      "step": 7667,
      "time_per_iteration": 2.588395118713379
    },
    {
      "auxiliary_loss_clip": 0.01168102,
      "auxiliary_loss_mlp": 0.01023914,
      "balance_loss_clip": 1.04435945,
      "balance_loss_mlp": 1.01616263,
      "epoch": 0.9220224854205495,
      "flos": 18077503996800.0,
      "grad_norm": 2.050566771544286,
      "language_loss": 0.79931855,
      "learning_rate": 6.336168956469867e-08,
      "loss": 0.82123864,
      "num_input_tokens_seen": 165795435,
      "step": 7668,
      "time_per_iteration": 3.570819854736328
    },
    {
      "auxiliary_loss_clip": 0.01164862,
      "auxiliary_loss_mlp": 0.01025459,
      "balance_loss_clip": 1.04428923,
      "balance_loss_mlp": 1.01874471,
      "epoch": 0.9221427283111886,
      "flos": 24790464875520.0,
      "grad_norm": 2.009520531835535,
      "language_loss": 0.71907508,
      "learning_rate": 6.316731488351168e-08,
      "loss": 0.74097824,
      "num_input_tokens_seen": 165816625,
      "step": 7669,
      "time_per_iteration": 3.6388967037200928
    },
    {
      "auxiliary_loss_clip": 0.01168564,
      "auxiliary_loss_mlp": 0.01024627,
      "balance_loss_clip": 1.04592991,
      "balance_loss_mlp": 1.01721859,
      "epoch": 0.9222629712018277,
      "flos": 13845036625920.0,
      "grad_norm": 1.9900039449785136,
      "language_loss": 0.63101363,
      "learning_rate": 6.297323401830334e-08,
      "loss": 0.65294546,
      "num_input_tokens_seen": 165835410,
      "step": 7670,
      "time_per_iteration": 2.5843565464019775
    },
    {
      "auxiliary_loss_clip": 0.01172781,
      "auxiliary_loss_mlp": 0.01025277,
      "balance_loss_clip": 1.04688501,
      "balance_loss_mlp": 1.01805925,
      "epoch": 0.9223832140924668,
      "flos": 21616177196160.0,
      "grad_norm": 2.1222423157535713,
      "language_loss": 0.68938839,
      "learning_rate": 6.277944699851523e-08,
      "loss": 0.71136898,
      "num_input_tokens_seen": 165854930,
      "step": 7671,
      "time_per_iteration": 2.653763771057129
    },
    {
      "auxiliary_loss_clip": 0.01164796,
      "auxiliary_loss_mlp": 0.01026747,
      "balance_loss_clip": 1.04643953,
      "balance_loss_mlp": 1.01916575,
      "epoch": 0.9225034569831059,
      "flos": 21142084561920.0,
      "grad_norm": 2.085780639169652,
      "language_loss": 0.73753864,
      "learning_rate": 6.25859538535447e-08,
      "loss": 0.75945401,
      "num_input_tokens_seen": 165875725,
      "step": 7672,
      "time_per_iteration": 2.576538562774658
    },
    {
      "auxiliary_loss_clip": 0.01171005,
      "auxiliary_loss_mlp": 0.0102136,
      "balance_loss_clip": 1.04418337,
      "balance_loss_mlp": 1.01377249,
      "epoch": 0.9226236998737449,
      "flos": 12495046844160.0,
      "grad_norm": 7.981177178026748,
      "language_loss": 0.77991879,
      "learning_rate": 6.239275461274474e-08,
      "loss": 0.80184245,
      "num_input_tokens_seen": 165892100,
      "step": 7673,
      "time_per_iteration": 3.56789231300354
    },
    {
      "auxiliary_loss_clip": 0.01170899,
      "auxiliary_loss_mlp": 0.01026355,
      "balance_loss_clip": 1.04706287,
      "balance_loss_mlp": 1.01915812,
      "epoch": 0.9227439427643841,
      "flos": 26214071581440.0,
      "grad_norm": 1.7162899288032334,
      "language_loss": 0.85874659,
      "learning_rate": 6.219984930542299e-08,
      "loss": 0.88071918,
      "num_input_tokens_seen": 165912840,
      "step": 7674,
      "time_per_iteration": 2.6480178833007812
    },
    {
      "auxiliary_loss_clip": 0.01171357,
      "auxiliary_loss_mlp": 0.01026012,
      "balance_loss_clip": 1.04599738,
      "balance_loss_mlp": 1.01891899,
      "epoch": 0.9228641856550232,
      "flos": 17967581400960.0,
      "grad_norm": 5.890348861600105,
      "language_loss": 0.75966227,
      "learning_rate": 6.200723796084383e-08,
      "loss": 0.781636,
      "num_input_tokens_seen": 165930935,
      "step": 7675,
      "time_per_iteration": 2.6273975372314453
    },
    {
      "auxiliary_loss_clip": 0.01085576,
      "auxiliary_loss_mlp": 0.0100258,
      "balance_loss_clip": 1.01012909,
      "balance_loss_mlp": 1.00188589,
      "epoch": 0.9229844285456622,
      "flos": 70420609710720.0,
      "grad_norm": 1.330585578740696,
      "language_loss": 0.62974483,
      "learning_rate": 6.181492060822546e-08,
      "loss": 0.65062636,
      "num_input_tokens_seen": 165991110,
      "step": 7676,
      "time_per_iteration": 3.0916121006011963
    },
    {
      "auxiliary_loss_clip": 0.0117472,
      "auxiliary_loss_mlp": 0.0102368,
      "balance_loss_clip": 1.04125977,
      "balance_loss_mlp": 1.01623583,
      "epoch": 0.9231046714363014,
      "flos": 17967832796160.0,
      "grad_norm": 2.115056373378294,
      "language_loss": 0.81870973,
      "learning_rate": 6.162289727674274e-08,
      "loss": 0.84069371,
      "num_input_tokens_seen": 166008790,
      "step": 7677,
      "time_per_iteration": 2.6032891273498535
    },
    {
      "auxiliary_loss_clip": 0.01173445,
      "auxiliary_loss_mlp": 0.01020524,
      "balance_loss_clip": 1.04248023,
      "balance_loss_mlp": 1.01380706,
      "epoch": 0.9232249143269404,
      "flos": 17858233422720.0,
      "grad_norm": 3.4953598687932823,
      "language_loss": 0.87992811,
      "learning_rate": 6.143116799552527e-08,
      "loss": 0.90186787,
      "num_input_tokens_seen": 166025035,
      "step": 7678,
      "time_per_iteration": 2.664029598236084
    },
    {
      "auxiliary_loss_clip": 0.01172289,
      "auxiliary_loss_mlp": 0.01020941,
      "balance_loss_clip": 1.04724622,
      "balance_loss_mlp": 1.01386571,
      "epoch": 0.9233451572175795,
      "flos": 23404384903680.0,
      "grad_norm": 3.053491403098985,
      "language_loss": 0.560408,
      "learning_rate": 6.123973279365802e-08,
      "loss": 0.58234036,
      "num_input_tokens_seen": 166044010,
      "step": 7679,
      "time_per_iteration": 2.59537410736084
    },
    {
      "auxiliary_loss_clip": 0.01173343,
      "auxiliary_loss_mlp": 0.01022301,
      "balance_loss_clip": 1.04741395,
      "balance_loss_mlp": 1.01539242,
      "epoch": 0.9234654001082186,
      "flos": 17999326045440.0,
      "grad_norm": 1.9321332000034235,
      "language_loss": 0.78101861,
      "learning_rate": 6.10485917001824e-08,
      "loss": 0.80297506,
      "num_input_tokens_seen": 166061865,
      "step": 7680,
      "time_per_iteration": 2.6381916999816895
    },
    {
      "auxiliary_loss_clip": 0.01174814,
      "auxiliary_loss_mlp": 0.01022914,
      "balance_loss_clip": 1.04426432,
      "balance_loss_mlp": 1.01604736,
      "epoch": 0.9235856429988577,
      "flos": 24750747411840.0,
      "grad_norm": 1.834618086873235,
      "language_loss": 0.81067491,
      "learning_rate": 6.085774474409322e-08,
      "loss": 0.83265209,
      "num_input_tokens_seen": 166082425,
      "step": 7681,
      "time_per_iteration": 2.643841028213501
    },
    {
      "auxiliary_loss_clip": 0.01172244,
      "auxiliary_loss_mlp": 0.0102602,
      "balance_loss_clip": 1.0461545,
      "balance_loss_mlp": 1.01886153,
      "epoch": 0.9237058858894968,
      "flos": 14099894599680.0,
      "grad_norm": 2.12914839328694,
      "language_loss": 0.70071173,
      "learning_rate": 6.066719195434267e-08,
      "loss": 0.7226944,
      "num_input_tokens_seen": 166100225,
      "step": 7682,
      "time_per_iteration": 2.6068320274353027
    },
    {
      "auxiliary_loss_clip": 0.01171421,
      "auxiliary_loss_mlp": 0.01025209,
      "balance_loss_clip": 1.04651809,
      "balance_loss_mlp": 1.01748419,
      "epoch": 0.9238261287801359,
      "flos": 28694529175680.0,
      "grad_norm": 1.8551272408402841,
      "language_loss": 0.66161346,
      "learning_rate": 6.047693335983717e-08,
      "loss": 0.68357968,
      "num_input_tokens_seen": 166122570,
      "step": 7683,
      "time_per_iteration": 2.634915351867676
    },
    {
      "auxiliary_loss_clip": 0.01171341,
      "auxiliary_loss_mlp": 0.01022439,
      "balance_loss_clip": 1.04509759,
      "balance_loss_mlp": 1.01511335,
      "epoch": 0.923946371670775,
      "flos": 23111856541440.0,
      "grad_norm": 3.9991157159982174,
      "language_loss": 0.82553244,
      "learning_rate": 6.028696898943853e-08,
      "loss": 0.84747022,
      "num_input_tokens_seen": 166141630,
      "step": 7684,
      "time_per_iteration": 2.638728618621826
    },
    {
      "auxiliary_loss_clip": 0.01172067,
      "auxiliary_loss_mlp": 0.01208462,
      "balance_loss_clip": 1.04218066,
      "balance_loss_mlp": 1.00077081,
      "epoch": 0.924066614561414,
      "flos": 21867120587520.0,
      "grad_norm": 2.2132620948267583,
      "language_loss": 0.7096293,
      "learning_rate": 6.00972988719648e-08,
      "loss": 0.73343468,
      "num_input_tokens_seen": 166159865,
      "step": 7685,
      "time_per_iteration": 3.5851566791534424
    },
    {
      "auxiliary_loss_clip": 0.01176315,
      "auxiliary_loss_mlp": 0.01208069,
      "balance_loss_clip": 1.04298615,
      "balance_loss_mlp": 1.00068903,
      "epoch": 0.9241868574520532,
      "flos": 28511887495680.0,
      "grad_norm": 3.547854350926138,
      "language_loss": 0.70673954,
      "learning_rate": 5.990792303618807e-08,
      "loss": 0.73058331,
      "num_input_tokens_seen": 166179445,
      "step": 7686,
      "time_per_iteration": 2.764315128326416
    },
    {
      "auxiliary_loss_clip": 0.01171977,
      "auxiliary_loss_mlp": 0.01021244,
      "balance_loss_clip": 1.04395282,
      "balance_loss_mlp": 1.01384664,
      "epoch": 0.9243071003426923,
      "flos": 30518324282880.0,
      "grad_norm": 1.8986485037648493,
      "language_loss": 0.69706392,
      "learning_rate": 5.971884151083695e-08,
      "loss": 0.71899617,
      "num_input_tokens_seen": 166201855,
      "step": 7687,
      "time_per_iteration": 2.7639925479888916
    },
    {
      "auxiliary_loss_clip": 0.01173512,
      "auxiliary_loss_mlp": 0.01025247,
      "balance_loss_clip": 1.04370689,
      "balance_loss_mlp": 1.01813591,
      "epoch": 0.9244273432333313,
      "flos": 28658331244800.0,
      "grad_norm": 1.8652921098835273,
      "language_loss": 0.74340832,
      "learning_rate": 5.9530054324595124e-08,
      "loss": 0.76539594,
      "num_input_tokens_seen": 166221970,
      "step": 7688,
      "time_per_iteration": 2.6724939346313477
    },
    {
      "auxiliary_loss_clip": 0.0106833,
      "auxiliary_loss_mlp": 0.01198675,
      "balance_loss_clip": 1.01246524,
      "balance_loss_mlp": 0.99995214,
      "epoch": 0.9245475861239704,
      "flos": 66230589237120.0,
      "grad_norm": 0.7169399733253757,
      "language_loss": 0.57555014,
      "learning_rate": 5.934156150610103e-08,
      "loss": 0.59822023,
      "num_input_tokens_seen": 166279335,
      "step": 7689,
      "time_per_iteration": 3.2512948513031006
    },
    {
      "auxiliary_loss_clip": 0.01169696,
      "auxiliary_loss_mlp": 0.01024714,
      "balance_loss_clip": 1.04382992,
      "balance_loss_mlp": 1.01684606,
      "epoch": 0.9246678290146095,
      "flos": 24239918142720.0,
      "grad_norm": 2.0998021301695395,
      "language_loss": 0.78973579,
      "learning_rate": 5.915336308394914e-08,
      "loss": 0.81167984,
      "num_input_tokens_seen": 166298170,
      "step": 7690,
      "time_per_iteration": 2.618666410446167
    },
    {
      "auxiliary_loss_clip": 0.01163847,
      "auxiliary_loss_mlp": 0.01022126,
      "balance_loss_clip": 1.04500747,
      "balance_loss_mlp": 1.01587009,
      "epoch": 0.9247880719052486,
      "flos": 18988808976000.0,
      "grad_norm": 1.5725337954922562,
      "language_loss": 0.77068156,
      "learning_rate": 5.89654590866886e-08,
      "loss": 0.79254127,
      "num_input_tokens_seen": 166317670,
      "step": 7691,
      "time_per_iteration": 2.6260790824890137
    },
    {
      "auxiliary_loss_clip": 0.01184755,
      "auxiliary_loss_mlp": 0.01024423,
      "balance_loss_clip": 1.04428935,
      "balance_loss_mlp": 1.01660299,
      "epoch": 0.9249083147958876,
      "flos": 24024095274240.0,
      "grad_norm": 1.9218208252754594,
      "language_loss": 0.88262999,
      "learning_rate": 5.877784954282483e-08,
      "loss": 0.90472174,
      "num_input_tokens_seen": 166337010,
      "step": 7692,
      "time_per_iteration": 2.747650384902954
    },
    {
      "auxiliary_loss_clip": 0.01172552,
      "auxiliary_loss_mlp": 0.01022204,
      "balance_loss_clip": 1.04602742,
      "balance_loss_mlp": 1.01431191,
      "epoch": 0.9250285576865268,
      "flos": 30773972355840.0,
      "grad_norm": 1.9899869056678519,
      "language_loss": 0.72540492,
      "learning_rate": 5.8590534480817963e-08,
      "loss": 0.74735248,
      "num_input_tokens_seen": 166358735,
      "step": 7693,
      "time_per_iteration": 2.7005116939544678
    },
    {
      "auxiliary_loss_clip": 0.01168084,
      "auxiliary_loss_mlp": 0.01024567,
      "balance_loss_clip": 1.04893541,
      "balance_loss_mlp": 1.01732516,
      "epoch": 0.9251488005771659,
      "flos": 10633581348480.0,
      "grad_norm": 2.352693323076527,
      "language_loss": 0.72739834,
      "learning_rate": 5.840351392908349e-08,
      "loss": 0.74932492,
      "num_input_tokens_seen": 166374455,
      "step": 7694,
      "time_per_iteration": 2.5489420890808105
    },
    {
      "auxiliary_loss_clip": 0.01177886,
      "auxiliary_loss_mlp": 0.01208117,
      "balance_loss_clip": 1.04406941,
      "balance_loss_mlp": 1.0007689,
      "epoch": 0.9252690434678049,
      "flos": 23586416052480.0,
      "grad_norm": 3.3347993907119116,
      "language_loss": 0.7117945,
      "learning_rate": 5.821678791599205e-08,
      "loss": 0.73565459,
      "num_input_tokens_seen": 166393900,
      "step": 7695,
      "time_per_iteration": 3.610821485519409
    },
    {
      "auxiliary_loss_clip": 0.01171143,
      "auxiliary_loss_mlp": 0.01023065,
      "balance_loss_clip": 1.04634476,
      "balance_loss_mlp": 1.01620126,
      "epoch": 0.9253892863584441,
      "flos": 21469158829440.0,
      "grad_norm": 1.7241885055431523,
      "language_loss": 0.80838627,
      "learning_rate": 5.803035646986965e-08,
      "loss": 0.83032835,
      "num_input_tokens_seen": 166413235,
      "step": 7696,
      "time_per_iteration": 3.6371757984161377
    },
    {
      "auxiliary_loss_clip": 0.01169891,
      "auxiliary_loss_mlp": 0.01025661,
      "balance_loss_clip": 1.04793525,
      "balance_loss_mlp": 1.01755214,
      "epoch": 0.9255095292490831,
      "flos": 17456680304640.0,
      "grad_norm": 2.166086928209733,
      "language_loss": 0.67411751,
      "learning_rate": 5.7844219618998766e-08,
      "loss": 0.696073,
      "num_input_tokens_seen": 166427560,
      "step": 7697,
      "time_per_iteration": 2.534740447998047
    },
    {
      "auxiliary_loss_clip": 0.01157951,
      "auxiliary_loss_mlp": 0.01022418,
      "balance_loss_clip": 1.03707135,
      "balance_loss_mlp": 1.01490748,
      "epoch": 0.9256297721397222,
      "flos": 24750675584640.0,
      "grad_norm": 3.116199965329018,
      "language_loss": 0.71231663,
      "learning_rate": 5.765837739161505e-08,
      "loss": 0.73412031,
      "num_input_tokens_seen": 166446680,
      "step": 7698,
      "time_per_iteration": 2.761286973953247
    },
    {
      "auxiliary_loss_clip": 0.01174716,
      "auxiliary_loss_mlp": 0.01021019,
      "balance_loss_clip": 1.04258204,
      "balance_loss_mlp": 1.01372945,
      "epoch": 0.9257500150303614,
      "flos": 23112215677440.0,
      "grad_norm": 1.6561167915861024,
      "language_loss": 0.74361491,
      "learning_rate": 5.7472829815911504e-08,
      "loss": 0.76557225,
      "num_input_tokens_seen": 166465505,
      "step": 7699,
      "time_per_iteration": 2.693824529647827
    },
    {
      "auxiliary_loss_clip": 0.01168189,
      "auxiliary_loss_mlp": 0.01030231,
      "balance_loss_clip": 1.04415607,
      "balance_loss_mlp": 1.02257216,
      "epoch": 0.9258702579210004,
      "flos": 22564685687040.0,
      "grad_norm": 1.6957489597613427,
      "language_loss": 0.81696045,
      "learning_rate": 5.7287576920035164e-08,
      "loss": 0.83894467,
      "num_input_tokens_seen": 166484520,
      "step": 7700,
      "time_per_iteration": 3.6198134422302246
    },
    {
      "auxiliary_loss_clip": 0.01171198,
      "auxiliary_loss_mlp": 0.01022511,
      "balance_loss_clip": 1.04265594,
      "balance_loss_mlp": 1.01551986,
      "epoch": 0.9259905008116395,
      "flos": 30004298703360.0,
      "grad_norm": 1.7741950761178644,
      "language_loss": 0.76693463,
      "learning_rate": 5.7102618732088435e-08,
      "loss": 0.78887171,
      "num_input_tokens_seen": 166503850,
      "step": 7701,
      "time_per_iteration": 2.7553670406341553
    },
    {
      "auxiliary_loss_clip": 0.01176953,
      "auxiliary_loss_mlp": 0.01025921,
      "balance_loss_clip": 1.04548764,
      "balance_loss_mlp": 1.01939774,
      "epoch": 0.9261107437022786,
      "flos": 24572128055040.0,
      "grad_norm": 1.7450721945817806,
      "language_loss": 0.74752629,
      "learning_rate": 5.6917955280130216e-08,
      "loss": 0.76955497,
      "num_input_tokens_seen": 166525330,
      "step": 7702,
      "time_per_iteration": 2.698840379714966
    },
    {
      "auxiliary_loss_clip": 0.01165801,
      "auxiliary_loss_mlp": 0.01025813,
      "balance_loss_clip": 1.04546356,
      "balance_loss_mlp": 1.01874375,
      "epoch": 0.9262309865929177,
      "flos": 22018448586240.0,
      "grad_norm": 2.429601343313273,
      "language_loss": 0.72443843,
      "learning_rate": 5.6733586592172755e-08,
      "loss": 0.74635458,
      "num_input_tokens_seen": 166544825,
      "step": 7703,
      "time_per_iteration": 2.6224796772003174
    },
    {
      "auxiliary_loss_clip": 0.01165227,
      "auxiliary_loss_mlp": 0.01207227,
      "balance_loss_clip": 1.0415771,
      "balance_loss_mlp": 1.00074959,
      "epoch": 0.9263512294835567,
      "flos": 20339481116160.0,
      "grad_norm": 1.9158744751024717,
      "language_loss": 0.8021785,
      "learning_rate": 5.6549512696185244e-08,
      "loss": 0.825903,
      "num_input_tokens_seen": 166563325,
      "step": 7704,
      "time_per_iteration": 2.635615825653076
    },
    {
      "auxiliary_loss_clip": 0.01164631,
      "auxiliary_loss_mlp": 0.01023535,
      "balance_loss_clip": 1.04678154,
      "balance_loss_mlp": 1.01590824,
      "epoch": 0.9264714723741959,
      "flos": 21215378263680.0,
      "grad_norm": 1.6597205789974692,
      "language_loss": 0.6832183,
      "learning_rate": 5.636573362009156e-08,
      "loss": 0.70509994,
      "num_input_tokens_seen": 166583385,
      "step": 7705,
      "time_per_iteration": 2.5769267082214355
    },
    {
      "auxiliary_loss_clip": 0.01169233,
      "auxiliary_loss_mlp": 0.01026531,
      "balance_loss_clip": 1.04822147,
      "balance_loss_mlp": 1.01898837,
      "epoch": 0.926591715264835,
      "flos": 18004964480640.0,
      "grad_norm": 4.29253166684937,
      "language_loss": 0.77236366,
      "learning_rate": 5.618224939177074e-08,
      "loss": 0.79432124,
      "num_input_tokens_seen": 166601290,
      "step": 7706,
      "time_per_iteration": 2.545243740081787
    },
    {
      "auxiliary_loss_clip": 0.01160844,
      "auxiliary_loss_mlp": 0.01023051,
      "balance_loss_clip": 1.04253578,
      "balance_loss_mlp": 1.01517737,
      "epoch": 0.926711958155474,
      "flos": 36167969825280.0,
      "grad_norm": 1.8605529695399354,
      "language_loss": 0.70090401,
      "learning_rate": 5.599906003905719e-08,
      "loss": 0.72274292,
      "num_input_tokens_seen": 166623835,
      "step": 7707,
      "time_per_iteration": 2.7166566848754883
    },
    {
      "auxiliary_loss_clip": 0.01163827,
      "auxiliary_loss_mlp": 0.01023774,
      "balance_loss_clip": 1.04669142,
      "balance_loss_mlp": 1.01623416,
      "epoch": 0.9268322010461132,
      "flos": 21032736583680.0,
      "grad_norm": 2.1149583255282973,
      "language_loss": 0.81553316,
      "learning_rate": 5.581616558974023e-08,
      "loss": 0.83740914,
      "num_input_tokens_seen": 166642400,
      "step": 7708,
      "time_per_iteration": 2.585855484008789
    },
    {
      "auxiliary_loss_clip": 0.01175667,
      "auxiliary_loss_mlp": 0.0120796,
      "balance_loss_clip": 1.04712582,
      "balance_loss_mlp": 1.000723,
      "epoch": 0.9269524439367522,
      "flos": 22964838174720.0,
      "grad_norm": 1.8792352071646812,
      "language_loss": 0.79283184,
      "learning_rate": 5.5633566071565444e-08,
      "loss": 0.81666809,
      "num_input_tokens_seen": 166661640,
      "step": 7709,
      "time_per_iteration": 2.6157402992248535
    },
    {
      "auxiliary_loss_clip": 0.01182006,
      "auxiliary_loss_mlp": 0.0102096,
      "balance_loss_clip": 1.0403744,
      "balance_loss_mlp": 1.01406336,
      "epoch": 0.9270726868273913,
      "flos": 41975551468800.0,
      "grad_norm": 1.8784556901939808,
      "language_loss": 0.70251268,
      "learning_rate": 5.5451261512232896e-08,
      "loss": 0.72454232,
      "num_input_tokens_seen": 166684320,
      "step": 7710,
      "time_per_iteration": 2.838758945465088
    },
    {
      "auxiliary_loss_clip": 0.01172822,
      "auxiliary_loss_mlp": 0.01024488,
      "balance_loss_clip": 1.04359436,
      "balance_loss_mlp": 1.01687074,
      "epoch": 0.9271929297180305,
      "flos": 19791771557760.0,
      "grad_norm": 2.827545963390506,
      "language_loss": 0.62852371,
      "learning_rate": 5.5269251939397576e-08,
      "loss": 0.65049684,
      "num_input_tokens_seen": 166703835,
      "step": 7711,
      "time_per_iteration": 3.5643246173858643
    },
    {
      "auxiliary_loss_clip": 0.01173339,
      "auxiliary_loss_mlp": 0.01020681,
      "balance_loss_clip": 1.03989017,
      "balance_loss_mlp": 1.01315308,
      "epoch": 0.9273131726086695,
      "flos": 19968343839360.0,
      "grad_norm": 1.9372527651699756,
      "language_loss": 0.76968753,
      "learning_rate": 5.508753738067073e-08,
      "loss": 0.79162776,
      "num_input_tokens_seen": 166723375,
      "step": 7712,
      "time_per_iteration": 2.6401865482330322
    },
    {
      "auxiliary_loss_clip": 0.01169676,
      "auxiliary_loss_mlp": 0.01023695,
      "balance_loss_clip": 1.04294157,
      "balance_loss_mlp": 1.01607788,
      "epoch": 0.9274334154993086,
      "flos": 23258587599360.0,
      "grad_norm": 3.6499432686804076,
      "language_loss": 0.79447031,
      "learning_rate": 5.4906117863617875e-08,
      "loss": 0.81640399,
      "num_input_tokens_seen": 166742760,
      "step": 7713,
      "time_per_iteration": 2.550313711166382
    },
    {
      "auxiliary_loss_clip": 0.01170016,
      "auxiliary_loss_mlp": 0.01020289,
      "balance_loss_clip": 1.04009402,
      "balance_loss_mlp": 1.01317489,
      "epoch": 0.9275536583899477,
      "flos": 31795343585280.0,
      "grad_norm": 1.9479215348493333,
      "language_loss": 0.77792346,
      "learning_rate": 5.4724993415760533e-08,
      "loss": 0.79982644,
      "num_input_tokens_seen": 166761115,
      "step": 7714,
      "time_per_iteration": 2.718170642852783
    },
    {
      "auxiliary_loss_clip": 0.01182572,
      "auxiliary_loss_mlp": 0.01208277,
      "balance_loss_clip": 1.04251981,
      "balance_loss_mlp": 1.00073147,
      "epoch": 0.9276739012805868,
      "flos": 18696998885760.0,
      "grad_norm": 2.167295946505079,
      "language_loss": 0.74985552,
      "learning_rate": 5.454416406457496e-08,
      "loss": 0.77376401,
      "num_input_tokens_seen": 166780210,
      "step": 7715,
      "time_per_iteration": 2.612879753112793
    },
    {
      "auxiliary_loss_clip": 0.01168455,
      "auxiliary_loss_mlp": 0.0102326,
      "balance_loss_clip": 1.04513597,
      "balance_loss_mlp": 1.01659298,
      "epoch": 0.9277941441712259,
      "flos": 13879079740800.0,
      "grad_norm": 2.326993547569469,
      "language_loss": 0.7383818,
      "learning_rate": 5.436362983749299e-08,
      "loss": 0.76029891,
      "num_input_tokens_seen": 166795380,
      "step": 7716,
      "time_per_iteration": 2.5333893299102783
    },
    {
      "auxiliary_loss_clip": 0.01166965,
      "auxiliary_loss_mlp": 0.01024985,
      "balance_loss_clip": 1.04391205,
      "balance_loss_mlp": 1.0183239,
      "epoch": 0.927914387061865,
      "flos": 23258659426560.0,
      "grad_norm": 2.1074550935325895,
      "language_loss": 0.64351809,
      "learning_rate": 5.418339076190137e-08,
      "loss": 0.66543758,
      "num_input_tokens_seen": 166814890,
      "step": 7717,
      "time_per_iteration": 2.7371532917022705
    },
    {
      "auxiliary_loss_clip": 0.01165012,
      "auxiliary_loss_mlp": 0.01018721,
      "balance_loss_clip": 1.04417896,
      "balance_loss_mlp": 1.01130605,
      "epoch": 0.9280346299525041,
      "flos": 18073733068800.0,
      "grad_norm": 3.001796784289003,
      "language_loss": 0.88754201,
      "learning_rate": 5.400344686514202e-08,
      "loss": 0.90937936,
      "num_input_tokens_seen": 166832475,
      "step": 7718,
      "time_per_iteration": 2.614698886871338
    },
    {
      "auxiliary_loss_clip": 0.01168596,
      "auxiliary_loss_mlp": 0.01022199,
      "balance_loss_clip": 1.04745889,
      "balance_loss_mlp": 1.01489198,
      "epoch": 0.9281548728431431,
      "flos": 22342901160960.0,
      "grad_norm": 2.0655460003671275,
      "language_loss": 0.66611671,
      "learning_rate": 5.38237981745131e-08,
      "loss": 0.6880247,
      "num_input_tokens_seen": 166850590,
      "step": 7719,
      "time_per_iteration": 2.550718069076538
    },
    {
      "auxiliary_loss_clip": 0.01172397,
      "auxiliary_loss_mlp": 0.01207943,
      "balance_loss_clip": 1.04666781,
      "balance_loss_mlp": 1.00070906,
      "epoch": 0.9282751157337822,
      "flos": 18843765857280.0,
      "grad_norm": 1.7238211953693579,
      "language_loss": 0.81665123,
      "learning_rate": 5.364444471726592e-08,
      "loss": 0.84045464,
      "num_input_tokens_seen": 166869795,
      "step": 7720,
      "time_per_iteration": 2.581981897354126
    },
    {
      "auxiliary_loss_clip": 0.01167864,
      "auxiliary_loss_mlp": 0.01021709,
      "balance_loss_clip": 1.04448783,
      "balance_loss_mlp": 1.0145061,
      "epoch": 0.9283953586244214,
      "flos": 25556834476800.0,
      "grad_norm": 2.4787196923548245,
      "language_loss": 0.80205512,
      "learning_rate": 5.346538652060939e-08,
      "loss": 0.82395089,
      "num_input_tokens_seen": 166891150,
      "step": 7721,
      "time_per_iteration": 3.553283929824829
    },
    {
      "auxiliary_loss_clip": 0.01166862,
      "auxiliary_loss_mlp": 0.01021103,
      "balance_loss_clip": 1.04426157,
      "balance_loss_mlp": 1.01413512,
      "epoch": 0.9285156015150604,
      "flos": 18223480869120.0,
      "grad_norm": 1.7935265684301298,
      "language_loss": 0.70364904,
      "learning_rate": 5.3286623611705994e-08,
      "loss": 0.72552866,
      "num_input_tokens_seen": 166909195,
      "step": 7722,
      "time_per_iteration": 2.5674288272857666
    },
    {
      "auxiliary_loss_clip": 0.01062898,
      "auxiliary_loss_mlp": 0.01001118,
      "balance_loss_clip": 1.01037216,
      "balance_loss_mlp": 1.00039387,
      "epoch": 0.9286358444056995,
      "flos": 66400017690240.0,
      "grad_norm": 0.805791879395036,
      "language_loss": 0.6055181,
      "learning_rate": 5.3108156017673824e-08,
      "loss": 0.62615836,
      "num_input_tokens_seen": 166970955,
      "step": 7723,
      "time_per_iteration": 4.152153730392456
    },
    {
      "auxiliary_loss_clip": 0.01179202,
      "auxiliary_loss_mlp": 0.01024915,
      "balance_loss_clip": 1.04586148,
      "balance_loss_mlp": 1.01688647,
      "epoch": 0.9287560872963386,
      "flos": 22345630594560.0,
      "grad_norm": 1.775263974111617,
      "language_loss": 0.71600562,
      "learning_rate": 5.2929983765586775e-08,
      "loss": 0.73804677,
      "num_input_tokens_seen": 166989735,
      "step": 7724,
      "time_per_iteration": 2.661404609680176
    },
    {
      "auxiliary_loss_clip": 0.01169194,
      "auxiliary_loss_mlp": 0.0102708,
      "balance_loss_clip": 1.05047214,
      "balance_loss_mlp": 1.0200882,
      "epoch": 0.9288763301869777,
      "flos": 25700225569920.0,
      "grad_norm": 1.83035986369728,
      "language_loss": 0.62779868,
      "learning_rate": 5.275210688247278e-08,
      "loss": 0.64976144,
      "num_input_tokens_seen": 167010060,
      "step": 7725,
      "time_per_iteration": 2.6011829376220703
    },
    {
      "auxiliary_loss_clip": 0.01176347,
      "auxiliary_loss_mlp": 0.01024146,
      "balance_loss_clip": 1.04236007,
      "balance_loss_mlp": 1.01691318,
      "epoch": 0.9289965730776167,
      "flos": 12312046028160.0,
      "grad_norm": 2.0360574468749757,
      "language_loss": 0.84964716,
      "learning_rate": 5.257452539531604e-08,
      "loss": 0.87165213,
      "num_input_tokens_seen": 167027130,
      "step": 7726,
      "time_per_iteration": 2.6453559398651123
    },
    {
      "auxiliary_loss_clip": 0.01168997,
      "auxiliary_loss_mlp": 0.01026575,
      "balance_loss_clip": 1.04437101,
      "balance_loss_mlp": 1.01903534,
      "epoch": 0.9291168159682559,
      "flos": 26685973486080.0,
      "grad_norm": 1.5073041283735422,
      "language_loss": 0.68436241,
      "learning_rate": 5.2397239331055445e-08,
      "loss": 0.70631814,
      "num_input_tokens_seen": 167049130,
      "step": 7727,
      "time_per_iteration": 3.579909324645996
    },
    {
      "auxiliary_loss_clip": 0.01170005,
      "auxiliary_loss_mlp": 0.01021857,
      "balance_loss_clip": 1.04549325,
      "balance_loss_mlp": 1.0143764,
      "epoch": 0.929237058858895,
      "flos": 14538256179840.0,
      "grad_norm": 2.4104475441950894,
      "language_loss": 0.81378096,
      "learning_rate": 5.2220248716585036e-08,
      "loss": 0.83569956,
      "num_input_tokens_seen": 167066810,
      "step": 7728,
      "time_per_iteration": 2.620479106903076
    },
    {
      "auxiliary_loss_clip": 0.01160909,
      "auxiliary_loss_mlp": 0.01027309,
      "balance_loss_clip": 1.04351234,
      "balance_loss_mlp": 1.01976275,
      "epoch": 0.929357301749534,
      "flos": 23835456023040.0,
      "grad_norm": 2.41544411203423,
      "language_loss": 0.75744247,
      "learning_rate": 5.204355357875445e-08,
      "loss": 0.77932465,
      "num_input_tokens_seen": 167085155,
      "step": 7729,
      "time_per_iteration": 2.668595790863037
    },
    {
      "auxiliary_loss_clip": 0.01168227,
      "auxiliary_loss_mlp": 0.01019976,
      "balance_loss_clip": 1.04185116,
      "balance_loss_mlp": 1.01261497,
      "epoch": 0.9294775446401732,
      "flos": 12969319046400.0,
      "grad_norm": 2.295668423921764,
      "language_loss": 0.70271474,
      "learning_rate": 5.1867153944367584e-08,
      "loss": 0.7245968,
      "num_input_tokens_seen": 167101545,
      "step": 7730,
      "time_per_iteration": 2.6462981700897217
    },
    {
      "auxiliary_loss_clip": 0.01180145,
      "auxiliary_loss_mlp": 0.01030137,
      "balance_loss_clip": 1.04472518,
      "balance_loss_mlp": 1.0229671,
      "epoch": 0.9295977875308122,
      "flos": 26211809024640.0,
      "grad_norm": 1.5402337913404183,
      "language_loss": 0.73656273,
      "learning_rate": 5.16910498401848e-08,
      "loss": 0.75866556,
      "num_input_tokens_seen": 167120995,
      "step": 7731,
      "time_per_iteration": 2.658599853515625
    },
    {
      "auxiliary_loss_clip": 0.0116469,
      "auxiliary_loss_mlp": 0.01026425,
      "balance_loss_clip": 1.0476433,
      "balance_loss_mlp": 1.01945758,
      "epoch": 0.9297180304214513,
      "flos": 16472297105280.0,
      "grad_norm": 2.82127952409276,
      "language_loss": 0.83384657,
      "learning_rate": 5.151524129292073e-08,
      "loss": 0.85575777,
      "num_input_tokens_seen": 167138890,
      "step": 7732,
      "time_per_iteration": 2.592879295349121
    },
    {
      "auxiliary_loss_clip": 0.01167434,
      "auxiliary_loss_mlp": 0.01025311,
      "balance_loss_clip": 1.04552746,
      "balance_loss_mlp": 1.01841521,
      "epoch": 0.9298382733120905,
      "flos": 24060436859520.0,
      "grad_norm": 1.947714031083173,
      "language_loss": 0.66858512,
      "learning_rate": 5.1339728329245155e-08,
      "loss": 0.69051254,
      "num_input_tokens_seen": 167159455,
      "step": 7733,
      "time_per_iteration": 2.637737989425659
    },
    {
      "auxiliary_loss_clip": 0.01171263,
      "auxiliary_loss_mlp": 0.01026458,
      "balance_loss_clip": 1.04787385,
      "balance_loss_mlp": 1.0184741,
      "epoch": 0.9299585162027295,
      "flos": 22127652910080.0,
      "grad_norm": 2.210154571246404,
      "language_loss": 0.79404491,
      "learning_rate": 5.116451097578367e-08,
      "loss": 0.8160221,
      "num_input_tokens_seen": 167178495,
      "step": 7734,
      "time_per_iteration": 2.636253595352173
    },
    {
      "auxiliary_loss_clip": 0.01172227,
      "auxiliary_loss_mlp": 0.01028695,
      "balance_loss_clip": 1.04215717,
      "balance_loss_mlp": 1.02142358,
      "epoch": 0.9300787590933686,
      "flos": 21471780522240.0,
      "grad_norm": 1.71939136230453,
      "language_loss": 0.74204516,
      "learning_rate": 5.0989589259115895e-08,
      "loss": 0.76405442,
      "num_input_tokens_seen": 167199380,
      "step": 7735,
      "time_per_iteration": 2.686676502227783
    },
    {
      "auxiliary_loss_clip": 0.01166288,
      "auxiliary_loss_mlp": 0.01024232,
      "balance_loss_clip": 1.0429287,
      "balance_loss_mlp": 1.01591086,
      "epoch": 0.9301990019840077,
      "flos": 17779588594560.0,
      "grad_norm": 1.9975577837956697,
      "language_loss": 0.71862656,
      "learning_rate": 5.081496320577816e-08,
      "loss": 0.74053174,
      "num_input_tokens_seen": 167216500,
      "step": 7736,
      "time_per_iteration": 2.664896011352539
    },
    {
      "auxiliary_loss_clip": 0.01071493,
      "auxiliary_loss_mlp": 0.01000061,
      "balance_loss_clip": 1.01210403,
      "balance_loss_mlp": 0.99926847,
      "epoch": 0.9303192448746468,
      "flos": 58896122307840.0,
      "grad_norm": 0.9113354632358505,
      "language_loss": 0.61196232,
      "learning_rate": 5.0640632842260835e-08,
      "loss": 0.63267785,
      "num_input_tokens_seen": 167276760,
      "step": 7737,
      "time_per_iteration": 3.2347254753112793
    },
    {
      "auxiliary_loss_clip": 0.01172025,
      "auxiliary_loss_mlp": 0.01207725,
      "balance_loss_clip": 1.04549742,
      "balance_loss_mlp": 1.00074005,
      "epoch": 0.9304394877652858,
      "flos": 57663522172800.0,
      "grad_norm": 1.517791036862482,
      "language_loss": 0.72435749,
      "learning_rate": 5.0466598195009426e-08,
      "loss": 0.748155,
      "num_input_tokens_seen": 167303630,
      "step": 7738,
      "time_per_iteration": 3.990811824798584
    },
    {
      "auxiliary_loss_clip": 0.01174383,
      "auxiliary_loss_mlp": 0.01026235,
      "balance_loss_clip": 1.04324365,
      "balance_loss_mlp": 1.01871276,
      "epoch": 0.930559730655925,
      "flos": 20996143603200.0,
      "grad_norm": 2.117200466313698,
      "language_loss": 0.69991726,
      "learning_rate": 5.0292859290425036e-08,
      "loss": 0.72192341,
      "num_input_tokens_seen": 167321500,
      "step": 7739,
      "time_per_iteration": 2.6340134143829346
    },
    {
      "auxiliary_loss_clip": 0.01165969,
      "auxiliary_loss_mlp": 0.01021751,
      "balance_loss_clip": 1.04809844,
      "balance_loss_mlp": 1.0150156,
      "epoch": 0.9306799735465641,
      "flos": 23258264376960.0,
      "grad_norm": 2.246678248024724,
      "language_loss": 0.77748275,
      "learning_rate": 5.011941615486348e-08,
      "loss": 0.79935992,
      "num_input_tokens_seen": 167340615,
      "step": 7740,
      "time_per_iteration": 2.72385311126709
    },
    {
      "auxiliary_loss_clip": 0.01165302,
      "auxiliary_loss_mlp": 0.01020359,
      "balance_loss_clip": 1.04603386,
      "balance_loss_mlp": 1.01318216,
      "epoch": 0.9308002164372031,
      "flos": 15231547560960.0,
      "grad_norm": 2.209254975073699,
      "language_loss": 0.8462109,
      "learning_rate": 4.994626881463659e-08,
      "loss": 0.8680675,
      "num_input_tokens_seen": 167356870,
      "step": 7741,
      "time_per_iteration": 2.588641881942749
    },
    {
      "auxiliary_loss_clip": 0.01161499,
      "auxiliary_loss_mlp": 0.01024012,
      "balance_loss_clip": 1.03893137,
      "balance_loss_mlp": 1.01672816,
      "epoch": 0.9309204593278423,
      "flos": 30847481539200.0,
      "grad_norm": 1.6188151073131725,
      "language_loss": 0.70755619,
      "learning_rate": 4.9773417296009814e-08,
      "loss": 0.72941136,
      "num_input_tokens_seen": 167378390,
      "step": 7742,
      "time_per_iteration": 2.788602828979492
    },
    {
      "auxiliary_loss_clip": 0.01175896,
      "auxiliary_loss_mlp": 0.01024916,
      "balance_loss_clip": 1.0476613,
      "balance_loss_mlp": 1.01746559,
      "epoch": 0.9310407022184813,
      "flos": 23037269950080.0,
      "grad_norm": 1.83396967090144,
      "language_loss": 0.6560818,
      "learning_rate": 4.960086162520527e-08,
      "loss": 0.67808998,
      "num_input_tokens_seen": 167398480,
      "step": 7743,
      "time_per_iteration": 2.6600053310394287
    },
    {
      "auxiliary_loss_clip": 0.01182172,
      "auxiliary_loss_mlp": 0.01022665,
      "balance_loss_clip": 1.04166174,
      "balance_loss_mlp": 1.01572752,
      "epoch": 0.9311609451091204,
      "flos": 22127976132480.0,
      "grad_norm": 2.990648344255757,
      "language_loss": 0.82532144,
      "learning_rate": 4.942860182839936e-08,
      "loss": 0.84736979,
      "num_input_tokens_seen": 167416825,
      "step": 7744,
      "time_per_iteration": 2.79996395111084
    },
    {
      "auxiliary_loss_clip": 0.01170542,
      "auxiliary_loss_mlp": 0.01023211,
      "balance_loss_clip": 1.04524946,
      "balance_loss_mlp": 1.01577282,
      "epoch": 0.9312811879997596,
      "flos": 21099206701440.0,
      "grad_norm": 2.8160117420666158,
      "language_loss": 0.79721367,
      "learning_rate": 4.925663793172341e-08,
      "loss": 0.81915122,
      "num_input_tokens_seen": 167434785,
      "step": 7745,
      "time_per_iteration": 2.739469051361084
    },
    {
      "auxiliary_loss_clip": 0.01071264,
      "auxiliary_loss_mlp": 0.01198569,
      "balance_loss_clip": 1.01055205,
      "balance_loss_mlp": 0.99981385,
      "epoch": 0.9314014308903986,
      "flos": 67148179096320.0,
      "grad_norm": 0.7868217349317467,
      "language_loss": 0.56494749,
      "learning_rate": 4.908496996126477e-08,
      "loss": 0.58764583,
      "num_input_tokens_seen": 167498245,
      "step": 7746,
      "time_per_iteration": 3.2487237453460693
    },
    {
      "auxiliary_loss_clip": 0.01168604,
      "auxiliary_loss_mlp": 0.01024028,
      "balance_loss_clip": 1.04842782,
      "balance_loss_mlp": 1.01646757,
      "epoch": 0.9315216737810377,
      "flos": 22565583527040.0,
      "grad_norm": 1.6772536042728758,
      "language_loss": 0.76471651,
      "learning_rate": 4.89135979430646e-08,
      "loss": 0.78664291,
      "num_input_tokens_seen": 167518290,
      "step": 7747,
      "time_per_iteration": 2.6228420734405518
    },
    {
      "auxiliary_loss_clip": 0.01167941,
      "auxiliary_loss_mlp": 0.01022474,
      "balance_loss_clip": 1.04978383,
      "balance_loss_mlp": 1.01468706,
      "epoch": 0.9316419166716768,
      "flos": 23984054588160.0,
      "grad_norm": 1.7082067660571487,
      "language_loss": 0.85522151,
      "learning_rate": 4.874252190312078e-08,
      "loss": 0.87712562,
      "num_input_tokens_seen": 167538675,
      "step": 7748,
      "time_per_iteration": 3.5264809131622314
    },
    {
      "auxiliary_loss_clip": 0.01171774,
      "auxiliary_loss_mlp": 0.01021685,
      "balance_loss_clip": 1.04537439,
      "balance_loss_mlp": 1.01439571,
      "epoch": 0.9317621595623159,
      "flos": 30230464688640.0,
      "grad_norm": 2.289529634354843,
      "language_loss": 0.64846253,
      "learning_rate": 4.857174186738477e-08,
      "loss": 0.67039716,
      "num_input_tokens_seen": 167562025,
      "step": 7749,
      "time_per_iteration": 2.7059857845306396
    },
    {
      "auxiliary_loss_clip": 0.01168972,
      "auxiliary_loss_mlp": 0.01024757,
      "balance_loss_clip": 1.04948032,
      "balance_loss_mlp": 1.01715732,
      "epoch": 0.931882402452955,
      "flos": 15742735966080.0,
      "grad_norm": 2.980699051392454,
      "language_loss": 0.73697269,
      "learning_rate": 4.840125786176408e-08,
      "loss": 0.75890994,
      "num_input_tokens_seen": 167578230,
      "step": 7750,
      "time_per_iteration": 2.567718029022217
    },
    {
      "auxiliary_loss_clip": 0.01168076,
      "auxiliary_loss_mlp": 0.01022565,
      "balance_loss_clip": 1.04372334,
      "balance_loss_mlp": 1.01543307,
      "epoch": 0.932002645343594,
      "flos": 28366521154560.0,
      "grad_norm": 1.9480228537867232,
      "language_loss": 0.77308619,
      "learning_rate": 4.823106991212067e-08,
      "loss": 0.79499263,
      "num_input_tokens_seen": 167597470,
      "step": 7751,
      "time_per_iteration": 3.5826258659362793
    },
    {
      "auxiliary_loss_clip": 0.01170742,
      "auxiliary_loss_mlp": 0.01021857,
      "balance_loss_clip": 1.04612768,
      "balance_loss_mlp": 1.01445436,
      "epoch": 0.9321228882342332,
      "flos": 15341146934400.0,
      "grad_norm": 2.5417796847426746,
      "language_loss": 0.83445036,
      "learning_rate": 4.806117804427212e-08,
      "loss": 0.85637641,
      "num_input_tokens_seen": 167615405,
      "step": 7752,
      "time_per_iteration": 2.607360601425171
    },
    {
      "auxiliary_loss_clip": 0.01163169,
      "auxiliary_loss_mlp": 0.01026579,
      "balance_loss_clip": 1.04312563,
      "balance_loss_mlp": 1.01869917,
      "epoch": 0.9322431311248722,
      "flos": 17895365107200.0,
      "grad_norm": 2.633136419098402,
      "language_loss": 0.64366865,
      "learning_rate": 4.7891582283990926e-08,
      "loss": 0.66556609,
      "num_input_tokens_seen": 167634130,
      "step": 7753,
      "time_per_iteration": 2.5691888332366943
    },
    {
      "auxiliary_loss_clip": 0.01173923,
      "auxiliary_loss_mlp": 0.01020495,
      "balance_loss_clip": 1.04241729,
      "balance_loss_mlp": 1.01348853,
      "epoch": 0.9323633740155113,
      "flos": 24169713010560.0,
      "grad_norm": 1.874137196992048,
      "language_loss": 0.72854465,
      "learning_rate": 4.772228265700473e-08,
      "loss": 0.75048888,
      "num_input_tokens_seen": 167654990,
      "step": 7754,
      "time_per_iteration": 3.708427667617798
    },
    {
      "auxiliary_loss_clip": 0.01171562,
      "auxiliary_loss_mlp": 0.01023378,
      "balance_loss_clip": 1.0457468,
      "balance_loss_mlp": 1.01590395,
      "epoch": 0.9324836169061504,
      "flos": 15043482927360.0,
      "grad_norm": 2.8002894299040273,
      "language_loss": 0.75704718,
      "learning_rate": 4.75532791889961e-08,
      "loss": 0.77899659,
      "num_input_tokens_seen": 167671690,
      "step": 7755,
      "time_per_iteration": 2.788161277770996
    },
    {
      "auxiliary_loss_clip": 0.01167386,
      "auxiliary_loss_mlp": 0.01026452,
      "balance_loss_clip": 1.04374337,
      "balance_loss_mlp": 1.0187515,
      "epoch": 0.9326038597967895,
      "flos": 18624890332800.0,
      "grad_norm": 2.2071883756530313,
      "language_loss": 0.65694034,
      "learning_rate": 4.738457190560252e-08,
      "loss": 0.67887878,
      "num_input_tokens_seen": 167690800,
      "step": 7756,
      "time_per_iteration": 2.6832222938537598
    },
    {
      "auxiliary_loss_clip": 0.01179254,
      "auxiliary_loss_mlp": 0.01024381,
      "balance_loss_clip": 1.04490042,
      "balance_loss_mlp": 1.01699567,
      "epoch": 0.9327241026874286,
      "flos": 18952646958720.0,
      "grad_norm": 2.1834803373063543,
      "language_loss": 0.79357231,
      "learning_rate": 4.721616083241664e-08,
      "loss": 0.81560862,
      "num_input_tokens_seen": 167709055,
      "step": 7757,
      "time_per_iteration": 2.6334311962127686
    },
    {
      "auxiliary_loss_clip": 0.01165951,
      "auxiliary_loss_mlp": 0.01025013,
      "balance_loss_clip": 1.04609895,
      "balance_loss_mlp": 1.01794374,
      "epoch": 0.9328443455780677,
      "flos": 29570282668800.0,
      "grad_norm": 1.7402456755357558,
      "language_loss": 0.77461994,
      "learning_rate": 4.7048045994986684e-08,
      "loss": 0.79652953,
      "num_input_tokens_seen": 167729915,
      "step": 7758,
      "time_per_iteration": 2.7196578979492188
    },
    {
      "auxiliary_loss_clip": 0.01175714,
      "auxiliary_loss_mlp": 0.01023301,
      "balance_loss_clip": 1.04817462,
      "balance_loss_mlp": 1.01590395,
      "epoch": 0.9329645884687068,
      "flos": 30081722469120.0,
      "grad_norm": 2.589077839308281,
      "language_loss": 0.91259265,
      "learning_rate": 4.688022741881559e-08,
      "loss": 0.93458283,
      "num_input_tokens_seen": 167750440,
      "step": 7759,
      "time_per_iteration": 2.6216530799865723
    },
    {
      "auxiliary_loss_clip": 0.01165833,
      "auxiliary_loss_mlp": 0.0102334,
      "balance_loss_clip": 1.04434276,
      "balance_loss_mlp": 1.01649427,
      "epoch": 0.9330848313593458,
      "flos": 21867982513920.0,
      "grad_norm": 1.7833987345851623,
      "language_loss": 0.75100768,
      "learning_rate": 4.671270512936076e-08,
      "loss": 0.77289939,
      "num_input_tokens_seen": 167769600,
      "step": 7760,
      "time_per_iteration": 2.596027135848999
    },
    {
      "auxiliary_loss_clip": 0.01168423,
      "auxiliary_loss_mlp": 0.01022914,
      "balance_loss_clip": 1.04209673,
      "balance_loss_mlp": 1.01564252,
      "epoch": 0.933205074249985,
      "flos": 22127221946880.0,
      "grad_norm": 1.8716149430901057,
      "language_loss": 0.83054119,
      "learning_rate": 4.6545479152035884e-08,
      "loss": 0.85245454,
      "num_input_tokens_seen": 167788770,
      "step": 7761,
      "time_per_iteration": 2.649724245071411
    },
    {
      "auxiliary_loss_clip": 0.01169035,
      "auxiliary_loss_mlp": 0.0102379,
      "balance_loss_clip": 1.04592228,
      "balance_loss_mlp": 1.01668477,
      "epoch": 0.9333253171406241,
      "flos": 15341254675200.0,
      "grad_norm": 1.9535403534546876,
      "language_loss": 0.76277399,
      "learning_rate": 4.637854951220821e-08,
      "loss": 0.7847023,
      "num_input_tokens_seen": 167805555,
      "step": 7762,
      "time_per_iteration": 2.5831663608551025
    },
    {
      "auxiliary_loss_clip": 0.01167845,
      "auxiliary_loss_mlp": 0.01022863,
      "balance_loss_clip": 1.04099119,
      "balance_loss_mlp": 1.01585364,
      "epoch": 0.9334455600312631,
      "flos": 15706142985600.0,
      "grad_norm": 1.8870651063561856,
      "language_loss": 0.74924397,
      "learning_rate": 4.621191623520171e-08,
      "loss": 0.77115107,
      "num_input_tokens_seen": 167823985,
      "step": 7763,
      "time_per_iteration": 2.6480751037597656
    },
    {
      "auxiliary_loss_clip": 0.01192775,
      "auxiliary_loss_mlp": 0.01023982,
      "balance_loss_clip": 1.04271722,
      "balance_loss_mlp": 1.01650786,
      "epoch": 0.9335658029219023,
      "flos": 22163563532160.0,
      "grad_norm": 2.3633132304305113,
      "language_loss": 0.84867859,
      "learning_rate": 4.604557934629372e-08,
      "loss": 0.87084615,
      "num_input_tokens_seen": 167843060,
      "step": 7764,
      "time_per_iteration": 3.684514045715332
    },
    {
      "auxiliary_loss_clip": 0.01168499,
      "auxiliary_loss_mlp": 0.01019892,
      "balance_loss_clip": 1.04606795,
      "balance_loss_mlp": 1.01296318,
      "epoch": 0.9336860458125413,
      "flos": 20266833859200.0,
      "grad_norm": 2.0288681476221875,
      "language_loss": 0.80612814,
      "learning_rate": 4.587953887071805e-08,
      "loss": 0.82801205,
      "num_input_tokens_seen": 167862880,
      "step": 7765,
      "time_per_iteration": 2.6706349849700928
    },
    {
      "auxiliary_loss_clip": 0.0116726,
      "auxiliary_loss_mlp": 0.01025158,
      "balance_loss_clip": 1.04101396,
      "balance_loss_mlp": 1.01758528,
      "epoch": 0.9338062887031804,
      "flos": 20919689504640.0,
      "grad_norm": 1.7909673343395571,
      "language_loss": 0.86152363,
      "learning_rate": 4.5713794833662554e-08,
      "loss": 0.88344783,
      "num_input_tokens_seen": 167882095,
      "step": 7766,
      "time_per_iteration": 2.6713736057281494
    },
    {
      "auxiliary_loss_clip": 0.01168227,
      "auxiliary_loss_mlp": 0.01025831,
      "balance_loss_clip": 1.04812121,
      "balance_loss_mlp": 1.01796353,
      "epoch": 0.9339265315938196,
      "flos": 23221635482880.0,
      "grad_norm": 1.9769753391179934,
      "language_loss": 0.63385409,
      "learning_rate": 4.5548347260270236e-08,
      "loss": 0.65579474,
      "num_input_tokens_seen": 167901385,
      "step": 7767,
      "time_per_iteration": 2.6541929244995117
    },
    {
      "auxiliary_loss_clip": 0.01169286,
      "auxiliary_loss_mlp": 0.01022827,
      "balance_loss_clip": 1.04199302,
      "balance_loss_mlp": 1.01588583,
      "epoch": 0.9340467744844586,
      "flos": 22820261932800.0,
      "grad_norm": 1.9677656500085474,
      "language_loss": 0.6966033,
      "learning_rate": 4.538319617564012e-08,
      "loss": 0.71852446,
      "num_input_tokens_seen": 167920405,
      "step": 7768,
      "time_per_iteration": 2.6758077144622803
    },
    {
      "auxiliary_loss_clip": 0.01169636,
      "auxiliary_loss_mlp": 0.0102298,
      "balance_loss_clip": 1.04219699,
      "balance_loss_mlp": 1.01556516,
      "epoch": 0.9341670173750977,
      "flos": 23660428026240.0,
      "grad_norm": 2.1031528375096444,
      "language_loss": 0.74627531,
      "learning_rate": 4.521834160482485e-08,
      "loss": 0.76820147,
      "num_input_tokens_seen": 167939145,
      "step": 7769,
      "time_per_iteration": 2.765845537185669
    },
    {
      "auxiliary_loss_clip": 0.01170312,
      "auxiliary_loss_mlp": 0.01028226,
      "balance_loss_clip": 1.04563725,
      "balance_loss_mlp": 1.02078772,
      "epoch": 0.9342872602657368,
      "flos": 24824256595200.0,
      "grad_norm": 1.7327525521146343,
      "language_loss": 0.82059282,
      "learning_rate": 4.5053783572832846e-08,
      "loss": 0.84257817,
      "num_input_tokens_seen": 167959325,
      "step": 7770,
      "time_per_iteration": 2.780447006225586
    },
    {
      "auxiliary_loss_clip": 0.01170194,
      "auxiliary_loss_mlp": 0.01027025,
      "balance_loss_clip": 1.04669476,
      "balance_loss_mlp": 1.01988173,
      "epoch": 0.9344075031563759,
      "flos": 25771831332480.0,
      "grad_norm": 2.214266231405263,
      "language_loss": 0.76606011,
      "learning_rate": 4.488952210462771e-08,
      "loss": 0.78803229,
      "num_input_tokens_seen": 167979530,
      "step": 7771,
      "time_per_iteration": 2.6557626724243164
    },
    {
      "auxiliary_loss_clip": 0.01163858,
      "auxiliary_loss_mlp": 0.01024613,
      "balance_loss_clip": 1.04663599,
      "balance_loss_mlp": 1.01759732,
      "epoch": 0.9345277460470149,
      "flos": 25551303782400.0,
      "grad_norm": 3.1213308894037226,
      "language_loss": 0.86136997,
      "learning_rate": 4.4725557225127495e-08,
      "loss": 0.88325465,
      "num_input_tokens_seen": 167997870,
      "step": 7772,
      "time_per_iteration": 2.6311450004577637
    },
    {
      "auxiliary_loss_clip": 0.01168528,
      "auxiliary_loss_mlp": 0.01028,
      "balance_loss_clip": 1.04698229,
      "balance_loss_mlp": 1.02146745,
      "epoch": 0.9346479889376541,
      "flos": 34313112432000.0,
      "grad_norm": 1.9407408137175757,
      "language_loss": 0.79334295,
      "learning_rate": 4.456188895920565e-08,
      "loss": 0.81530821,
      "num_input_tokens_seen": 168019625,
      "step": 7773,
      "time_per_iteration": 2.7449324131011963
    },
    {
      "auxiliary_loss_clip": 0.01168293,
      "auxiliary_loss_mlp": 0.01023923,
      "balance_loss_clip": 1.04874492,
      "balance_loss_mlp": 1.01651454,
      "epoch": 0.9347682318282932,
      "flos": 19093739581440.0,
      "grad_norm": 7.494462929241311,
      "language_loss": 0.85708356,
      "learning_rate": 4.439851733169031e-08,
      "loss": 0.87900573,
      "num_input_tokens_seen": 168037415,
      "step": 7774,
      "time_per_iteration": 3.533856153488159
    },
    {
      "auxiliary_loss_clip": 0.01174593,
      "auxiliary_loss_mlp": 0.01026334,
      "balance_loss_clip": 1.04225731,
      "balance_loss_mlp": 1.01926184,
      "epoch": 0.9348884747189322,
      "flos": 26249587153920.0,
      "grad_norm": 4.266102519992353,
      "language_loss": 0.69901121,
      "learning_rate": 4.4235442367365204e-08,
      "loss": 0.72102046,
      "num_input_tokens_seen": 168057725,
      "step": 7775,
      "time_per_iteration": 2.6607680320739746
    },
    {
      "auxiliary_loss_clip": 0.0116632,
      "auxiliary_loss_mlp": 0.0102469,
      "balance_loss_clip": 1.04102135,
      "balance_loss_mlp": 1.01694727,
      "epoch": 0.9350087176095714,
      "flos": 18333080242560.0,
      "grad_norm": 2.658224047242209,
      "language_loss": 0.79594386,
      "learning_rate": 4.4072664090968545e-08,
      "loss": 0.81785393,
      "num_input_tokens_seen": 168076110,
      "step": 7776,
      "time_per_iteration": 2.647108793258667
    },
    {
      "auxiliary_loss_clip": 0.01171982,
      "auxiliary_loss_mlp": 0.01027157,
      "balance_loss_clip": 1.04204369,
      "balance_loss_mlp": 1.01992106,
      "epoch": 0.9351289605002104,
      "flos": 19318253541120.0,
      "grad_norm": 1.9505517082452228,
      "language_loss": 0.84327191,
      "learning_rate": 4.391018252719347e-08,
      "loss": 0.86526334,
      "num_input_tokens_seen": 168095905,
      "step": 7777,
      "time_per_iteration": 3.5491716861724854
    },
    {
      "auxiliary_loss_clip": 0.01172072,
      "auxiliary_loss_mlp": 0.01027122,
      "balance_loss_clip": 1.04273534,
      "balance_loss_mlp": 1.01936769,
      "epoch": 0.9352492033908495,
      "flos": 18799990156800.0,
      "grad_norm": 2.175587671269986,
      "language_loss": 0.69387865,
      "learning_rate": 4.374799770068849e-08,
      "loss": 0.71587062,
      "num_input_tokens_seen": 168112580,
      "step": 7778,
      "time_per_iteration": 2.620116710662842
    },
    {
      "auxiliary_loss_clip": 0.0116609,
      "auxiliary_loss_mlp": 0.01022654,
      "balance_loss_clip": 1.04587448,
      "balance_loss_mlp": 1.01527512,
      "epoch": 0.9353694462814887,
      "flos": 29530134241920.0,
      "grad_norm": 2.1362289651229402,
      "language_loss": 0.75026703,
      "learning_rate": 4.358610963605658e-08,
      "loss": 0.77215445,
      "num_input_tokens_seen": 168133030,
      "step": 7779,
      "time_per_iteration": 2.721810817718506
    },
    {
      "auxiliary_loss_clip": 0.01169516,
      "auxiliary_loss_mlp": 0.01031726,
      "balance_loss_clip": 1.0486238,
      "balance_loss_mlp": 1.02429926,
      "epoch": 0.9354896891721277,
      "flos": 30665450390400.0,
      "grad_norm": 5.6839910519729155,
      "language_loss": 0.68689662,
      "learning_rate": 4.342451835785677e-08,
      "loss": 0.70890903,
      "num_input_tokens_seen": 168153940,
      "step": 7780,
      "time_per_iteration": 3.598115921020508
    },
    {
      "auxiliary_loss_clip": 0.01168184,
      "auxiliary_loss_mlp": 0.01019957,
      "balance_loss_clip": 1.04370689,
      "balance_loss_mlp": 1.01281965,
      "epoch": 0.9356099320627668,
      "flos": 19463907191040.0,
      "grad_norm": 1.6571173848886593,
      "language_loss": 0.75048673,
      "learning_rate": 4.3263223890601665e-08,
      "loss": 0.77236819,
      "num_input_tokens_seen": 168172650,
      "step": 7781,
      "time_per_iteration": 2.923872709274292
    },
    {
      "auxiliary_loss_clip": 0.01163365,
      "auxiliary_loss_mlp": 0.01207717,
      "balance_loss_clip": 1.04633975,
      "balance_loss_mlp": 1.00071871,
      "epoch": 0.9357301749534058,
      "flos": 19098156954240.0,
      "grad_norm": 1.9399864866309835,
      "language_loss": 0.79446024,
      "learning_rate": 4.31022262587597e-08,
      "loss": 0.81817102,
      "num_input_tokens_seen": 168191325,
      "step": 7782,
      "time_per_iteration": 2.691876173019409
    },
    {
      "auxiliary_loss_clip": 0.01169891,
      "auxiliary_loss_mlp": 0.01028401,
      "balance_loss_clip": 1.0471251,
      "balance_loss_mlp": 1.02022362,
      "epoch": 0.935850417844045,
      "flos": 23550361776000.0,
      "grad_norm": 2.106271268282558,
      "language_loss": 0.66067922,
      "learning_rate": 4.2941525486754225e-08,
      "loss": 0.68266207,
      "num_input_tokens_seen": 168211645,
      "step": 7783,
      "time_per_iteration": 2.641573190689087
    },
    {
      "auxiliary_loss_clip": 0.01167968,
      "auxiliary_loss_mlp": 0.01021547,
      "balance_loss_clip": 1.04199958,
      "balance_loss_mlp": 1.01484179,
      "epoch": 0.935970660734684,
      "flos": 18588333265920.0,
      "grad_norm": 1.8397708551236442,
      "language_loss": 0.79356122,
      "learning_rate": 4.278112159896286e-08,
      "loss": 0.81545633,
      "num_input_tokens_seen": 168229485,
      "step": 7784,
      "time_per_iteration": 2.7653279304504395
    },
    {
      "auxiliary_loss_clip": 0.01163422,
      "auxiliary_loss_mlp": 0.01019881,
      "balance_loss_clip": 1.04021299,
      "balance_loss_mlp": 1.01315141,
      "epoch": 0.9360909036253231,
      "flos": 20631255292800.0,
      "grad_norm": 1.9508610687425991,
      "language_loss": 0.67778081,
      "learning_rate": 4.2621014619719896e-08,
      "loss": 0.69961381,
      "num_input_tokens_seen": 168247250,
      "step": 7785,
      "time_per_iteration": 2.620356559753418
    },
    {
      "auxiliary_loss_clip": 0.01073341,
      "auxiliary_loss_mlp": 0.01001923,
      "balance_loss_clip": 1.00989485,
      "balance_loss_mlp": 1.00117481,
      "epoch": 0.9362111465159623,
      "flos": 61791421052160.0,
      "grad_norm": 0.7204262964119662,
      "language_loss": 0.58583343,
      "learning_rate": 4.246120457331215e-08,
      "loss": 0.6065861,
      "num_input_tokens_seen": 168309425,
      "step": 7786,
      "time_per_iteration": 3.212420701980591
    },
    {
      "auxiliary_loss_clip": 0.011652,
      "auxiliary_loss_mlp": 0.01025511,
      "balance_loss_clip": 1.04423308,
      "balance_loss_mlp": 1.01792324,
      "epoch": 0.9363313894066013,
      "flos": 24170395368960.0,
      "grad_norm": 1.9694097912566024,
      "language_loss": 0.71829891,
      "learning_rate": 4.2301691483983325e-08,
      "loss": 0.740206,
      "num_input_tokens_seen": 168329545,
      "step": 7787,
      "time_per_iteration": 2.652975559234619
    },
    {
      "auxiliary_loss_clip": 0.01171909,
      "auxiliary_loss_mlp": 0.01021216,
      "balance_loss_clip": 1.04579365,
      "balance_loss_mlp": 1.01372397,
      "epoch": 0.9364516322972404,
      "flos": 20120354196480.0,
      "grad_norm": 1.6839434752086913,
      "language_loss": 0.760674,
      "learning_rate": 4.214247537593163e-08,
      "loss": 0.78260529,
      "num_input_tokens_seen": 168348795,
      "step": 7788,
      "time_per_iteration": 2.6050894260406494
    },
    {
      "auxiliary_loss_clip": 0.01172324,
      "auxiliary_loss_mlp": 0.01033096,
      "balance_loss_clip": 1.04391444,
      "balance_loss_mlp": 1.02571702,
      "epoch": 0.9365718751878795,
      "flos": 20703758895360.0,
      "grad_norm": 2.1519519523062414,
      "language_loss": 0.80776942,
      "learning_rate": 4.1983556273309293e-08,
      "loss": 0.82982361,
      "num_input_tokens_seen": 168367545,
      "step": 7789,
      "time_per_iteration": 2.694636583328247
    },
    {
      "auxiliary_loss_clip": 0.0116924,
      "auxiliary_loss_mlp": 0.01027931,
      "balance_loss_clip": 1.04764581,
      "balance_loss_mlp": 1.02019405,
      "epoch": 0.9366921180785186,
      "flos": 18655270260480.0,
      "grad_norm": 10.061495875290207,
      "language_loss": 0.68973196,
      "learning_rate": 4.182493420022526e-08,
      "loss": 0.71170372,
      "num_input_tokens_seen": 168383215,
      "step": 7790,
      "time_per_iteration": 2.657287359237671
    },
    {
      "auxiliary_loss_clip": 0.01172547,
      "auxiliary_loss_mlp": 0.01022025,
      "balance_loss_clip": 1.04097939,
      "balance_loss_mlp": 1.01505756,
      "epoch": 0.9368123609691577,
      "flos": 25774955815680.0,
      "grad_norm": 2.2336380406674703,
      "language_loss": 0.786605,
      "learning_rate": 4.166660918074139e-08,
      "loss": 0.80855078,
      "num_input_tokens_seen": 168403120,
      "step": 7791,
      "time_per_iteration": 3.6993966102600098
    },
    {
      "auxiliary_loss_clip": 0.01168671,
      "auxiliary_loss_mlp": 0.01024732,
      "balance_loss_clip": 1.04118419,
      "balance_loss_mlp": 1.01762748,
      "epoch": 0.9369326038597968,
      "flos": 25553386771200.0,
      "grad_norm": 2.223938927767409,
      "language_loss": 0.73512244,
      "learning_rate": 4.15085812388758e-08,
      "loss": 0.75705647,
      "num_input_tokens_seen": 168425340,
      "step": 7792,
      "time_per_iteration": 2.8708267211914062
    },
    {
      "auxiliary_loss_clip": 0.01171296,
      "auxiliary_loss_mlp": 0.01023781,
      "balance_loss_clip": 1.04453731,
      "balance_loss_mlp": 1.01622915,
      "epoch": 0.9370528467504359,
      "flos": 23220019370880.0,
      "grad_norm": 1.6657316815955951,
      "language_loss": 0.78322214,
      "learning_rate": 4.135085039860153e-08,
      "loss": 0.80517292,
      "num_input_tokens_seen": 168444740,
      "step": 7793,
      "time_per_iteration": 2.6559433937072754
    },
    {
      "auxiliary_loss_clip": 0.01170901,
      "auxiliary_loss_mlp": 0.01021257,
      "balance_loss_clip": 1.04659879,
      "balance_loss_mlp": 1.01378834,
      "epoch": 0.9371730896410749,
      "flos": 24967468120320.0,
      "grad_norm": 2.6557178781724917,
      "language_loss": 0.7873773,
      "learning_rate": 4.1193416683845906e-08,
      "loss": 0.80929893,
      "num_input_tokens_seen": 168463670,
      "step": 7794,
      "time_per_iteration": 2.613276720046997
    },
    {
      "auxiliary_loss_clip": 0.01176951,
      "auxiliary_loss_mlp": 0.01024895,
      "balance_loss_clip": 1.04529166,
      "balance_loss_mlp": 1.01846409,
      "epoch": 0.9372933325317141,
      "flos": 15553091134080.0,
      "grad_norm": 2.325211742778355,
      "language_loss": 0.83753258,
      "learning_rate": 4.103628011849136e-08,
      "loss": 0.85955107,
      "num_input_tokens_seen": 168479030,
      "step": 7795,
      "time_per_iteration": 2.722339630126953
    },
    {
      "auxiliary_loss_clip": 0.01172295,
      "auxiliary_loss_mlp": 0.0102372,
      "balance_loss_clip": 1.04442346,
      "balance_loss_mlp": 1.01652277,
      "epoch": 0.9374135754223532,
      "flos": 21871861182720.0,
      "grad_norm": 1.9695229554129132,
      "language_loss": 0.75866705,
      "learning_rate": 4.0879440726375506e-08,
      "loss": 0.78062725,
      "num_input_tokens_seen": 168496815,
      "step": 7796,
      "time_per_iteration": 2.730821371078491
    },
    {
      "auxiliary_loss_clip": 0.01167344,
      "auxiliary_loss_mlp": 0.01022868,
      "balance_loss_clip": 1.03979337,
      "balance_loss_mlp": 1.01562619,
      "epoch": 0.9375338183129922,
      "flos": 22631048064000.0,
      "grad_norm": 3.8236421702954155,
      "language_loss": 0.56477648,
      "learning_rate": 4.0722898531291074e-08,
      "loss": 0.58667856,
      "num_input_tokens_seen": 168514055,
      "step": 7797,
      "time_per_iteration": 2.6582157611846924
    },
    {
      "auxiliary_loss_clip": 0.01177099,
      "auxiliary_loss_mlp": 0.01024427,
      "balance_loss_clip": 1.04455197,
      "balance_loss_mlp": 1.01688755,
      "epoch": 0.9376540612036314,
      "flos": 26104292640000.0,
      "grad_norm": 1.7180205603428886,
      "language_loss": 0.76825452,
      "learning_rate": 4.0566653556985295e-08,
      "loss": 0.79026985,
      "num_input_tokens_seen": 168534600,
      "step": 7798,
      "time_per_iteration": 2.7513904571533203
    },
    {
      "auxiliary_loss_clip": 0.01185444,
      "auxiliary_loss_mlp": 0.01031496,
      "balance_loss_clip": 1.03980398,
      "balance_loss_mlp": 1.02330935,
      "epoch": 0.9377743040942704,
      "flos": 19717580016000.0,
      "grad_norm": 2.308084091335419,
      "language_loss": 0.81527996,
      "learning_rate": 4.0410705827159886e-08,
      "loss": 0.83744931,
      "num_input_tokens_seen": 168551895,
      "step": 7799,
      "time_per_iteration": 2.704897403717041
    },
    {
      "auxiliary_loss_clip": 0.01167858,
      "auxiliary_loss_mlp": 0.01022857,
      "balance_loss_clip": 1.04155111,
      "balance_loss_mlp": 1.01535273,
      "epoch": 0.9378945469849095,
      "flos": 15267530010240.0,
      "grad_norm": 2.0095871063489232,
      "language_loss": 0.71437323,
      "learning_rate": 4.0255055365472356e-08,
      "loss": 0.73628038,
      "num_input_tokens_seen": 168569990,
      "step": 7800,
      "time_per_iteration": 2.641146659851074
    },
    {
      "auxiliary_loss_clip": 0.01180094,
      "auxiliary_loss_mlp": 0.01027253,
      "balance_loss_clip": 1.03814912,
      "balance_loss_mlp": 1.01993918,
      "epoch": 0.9380147898755486,
      "flos": 20591394174720.0,
      "grad_norm": 2.0413544432647153,
      "language_loss": 0.74531472,
      "learning_rate": 4.009970219553471e-08,
      "loss": 0.76738822,
      "num_input_tokens_seen": 168586940,
      "step": 7801,
      "time_per_iteration": 3.5511956214904785
    },
    {
      "auxiliary_loss_clip": 0.01173573,
      "auxiliary_loss_mlp": 0.01026775,
      "balance_loss_clip": 1.04640627,
      "balance_loss_mlp": 1.01870418,
      "epoch": 0.9381350327661877,
      "flos": 26281116316800.0,
      "grad_norm": 3.661766854075664,
      "language_loss": 0.76418614,
      "learning_rate": 3.99446463409141e-08,
      "loss": 0.78618968,
      "num_input_tokens_seen": 168604795,
      "step": 7802,
      "time_per_iteration": 2.662249803543091
    },
    {
      "auxiliary_loss_clip": 0.0117294,
      "auxiliary_loss_mlp": 0.01024086,
      "balance_loss_clip": 1.04383564,
      "balance_loss_mlp": 1.01640344,
      "epoch": 0.9382552756568268,
      "flos": 23586344225280.0,
      "grad_norm": 2.119332285583555,
      "language_loss": 0.68961871,
      "learning_rate": 3.978988782513215e-08,
      "loss": 0.71158898,
      "num_input_tokens_seen": 168622290,
      "step": 7803,
      "time_per_iteration": 3.564380407333374
    },
    {
      "auxiliary_loss_clip": 0.01172367,
      "auxiliary_loss_mlp": 0.01020422,
      "balance_loss_clip": 1.04518533,
      "balance_loss_mlp": 1.01309371,
      "epoch": 0.9383755185474659,
      "flos": 28438809275520.0,
      "grad_norm": 3.106206187347868,
      "language_loss": 0.76534081,
      "learning_rate": 3.963542667166586e-08,
      "loss": 0.78726864,
      "num_input_tokens_seen": 168642395,
      "step": 7804,
      "time_per_iteration": 2.6651227474212646
    },
    {
      "auxiliary_loss_clip": 0.01176723,
      "auxiliary_loss_mlp": 0.01024881,
      "balance_loss_clip": 1.04698956,
      "balance_loss_mlp": 1.01790452,
      "epoch": 0.938495761438105,
      "flos": 20449583280000.0,
      "grad_norm": 2.331523255472263,
      "language_loss": 0.68357682,
      "learning_rate": 3.9481262903946486e-08,
      "loss": 0.70559287,
      "num_input_tokens_seen": 168661840,
      "step": 7805,
      "time_per_iteration": 2.6628737449645996
    },
    {
      "auxiliary_loss_clip": 0.01093435,
      "auxiliary_loss_mlp": 0.01004749,
      "balance_loss_clip": 1.01180184,
      "balance_loss_mlp": 1.00400949,
      "epoch": 0.938616004328744,
      "flos": 69302711658240.0,
      "grad_norm": 0.775260695774372,
      "language_loss": 0.54462165,
      "learning_rate": 3.932739654536066e-08,
      "loss": 0.56560349,
      "num_input_tokens_seen": 168724540,
      "step": 7806,
      "time_per_iteration": 3.2064712047576904
    },
    {
      "auxiliary_loss_clip": 0.01168808,
      "auxiliary_loss_mlp": 0.01026424,
      "balance_loss_clip": 1.0473187,
      "balance_loss_mlp": 1.0196383,
      "epoch": 0.9387362472193832,
      "flos": 18911636605440.0,
      "grad_norm": 2.157636484109322,
      "language_loss": 0.74161565,
      "learning_rate": 3.917382761925014e-08,
      "loss": 0.76356798,
      "num_input_tokens_seen": 168740375,
      "step": 7807,
      "time_per_iteration": 3.5389790534973145
    },
    {
      "auxiliary_loss_clip": 0.01166389,
      "auxiliary_loss_mlp": 0.01026968,
      "balance_loss_clip": 1.0460794,
      "balance_loss_mlp": 1.01999998,
      "epoch": 0.9388564901100223,
      "flos": 26501967089280.0,
      "grad_norm": 1.7343640290128817,
      "language_loss": 0.79653966,
      "learning_rate": 3.9020556148910754e-08,
      "loss": 0.81847322,
      "num_input_tokens_seen": 168759730,
      "step": 7808,
      "time_per_iteration": 2.6952590942382812
    },
    {
      "auxiliary_loss_clip": 0.01079333,
      "auxiliary_loss_mlp": 0.01000803,
      "balance_loss_clip": 1.01129949,
      "balance_loss_mlp": 0.99998063,
      "epoch": 0.9389767330006613,
      "flos": 58941083157120.0,
      "grad_norm": 0.7072581496550967,
      "language_loss": 0.56669533,
      "learning_rate": 3.8867582157593895e-08,
      "loss": 0.5874967,
      "num_input_tokens_seen": 168813935,
      "step": 7809,
      "time_per_iteration": 2.959533452987671
    },
    {
      "auxiliary_loss_clip": 0.01169699,
      "auxiliary_loss_mlp": 0.0102223,
      "balance_loss_clip": 1.04820859,
      "balance_loss_mlp": 1.01499677,
      "epoch": 0.9390969758913005,
      "flos": 31102554994560.0,
      "grad_norm": 1.8421733596910606,
      "language_loss": 0.76277816,
      "learning_rate": 3.871490566850544e-08,
      "loss": 0.78469741,
      "num_input_tokens_seen": 168838145,
      "step": 7810,
      "time_per_iteration": 2.6962690353393555
    },
    {
      "auxiliary_loss_clip": 0.01168543,
      "auxiliary_loss_mlp": 0.01022749,
      "balance_loss_clip": 1.04497361,
      "balance_loss_mlp": 1.01519668,
      "epoch": 0.9392172187819395,
      "flos": 22419391173120.0,
      "grad_norm": 1.732843299286383,
      "language_loss": 0.70888597,
      "learning_rate": 3.856252670480642e-08,
      "loss": 0.73079884,
      "num_input_tokens_seen": 168856805,
      "step": 7811,
      "time_per_iteration": 2.7203917503356934
    },
    {
      "auxiliary_loss_clip": 0.01167784,
      "auxiliary_loss_mlp": 0.01027152,
      "balance_loss_clip": 1.04118466,
      "balance_loss_mlp": 1.01942158,
      "epoch": 0.9393374616725786,
      "flos": 19719483436800.0,
      "grad_norm": 1.937333041003612,
      "language_loss": 0.81560075,
      "learning_rate": 3.841044528961279e-08,
      "loss": 0.83755016,
      "num_input_tokens_seen": 168874600,
      "step": 7812,
      "time_per_iteration": 2.6497886180877686
    },
    {
      "auxiliary_loss_clip": 0.01166339,
      "auxiliary_loss_mlp": 0.01024936,
      "balance_loss_clip": 1.04535747,
      "balance_loss_mlp": 1.01750374,
      "epoch": 0.9394577045632178,
      "flos": 24170215800960.0,
      "grad_norm": 2.9195352044995793,
      "language_loss": 0.78885603,
      "learning_rate": 3.825866144599477e-08,
      "loss": 0.81076872,
      "num_input_tokens_seen": 168893655,
      "step": 7813,
      "time_per_iteration": 2.604292154312134
    },
    {
      "auxiliary_loss_clip": 0.01172336,
      "auxiliary_loss_mlp": 0.01020226,
      "balance_loss_clip": 1.04341078,
      "balance_loss_mlp": 1.01279378,
      "epoch": 0.9395779474538568,
      "flos": 19023929498880.0,
      "grad_norm": 1.9927541167069105,
      "language_loss": 0.75496638,
      "learning_rate": 3.8107175196978145e-08,
      "loss": 0.77689201,
      "num_input_tokens_seen": 168909960,
      "step": 7814,
      "time_per_iteration": 2.65568208694458
    },
    {
      "auxiliary_loss_clip": 0.01168951,
      "auxiliary_loss_mlp": 0.0102547,
      "balance_loss_clip": 1.04194021,
      "balance_loss_mlp": 1.01869845,
      "epoch": 0.9396981903444959,
      "flos": 14319129260160.0,
      "grad_norm": 1.859129279212071,
      "language_loss": 0.76619756,
      "learning_rate": 3.7955986565542996e-08,
      "loss": 0.78814185,
      "num_input_tokens_seen": 168928040,
      "step": 7815,
      "time_per_iteration": 2.6571688652038574
    },
    {
      "auxiliary_loss_clip": 0.01174389,
      "auxiliary_loss_mlp": 0.01030884,
      "balance_loss_clip": 1.04274189,
      "balance_loss_mlp": 1.02339149,
      "epoch": 0.9398184332351349,
      "flos": 34787564202240.0,
      "grad_norm": 2.465675959622592,
      "language_loss": 0.68248355,
      "learning_rate": 3.780509557462497e-08,
      "loss": 0.70453632,
      "num_input_tokens_seen": 168948240,
      "step": 7816,
      "time_per_iteration": 2.7709412574768066
    },
    {
      "auxiliary_loss_clip": 0.01168254,
      "auxiliary_loss_mlp": 0.01022157,
      "balance_loss_clip": 1.04192853,
      "balance_loss_mlp": 1.01420617,
      "epoch": 0.9399386761257741,
      "flos": 25372253462400.0,
      "grad_norm": 2.9829092031976576,
      "language_loss": 0.75715876,
      "learning_rate": 3.765450224711375e-08,
      "loss": 0.77906287,
      "num_input_tokens_seen": 168968745,
      "step": 7817,
      "time_per_iteration": 2.6606407165527344
    },
    {
      "auxiliary_loss_clip": 0.01165378,
      "auxiliary_loss_mlp": 0.01023511,
      "balance_loss_clip": 1.04440165,
      "balance_loss_mlp": 1.01601291,
      "epoch": 0.9400589190164131,
      "flos": 27304965584640.0,
      "grad_norm": 1.7435806085300396,
      "language_loss": 0.8004002,
      "learning_rate": 3.750420660585396e-08,
      "loss": 0.82228911,
      "num_input_tokens_seen": 168990685,
      "step": 7818,
      "time_per_iteration": 3.617560863494873
    },
    {
      "auxiliary_loss_clip": 0.01165543,
      "auxiliary_loss_mlp": 0.01023045,
      "balance_loss_clip": 1.04804826,
      "balance_loss_mlp": 1.01586843,
      "epoch": 0.9401791619070522,
      "flos": 23399859790080.0,
      "grad_norm": 2.402950357711973,
      "language_loss": 0.79803574,
      "learning_rate": 3.735420867364603e-08,
      "loss": 0.81992161,
      "num_input_tokens_seen": 169011665,
      "step": 7819,
      "time_per_iteration": 2.595315456390381
    },
    {
      "auxiliary_loss_clip": 0.01168383,
      "auxiliary_loss_mlp": 0.01020762,
      "balance_loss_clip": 1.03489745,
      "balance_loss_mlp": 1.01393771,
      "epoch": 0.9402994047976914,
      "flos": 35881403120640.0,
      "grad_norm": 1.6509477333218867,
      "language_loss": 0.61732745,
      "learning_rate": 3.7204508473244186e-08,
      "loss": 0.63921893,
      "num_input_tokens_seen": 169035290,
      "step": 7820,
      "time_per_iteration": 2.8121488094329834
    },
    {
      "auxiliary_loss_clip": 0.0117679,
      "auxiliary_loss_mlp": 0.01020129,
      "balance_loss_clip": 1.03934455,
      "balance_loss_mlp": 1.01342964,
      "epoch": 0.9404196476883304,
      "flos": 22236821320320.0,
      "grad_norm": 1.8127963888525007,
      "language_loss": 0.6923064,
      "learning_rate": 3.7055106027357395e-08,
      "loss": 0.7142756,
      "num_input_tokens_seen": 169055155,
      "step": 7821,
      "time_per_iteration": 2.7823140621185303
    },
    {
      "auxiliary_loss_clip": 0.01164632,
      "auxiliary_loss_mlp": 0.01024253,
      "balance_loss_clip": 1.04552364,
      "balance_loss_mlp": 1.01653731,
      "epoch": 0.9405398905789695,
      "flos": 18915802583040.0,
      "grad_norm": 2.2613696266003105,
      "language_loss": 0.72016549,
      "learning_rate": 3.690600135865063e-08,
      "loss": 0.74205434,
      "num_input_tokens_seen": 169072080,
      "step": 7822,
      "time_per_iteration": 2.6529765129089355
    },
    {
      "auxiliary_loss_clip": 0.01090686,
      "auxiliary_loss_mlp": 0.01001396,
      "balance_loss_clip": 1.01073122,
      "balance_loss_mlp": 1.00065422,
      "epoch": 0.9406601334696086,
      "flos": 70274130048000.0,
      "grad_norm": 0.786440374934692,
      "language_loss": 0.57997108,
      "learning_rate": 3.675719448974246e-08,
      "loss": 0.60089189,
      "num_input_tokens_seen": 169137170,
      "step": 7823,
      "time_per_iteration": 3.284353256225586
    },
    {
      "auxiliary_loss_clip": 0.01172047,
      "auxiliary_loss_mlp": 0.01207425,
      "balance_loss_clip": 1.04067731,
      "balance_loss_mlp": 1.00069368,
      "epoch": 0.9407803763602477,
      "flos": 22165071903360.0,
      "grad_norm": 2.957905343064848,
      "language_loss": 0.60055852,
      "learning_rate": 3.6608685443207054e-08,
      "loss": 0.62435317,
      "num_input_tokens_seen": 169156320,
      "step": 7824,
      "time_per_iteration": 2.6643028259277344
    },
    {
      "auxiliary_loss_clip": 0.01173176,
      "auxiliary_loss_mlp": 0.01022341,
      "balance_loss_clip": 1.04125834,
      "balance_loss_mlp": 1.0155555,
      "epoch": 0.9409006192508867,
      "flos": 18879496911360.0,
      "grad_norm": 5.096794450004414,
      "language_loss": 0.66887522,
      "learning_rate": 3.646047424157306e-08,
      "loss": 0.69083035,
      "num_input_tokens_seen": 169173295,
      "step": 7825,
      "time_per_iteration": 2.667395830154419
    },
    {
      "auxiliary_loss_clip": 0.01172554,
      "auxiliary_loss_mlp": 0.01025384,
      "balance_loss_clip": 1.04520535,
      "balance_loss_mlp": 1.01745677,
      "epoch": 0.9410208621415259,
      "flos": 23368258800000.0,
      "grad_norm": 5.23773209415688,
      "language_loss": 0.68931806,
      "learning_rate": 3.631256090732382e-08,
      "loss": 0.71129751,
      "num_input_tokens_seen": 169193755,
      "step": 7826,
      "time_per_iteration": 2.5838332176208496
    },
    {
      "auxiliary_loss_clip": 0.01173268,
      "auxiliary_loss_mlp": 0.0102358,
      "balance_loss_clip": 1.04405618,
      "balance_loss_mlp": 1.0170505,
      "epoch": 0.941141105032165,
      "flos": 22742227635840.0,
      "grad_norm": 1.6611938907697805,
      "language_loss": 0.82873929,
      "learning_rate": 3.6164945462897833e-08,
      "loss": 0.85070777,
      "num_input_tokens_seen": 169213045,
      "step": 7827,
      "time_per_iteration": 3.583796977996826
    },
    {
      "auxiliary_loss_clip": 0.01168261,
      "auxiliary_loss_mlp": 0.01207238,
      "balance_loss_clip": 1.04706061,
      "balance_loss_mlp": 1.00068069,
      "epoch": 0.941261347922804,
      "flos": 20704908130560.0,
      "grad_norm": 2.1673382382597723,
      "language_loss": 0.75777167,
      "learning_rate": 3.6017627930687856e-08,
      "loss": 0.78152663,
      "num_input_tokens_seen": 169232870,
      "step": 7828,
      "time_per_iteration": 2.6214723587036133
    },
    {
      "auxiliary_loss_clip": 0.01172511,
      "auxiliary_loss_mlp": 0.0102004,
      "balance_loss_clip": 1.03901434,
      "balance_loss_mlp": 1.01292288,
      "epoch": 0.9413815908134432,
      "flos": 19421998997760.0,
      "grad_norm": 4.2057109383215225,
      "language_loss": 0.77053046,
      "learning_rate": 3.587060833304267e-08,
      "loss": 0.79245603,
      "num_input_tokens_seen": 169251060,
      "step": 7829,
      "time_per_iteration": 2.622770309448242
    },
    {
      "auxiliary_loss_clip": 0.01172259,
      "auxiliary_loss_mlp": 0.0102886,
      "balance_loss_clip": 1.04721928,
      "balance_loss_mlp": 1.02123618,
      "epoch": 0.9415018337040822,
      "flos": 17493452853120.0,
      "grad_norm": 2.073995843051633,
      "language_loss": 0.64140129,
      "learning_rate": 3.5723886692264225e-08,
      "loss": 0.66341245,
      "num_input_tokens_seen": 169268600,
      "step": 7830,
      "time_per_iteration": 3.552737236022949
    },
    {
      "auxiliary_loss_clip": 0.01169398,
      "auxiliary_loss_mlp": 0.01024291,
      "balance_loss_clip": 1.04277503,
      "balance_loss_mlp": 1.01743102,
      "epoch": 0.9416220765947213,
      "flos": 31831613343360.0,
      "grad_norm": 1.9170692081020655,
      "language_loss": 0.62285876,
      "learning_rate": 3.557746303061071e-08,
      "loss": 0.64479566,
      "num_input_tokens_seen": 169290355,
      "step": 7831,
      "time_per_iteration": 2.6838810443878174
    },
    {
      "auxiliary_loss_clip": 0.01168364,
      "auxiliary_loss_mlp": 0.01023058,
      "balance_loss_clip": 1.04255271,
      "balance_loss_mlp": 1.01595902,
      "epoch": 0.9417423194853605,
      "flos": 23511973115520.0,
      "grad_norm": 2.811634776270239,
      "language_loss": 0.72279978,
      "learning_rate": 3.543133737029391e-08,
      "loss": 0.74471396,
      "num_input_tokens_seen": 169310865,
      "step": 7832,
      "time_per_iteration": 2.6585769653320312
    },
    {
      "auxiliary_loss_clip": 0.01172657,
      "auxiliary_loss_mlp": 0.01025004,
      "balance_loss_clip": 1.04558885,
      "balance_loss_mlp": 1.01759553,
      "epoch": 0.9418625623759995,
      "flos": 23915106432000.0,
      "grad_norm": 2.2380025907820964,
      "language_loss": 0.69158781,
      "learning_rate": 3.5285509733481214e-08,
      "loss": 0.7135644,
      "num_input_tokens_seen": 169330590,
      "step": 7833,
      "time_per_iteration": 2.7196409702301025
    },
    {
      "auxiliary_loss_clip": 0.01165325,
      "auxiliary_loss_mlp": 0.01027215,
      "balance_loss_clip": 1.04404223,
      "balance_loss_mlp": 1.01916873,
      "epoch": 0.9419828052666386,
      "flos": 18076965292800.0,
      "grad_norm": 2.0931774167593513,
      "language_loss": 0.76666212,
      "learning_rate": 3.513998014229469e-08,
      "loss": 0.78858757,
      "num_input_tokens_seen": 169349540,
      "step": 7834,
      "time_per_iteration": 3.430319309234619
    },
    {
      "auxiliary_loss_clip": 0.01173508,
      "auxiliary_loss_mlp": 0.01024642,
      "balance_loss_clip": 1.04603696,
      "balance_loss_mlp": 1.0175879,
      "epoch": 0.9421030481572777,
      "flos": 17712328377600.0,
      "grad_norm": 2.343651045245585,
      "language_loss": 0.86237657,
      "learning_rate": 3.499474861881069e-08,
      "loss": 0.88435811,
      "num_input_tokens_seen": 169366765,
      "step": 7835,
      "time_per_iteration": 2.598116874694824
    },
    {
      "auxiliary_loss_clip": 0.01180164,
      "auxiliary_loss_mlp": 0.01020236,
      "balance_loss_clip": 1.04081833,
      "balance_loss_mlp": 1.01300931,
      "epoch": 0.9422232910479168,
      "flos": 20194114775040.0,
      "grad_norm": 2.024426653298815,
      "language_loss": 0.68095398,
      "learning_rate": 3.4849815185061136e-08,
      "loss": 0.70295799,
      "num_input_tokens_seen": 169386655,
      "step": 7836,
      "time_per_iteration": 2.6808269023895264
    },
    {
      "auxiliary_loss_clip": 0.01165302,
      "auxiliary_loss_mlp": 0.01023435,
      "balance_loss_clip": 1.04221416,
      "balance_loss_mlp": 1.01653862,
      "epoch": 0.9423435339385559,
      "flos": 18442571875200.0,
      "grad_norm": 2.2516240488609225,
      "language_loss": 0.76109135,
      "learning_rate": 3.470517986303223e-08,
      "loss": 0.78297877,
      "num_input_tokens_seen": 169405640,
      "step": 7837,
      "time_per_iteration": 2.613650321960449
    },
    {
      "auxiliary_loss_clip": 0.01173075,
      "auxiliary_loss_mlp": 0.01031279,
      "balance_loss_clip": 1.04579663,
      "balance_loss_mlp": 1.02391744,
      "epoch": 0.942463776829195,
      "flos": 20080636732800.0,
      "grad_norm": 2.289896656725464,
      "language_loss": 0.79507565,
      "learning_rate": 3.4560842674664856e-08,
      "loss": 0.81711924,
      "num_input_tokens_seen": 169424155,
      "step": 7838,
      "time_per_iteration": 2.656808614730835
    },
    {
      "auxiliary_loss_clip": 0.01170546,
      "auxiliary_loss_mlp": 0.01022317,
      "balance_loss_clip": 1.04383981,
      "balance_loss_mlp": 1.01474154,
      "epoch": 0.9425840197198341,
      "flos": 22636255536000.0,
      "grad_norm": 2.155507598549225,
      "language_loss": 0.75312299,
      "learning_rate": 3.441680364185506e-08,
      "loss": 0.77505159,
      "num_input_tokens_seen": 169444025,
      "step": 7839,
      "time_per_iteration": 2.670708179473877
    },
    {
      "auxiliary_loss_clip": 0.0117298,
      "auxiliary_loss_mlp": 0.01026288,
      "balance_loss_clip": 1.04644728,
      "balance_loss_mlp": 1.0191716,
      "epoch": 0.9427042626104731,
      "flos": 19937892084480.0,
      "grad_norm": 2.679687269739271,
      "language_loss": 0.75344867,
      "learning_rate": 3.427306278645314e-08,
      "loss": 0.77544135,
      "num_input_tokens_seen": 169462480,
      "step": 7840,
      "time_per_iteration": 2.6372458934783936
    },
    {
      "auxiliary_loss_clip": 0.01175656,
      "auxiliary_loss_mlp": 0.01024867,
      "balance_loss_clip": 1.0416441,
      "balance_loss_mlp": 1.0176909,
      "epoch": 0.9428245055011123,
      "flos": 22856998567680.0,
      "grad_norm": 2.3769963636842624,
      "language_loss": 0.73214066,
      "learning_rate": 3.4129620130264767e-08,
      "loss": 0.75414586,
      "num_input_tokens_seen": 169480840,
      "step": 7841,
      "time_per_iteration": 2.6257152557373047
    },
    {
      "auxiliary_loss_clip": 0.0117664,
      "auxiliary_loss_mlp": 0.0120729,
      "balance_loss_clip": 1.04732537,
      "balance_loss_mlp": 1.00071073,
      "epoch": 0.9429447483917514,
      "flos": 20951757371520.0,
      "grad_norm": 5.18463424265482,
      "language_loss": 0.77814543,
      "learning_rate": 3.398647569505009e-08,
      "loss": 0.80198473,
      "num_input_tokens_seen": 169498265,
      "step": 7842,
      "time_per_iteration": 2.6103570461273193
    },
    {
      "auxiliary_loss_clip": 0.01178827,
      "auxiliary_loss_mlp": 0.01022282,
      "balance_loss_clip": 1.04301584,
      "balance_loss_mlp": 1.0147121,
      "epoch": 0.9430649912823904,
      "flos": 18843658116480.0,
      "grad_norm": 33.74280651345444,
      "language_loss": 0.7502287,
      "learning_rate": 3.384362950252373e-08,
      "loss": 0.7722398,
      "num_input_tokens_seen": 169515235,
      "step": 7843,
      "time_per_iteration": 2.6361799240112305
    },
    {
      "auxiliary_loss_clip": 0.01167029,
      "auxiliary_loss_mlp": 0.0102178,
      "balance_loss_clip": 1.04138947,
      "balance_loss_mlp": 1.01444602,
      "epoch": 0.9431852341730296,
      "flos": 32556038837760.0,
      "grad_norm": 4.921304196819366,
      "language_loss": 0.57037801,
      "learning_rate": 3.3701081574355473e-08,
      "loss": 0.59226608,
      "num_input_tokens_seen": 169537195,
      "step": 7844,
      "time_per_iteration": 2.741123676300049
    },
    {
      "auxiliary_loss_clip": 0.01076984,
      "auxiliary_loss_mlp": 0.01001924,
      "balance_loss_clip": 1.01008546,
      "balance_loss_mlp": 1.00120878,
      "epoch": 0.9433054770636686,
      "flos": 66904490252160.0,
      "grad_norm": 0.6390888279357521,
      "language_loss": 0.51646686,
      "learning_rate": 3.3558831932169796e-08,
      "loss": 0.537256,
      "num_input_tokens_seen": 169605865,
      "step": 7845,
      "time_per_iteration": 4.16228985786438
    },
    {
      "auxiliary_loss_clip": 0.01166991,
      "auxiliary_loss_mlp": 0.01025247,
      "balance_loss_clip": 1.04484749,
      "balance_loss_mlp": 1.01814246,
      "epoch": 0.9434257199543077,
      "flos": 26140346916480.0,
      "grad_norm": 1.8778829359315081,
      "language_loss": 0.88231438,
      "learning_rate": 3.341688059754588e-08,
      "loss": 0.90423679,
      "num_input_tokens_seen": 169621520,
      "step": 7846,
      "time_per_iteration": 2.6791481971740723
    },
    {
      "auxiliary_loss_clip": 0.0117947,
      "auxiliary_loss_mlp": 0.01207061,
      "balance_loss_clip": 1.04154015,
      "balance_loss_mlp": 1.00076485,
      "epoch": 0.9435459628449467,
      "flos": 25003486483200.0,
      "grad_norm": 2.129510275081448,
      "language_loss": 0.77627569,
      "learning_rate": 3.327522759201762e-08,
      "loss": 0.80014098,
      "num_input_tokens_seen": 169641390,
      "step": 7847,
      "time_per_iteration": 2.654139757156372
    },
    {
      "auxiliary_loss_clip": 0.01171857,
      "auxiliary_loss_mlp": 0.01024877,
      "balance_loss_clip": 1.04231286,
      "balance_loss_mlp": 1.01722991,
      "epoch": 0.9436662057355859,
      "flos": 22163240309760.0,
      "grad_norm": 2.1556195373637,
      "language_loss": 0.66956878,
      "learning_rate": 3.313387293707359e-08,
      "loss": 0.69153607,
      "num_input_tokens_seen": 169660095,
      "step": 7848,
      "time_per_iteration": 2.6363677978515625
    },
    {
      "auxiliary_loss_clip": 0.01169641,
      "auxiliary_loss_mlp": 0.01027032,
      "balance_loss_clip": 1.04295158,
      "balance_loss_mlp": 1.0191251,
      "epoch": 0.943786448626225,
      "flos": 20118522602880.0,
      "grad_norm": 1.8827500452399597,
      "language_loss": 0.68566453,
      "learning_rate": 3.29928166541571e-08,
      "loss": 0.70763123,
      "num_input_tokens_seen": 169679050,
      "step": 7849,
      "time_per_iteration": 2.649028778076172
    },
    {
      "auxiliary_loss_clip": 0.01160055,
      "auxiliary_loss_mlp": 0.01023027,
      "balance_loss_clip": 1.04221594,
      "balance_loss_mlp": 1.01573181,
      "epoch": 0.943906691516864,
      "flos": 22090808534400.0,
      "grad_norm": 2.0894604545622406,
      "language_loss": 0.80775613,
      "learning_rate": 3.2852058764666346e-08,
      "loss": 0.82958698,
      "num_input_tokens_seen": 169698150,
      "step": 7850,
      "time_per_iteration": 2.631256103515625
    },
    {
      "auxiliary_loss_clip": 0.01160663,
      "auxiliary_loss_mlp": 0.01024096,
      "balance_loss_clip": 1.04361737,
      "balance_loss_mlp": 1.0172863,
      "epoch": 0.9440269344075032,
      "flos": 35298501212160.0,
      "grad_norm": 2.3385594614930456,
      "language_loss": 0.68648362,
      "learning_rate": 3.2711599289954264e-08,
      "loss": 0.70833123,
      "num_input_tokens_seen": 169722185,
      "step": 7851,
      "time_per_iteration": 2.8083860874176025
    },
    {
      "auxiliary_loss_clip": 0.01180566,
      "auxiliary_loss_mlp": 0.01028457,
      "balance_loss_clip": 1.04058409,
      "balance_loss_mlp": 1.02148879,
      "epoch": 0.9441471772981422,
      "flos": 19238136255360.0,
      "grad_norm": 2.332271176015003,
      "language_loss": 0.77885437,
      "learning_rate": 3.257143825132847e-08,
      "loss": 0.80094463,
      "num_input_tokens_seen": 169740355,
      "step": 7852,
      "time_per_iteration": 2.669107675552368
    },
    {
      "auxiliary_loss_clip": 0.01170777,
      "auxiliary_loss_mlp": 0.01020192,
      "balance_loss_clip": 1.0437839,
      "balance_loss_mlp": 1.01303053,
      "epoch": 0.9442674201887813,
      "flos": 25739799379200.0,
      "grad_norm": 1.9981753024794946,
      "language_loss": 0.76645529,
      "learning_rate": 3.243157567005106e-08,
      "loss": 0.78836501,
      "num_input_tokens_seen": 169758535,
      "step": 7853,
      "time_per_iteration": 2.684828758239746
    },
    {
      "auxiliary_loss_clip": 0.01171772,
      "auxiliary_loss_mlp": 0.01023804,
      "balance_loss_clip": 1.0508976,
      "balance_loss_mlp": 1.01631725,
      "epoch": 0.9443876630794205,
      "flos": 15523321737600.0,
      "grad_norm": 3.5906572700707593,
      "language_loss": 0.64126813,
      "learning_rate": 3.2292011567339296e-08,
      "loss": 0.66322386,
      "num_input_tokens_seen": 169776340,
      "step": 7854,
      "time_per_iteration": 3.4913504123687744
    },
    {
      "auxiliary_loss_clip": 0.01169069,
      "auxiliary_loss_mlp": 0.01207324,
      "balance_loss_clip": 1.04459524,
      "balance_loss_mlp": 1.00067496,
      "epoch": 0.9445079059700595,
      "flos": 13400821128960.0,
      "grad_norm": 2.0273555172142106,
      "language_loss": 0.55917299,
      "learning_rate": 3.21527459643649e-08,
      "loss": 0.58293688,
      "num_input_tokens_seen": 169793225,
      "step": 7855,
      "time_per_iteration": 2.719740152359009
    },
    {
      "auxiliary_loss_clip": 0.01171374,
      "auxiliary_loss_mlp": 0.01024851,
      "balance_loss_clip": 1.04592752,
      "balance_loss_mlp": 1.01739454,
      "epoch": 0.9446281488606986,
      "flos": 23659242877440.0,
      "grad_norm": 1.8985376171522603,
      "language_loss": 0.74087977,
      "learning_rate": 3.2013778882254536e-08,
      "loss": 0.76284206,
      "num_input_tokens_seen": 169812020,
      "step": 7856,
      "time_per_iteration": 2.70186448097229
    },
    {
      "auxiliary_loss_clip": 0.01160619,
      "auxiliary_loss_mlp": 0.01031302,
      "balance_loss_clip": 1.04421473,
      "balance_loss_mlp": 1.02427459,
      "epoch": 0.9447483917513377,
      "flos": 25557337267200.0,
      "grad_norm": 1.9378541389326902,
      "language_loss": 0.76078355,
      "learning_rate": 3.1875110342088676e-08,
      "loss": 0.78270274,
      "num_input_tokens_seen": 169833470,
      "step": 7857,
      "time_per_iteration": 3.6294796466827393
    },
    {
      "auxiliary_loss_clip": 0.01164978,
      "auxiliary_loss_mlp": 0.01024438,
      "balance_loss_clip": 1.0445931,
      "balance_loss_mlp": 1.01733947,
      "epoch": 0.9448686346419768,
      "flos": 24535463247360.0,
      "grad_norm": 1.8851587068633875,
      "language_loss": 0.65781945,
      "learning_rate": 3.1736740364904035e-08,
      "loss": 0.67971361,
      "num_input_tokens_seen": 169854000,
      "step": 7858,
      "time_per_iteration": 2.7967398166656494
    },
    {
      "auxiliary_loss_clip": 0.01172869,
      "auxiliary_loss_mlp": 0.01207539,
      "balance_loss_clip": 1.04066682,
      "balance_loss_mlp": 1.00075412,
      "epoch": 0.9449888775326158,
      "flos": 14721256995840.0,
      "grad_norm": 2.266090398582784,
      "language_loss": 0.77239478,
      "learning_rate": 3.159866897169094e-08,
      "loss": 0.79619884,
      "num_input_tokens_seen": 169872200,
      "step": 7859,
      "time_per_iteration": 2.7177398204803467
    },
    {
      "auxiliary_loss_clip": 0.0118234,
      "auxiliary_loss_mlp": 0.01025114,
      "balance_loss_clip": 1.04535568,
      "balance_loss_mlp": 1.01798487,
      "epoch": 0.945109120423255,
      "flos": 15447873219840.0,
      "grad_norm": 2.6028414474988875,
      "language_loss": 0.76098311,
      "learning_rate": 3.146089618339487e-08,
      "loss": 0.78305757,
      "num_input_tokens_seen": 169889055,
      "step": 7860,
      "time_per_iteration": 3.5007190704345703
    },
    {
      "auxiliary_loss_clip": 0.01175207,
      "auxiliary_loss_mlp": 0.01019648,
      "balance_loss_clip": 1.04277647,
      "balance_loss_mlp": 1.01220298,
      "epoch": 0.9452293633138941,
      "flos": 25448097029760.0,
      "grad_norm": 2.3859843007064288,
      "language_loss": 0.68299663,
      "learning_rate": 3.132342202091554e-08,
      "loss": 0.70494509,
      "num_input_tokens_seen": 169909280,
      "step": 7861,
      "time_per_iteration": 2.6983447074890137
    },
    {
      "auxiliary_loss_clip": 0.01167392,
      "auxiliary_loss_mlp": 0.010264,
      "balance_loss_clip": 1.04724956,
      "balance_loss_mlp": 1.01900935,
      "epoch": 0.9453496062045331,
      "flos": 21215342350080.0,
      "grad_norm": 2.214597004376131,
      "language_loss": 0.68716371,
      "learning_rate": 3.1186246505107595e-08,
      "loss": 0.70910162,
      "num_input_tokens_seen": 169928420,
      "step": 7862,
      "time_per_iteration": 2.5619142055511475
    },
    {
      "auxiliary_loss_clip": 0.01170325,
      "auxiliary_loss_mlp": 0.0102421,
      "balance_loss_clip": 1.04717493,
      "balance_loss_mlp": 1.01645589,
      "epoch": 0.9454698490951723,
      "flos": 20010898477440.0,
      "grad_norm": 2.1824827467914636,
      "language_loss": 0.83673131,
      "learning_rate": 3.104936965678084e-08,
      "loss": 0.85867661,
      "num_input_tokens_seen": 169946750,
      "step": 7863,
      "time_per_iteration": 2.5627267360687256
    },
    {
      "auxiliary_loss_clip": 0.01168233,
      "auxiliary_loss_mlp": 0.01020283,
      "balance_loss_clip": 1.04440498,
      "balance_loss_mlp": 1.01277256,
      "epoch": 0.9455900919858113,
      "flos": 21069652786560.0,
      "grad_norm": 2.082964871823296,
      "language_loss": 0.81916499,
      "learning_rate": 3.091279149669956e-08,
      "loss": 0.84105021,
      "num_input_tokens_seen": 169965540,
      "step": 7864,
      "time_per_iteration": 2.517472505569458
    },
    {
      "auxiliary_loss_clip": 0.01168071,
      "auxiliary_loss_mlp": 0.01207355,
      "balance_loss_clip": 1.04592228,
      "balance_loss_mlp": 1.00075257,
      "epoch": 0.9457103348764504,
      "flos": 20740854666240.0,
      "grad_norm": 2.0403699058982525,
      "language_loss": 0.73942697,
      "learning_rate": 3.0776512045581624e-08,
      "loss": 0.76318127,
      "num_input_tokens_seen": 169984330,
      "step": 7865,
      "time_per_iteration": 2.6297202110290527
    },
    {
      "auxiliary_loss_clip": 0.01163979,
      "auxiliary_loss_mlp": 0.01026695,
      "balance_loss_clip": 1.04306507,
      "balance_loss_mlp": 1.01882124,
      "epoch": 0.9458305777670896,
      "flos": 21428363957760.0,
      "grad_norm": 1.8802110247985695,
      "language_loss": 0.77590793,
      "learning_rate": 3.0640531324101384e-08,
      "loss": 0.79781473,
      "num_input_tokens_seen": 170002095,
      "step": 7866,
      "time_per_iteration": 2.6153340339660645
    },
    {
      "auxiliary_loss_clip": 0.01173036,
      "auxiliary_loss_mlp": 0.0102472,
      "balance_loss_clip": 1.04912448,
      "balance_loss_mlp": 1.01671469,
      "epoch": 0.9459508206577286,
      "flos": 20011185786240.0,
      "grad_norm": 1.9127180135914061,
      "language_loss": 0.76576215,
      "learning_rate": 3.0504849352886554e-08,
      "loss": 0.78773963,
      "num_input_tokens_seen": 170020240,
      "step": 7867,
      "time_per_iteration": 2.61962628364563
    },
    {
      "auxiliary_loss_clip": 0.01167618,
      "auxiliary_loss_mlp": 0.01022074,
      "balance_loss_clip": 1.04589474,
      "balance_loss_mlp": 1.01508832,
      "epoch": 0.9460710635483677,
      "flos": 12166428291840.0,
      "grad_norm": 4.2293606503786885,
      "language_loss": 0.71681976,
      "learning_rate": 3.036946615252023e-08,
      "loss": 0.73871672,
      "num_input_tokens_seen": 170035770,
      "step": 7868,
      "time_per_iteration": 2.5016164779663086
    },
    {
      "auxiliary_loss_clip": 0.01174965,
      "auxiliary_loss_mlp": 0.01025903,
      "balance_loss_clip": 1.04439759,
      "balance_loss_mlp": 1.018435,
      "epoch": 0.9461913064390068,
      "flos": 34276196229120.0,
      "grad_norm": 3.2511657992913965,
      "language_loss": 0.67017144,
      "learning_rate": 3.0234381743539984e-08,
      "loss": 0.69218016,
      "num_input_tokens_seen": 170053385,
      "step": 7869,
      "time_per_iteration": 2.734100341796875
    },
    {
      "auxiliary_loss_clip": 0.01177017,
      "auxiliary_loss_mlp": 0.01024681,
      "balance_loss_clip": 1.04427552,
      "balance_loss_mlp": 1.01732612,
      "epoch": 0.9463115493296459,
      "flos": 19463763536640.0,
      "grad_norm": 2.290918529166136,
      "language_loss": 0.80546337,
      "learning_rate": 3.0099596146437863e-08,
      "loss": 0.82748032,
      "num_input_tokens_seen": 170070490,
      "step": 7870,
      "time_per_iteration": 2.60395884513855
    },
    {
      "auxiliary_loss_clip": 0.01062374,
      "auxiliary_loss_mlp": 0.01001096,
      "balance_loss_clip": 1.00997078,
      "balance_loss_mlp": 1.00037515,
      "epoch": 0.946431792220285,
      "flos": 70570824387840.0,
      "grad_norm": 0.7753231160567351,
      "language_loss": 0.60024136,
      "learning_rate": 2.996510938166086e-08,
      "loss": 0.62087607,
      "num_input_tokens_seen": 170133465,
      "step": 7871,
      "time_per_iteration": 4.179561138153076
    },
    {
      "auxiliary_loss_clip": 0.01167051,
      "auxiliary_loss_mlp": 0.01022542,
      "balance_loss_clip": 1.04748225,
      "balance_loss_mlp": 1.01563096,
      "epoch": 0.9465520351109241,
      "flos": 18947906363520.0,
      "grad_norm": 1.9744067501070521,
      "language_loss": 0.73629606,
      "learning_rate": 2.983092146960997e-08,
      "loss": 0.758192,
      "num_input_tokens_seen": 170150810,
      "step": 7872,
      "time_per_iteration": 2.651447057723999
    },
    {
      "auxiliary_loss_clip": 0.01172704,
      "auxiliary_loss_mlp": 0.0103116,
      "balance_loss_clip": 1.04299879,
      "balance_loss_mlp": 1.02279758,
      "epoch": 0.9466722780015632,
      "flos": 19135647774720.0,
      "grad_norm": 3.541309007495558,
      "language_loss": 0.80033374,
      "learning_rate": 2.9697032430642256e-08,
      "loss": 0.82237244,
      "num_input_tokens_seen": 170169025,
      "step": 7873,
      "time_per_iteration": 2.615567207336426
    },
    {
      "auxiliary_loss_clip": 0.01164497,
      "auxiliary_loss_mlp": 0.01021959,
      "balance_loss_clip": 1.04748917,
      "balance_loss_mlp": 1.01520014,
      "epoch": 0.9467925208922022,
      "flos": 17237912520960.0,
      "grad_norm": 2.6094179958559454,
      "language_loss": 0.73517287,
      "learning_rate": 2.9563442285067906e-08,
      "loss": 0.75703746,
      "num_input_tokens_seen": 170186070,
      "step": 7874,
      "time_per_iteration": 2.615297317504883
    },
    {
      "auxiliary_loss_clip": 0.01171481,
      "auxiliary_loss_mlp": 0.01025534,
      "balance_loss_clip": 1.04679346,
      "balance_loss_mlp": 1.01788688,
      "epoch": 0.9469127637828414,
      "flos": 29169016859520.0,
      "grad_norm": 3.584976190732332,
      "language_loss": 0.79697263,
      "learning_rate": 2.943015105315294e-08,
      "loss": 0.81894279,
      "num_input_tokens_seen": 170206265,
      "step": 7875,
      "time_per_iteration": 2.5778422355651855
    },
    {
      "auxiliary_loss_clip": 0.01177441,
      "auxiliary_loss_mlp": 0.01022522,
      "balance_loss_clip": 1.03935456,
      "balance_loss_mlp": 1.01424313,
      "epoch": 0.9470330066734804,
      "flos": 26030460234240.0,
      "grad_norm": 2.529133230395744,
      "language_loss": 0.66902161,
      "learning_rate": 2.929715875511718e-08,
      "loss": 0.6910212,
      "num_input_tokens_seen": 170225300,
      "step": 7876,
      "time_per_iteration": 2.6727187633514404
    },
    {
      "auxiliary_loss_clip": 0.01168953,
      "auxiliary_loss_mlp": 0.01022084,
      "balance_loss_clip": 1.04255641,
      "balance_loss_mlp": 1.01463342,
      "epoch": 0.9471532495641195,
      "flos": 23440906056960.0,
      "grad_norm": 1.829254048687758,
      "language_loss": 0.70147562,
      "learning_rate": 2.9164465411135375e-08,
      "loss": 0.72338605,
      "num_input_tokens_seen": 170245070,
      "step": 7877,
      "time_per_iteration": 2.5786263942718506
    },
    {
      "auxiliary_loss_clip": 0.01172068,
      "auxiliary_loss_mlp": 0.01023655,
      "balance_loss_clip": 1.048558,
      "balance_loss_mlp": 1.0164609,
      "epoch": 0.9472734924547586,
      "flos": 15815850099840.0,
      "grad_norm": 1.9643127754484604,
      "language_loss": 0.80835891,
      "learning_rate": 2.9032071041337426e-08,
      "loss": 0.83031607,
      "num_input_tokens_seen": 170263305,
      "step": 7878,
      "time_per_iteration": 2.6287293434143066
    },
    {
      "auxiliary_loss_clip": 0.01162964,
      "auxiliary_loss_mlp": 0.01026744,
      "balance_loss_clip": 1.04284716,
      "balance_loss_mlp": 1.01977301,
      "epoch": 0.9473937353453977,
      "flos": 11181793697280.0,
      "grad_norm": 1.6290964359344318,
      "language_loss": 0.73103082,
      "learning_rate": 2.889997566580704e-08,
      "loss": 0.7529279,
      "num_input_tokens_seen": 170281460,
      "step": 7879,
      "time_per_iteration": 2.5892598628997803
    },
    {
      "auxiliary_loss_clip": 0.01167721,
      "auxiliary_loss_mlp": 0.01025721,
      "balance_loss_clip": 1.04708254,
      "balance_loss_mlp": 1.01773405,
      "epoch": 0.9475139782360368,
      "flos": 25775530433280.0,
      "grad_norm": 1.8608034241662517,
      "language_loss": 0.70296657,
      "learning_rate": 2.8768179304583086e-08,
      "loss": 0.72490096,
      "num_input_tokens_seen": 170303515,
      "step": 7880,
      "time_per_iteration": 3.506901502609253
    },
    {
      "auxiliary_loss_clip": 0.01173344,
      "auxiliary_loss_mlp": 0.01027188,
      "balance_loss_clip": 1.04395831,
      "balance_loss_mlp": 1.02018142,
      "epoch": 0.9476342211266758,
      "flos": 22820046451200.0,
      "grad_norm": 1.7928514658966583,
      "language_loss": 0.73724675,
      "learning_rate": 2.8636681977659117e-08,
      "loss": 0.75925207,
      "num_input_tokens_seen": 170323165,
      "step": 7881,
      "time_per_iteration": 2.6854703426361084
    },
    {
      "auxiliary_loss_clip": 0.01176949,
      "auxiliary_loss_mlp": 0.01026535,
      "balance_loss_clip": 1.04513621,
      "balance_loss_mlp": 1.01915574,
      "epoch": 0.947754464017315,
      "flos": 20193611984640.0,
      "grad_norm": 2.1855885820154084,
      "language_loss": 0.77867377,
      "learning_rate": 2.850548370498318e-08,
      "loss": 0.80070865,
      "num_input_tokens_seen": 170341005,
      "step": 7882,
      "time_per_iteration": 2.699801445007324
    },
    {
      "auxiliary_loss_clip": 0.0116544,
      "auxiliary_loss_mlp": 0.01020961,
      "balance_loss_clip": 1.04222262,
      "balance_loss_mlp": 1.01428509,
      "epoch": 0.9478747069079541,
      "flos": 24717925359360.0,
      "grad_norm": 1.5436696985277516,
      "language_loss": 0.71395016,
      "learning_rate": 2.8374584506457798e-08,
      "loss": 0.73581409,
      "num_input_tokens_seen": 170362280,
      "step": 7883,
      "time_per_iteration": 3.571105480194092
    },
    {
      "auxiliary_loss_clip": 0.01170707,
      "auxiliary_loss_mlp": 0.0102152,
      "balance_loss_clip": 1.04599953,
      "balance_loss_mlp": 1.01372957,
      "epoch": 0.9479949497985931,
      "flos": 21361355136000.0,
      "grad_norm": 2.376294301871295,
      "language_loss": 0.67448097,
      "learning_rate": 2.824398440193998e-08,
      "loss": 0.69640326,
      "num_input_tokens_seen": 170381080,
      "step": 7884,
      "time_per_iteration": 2.665313959121704
    },
    {
      "auxiliary_loss_clip": 0.01172106,
      "auxiliary_loss_mlp": 0.01024786,
      "balance_loss_clip": 1.0418551,
      "balance_loss_mlp": 1.01678729,
      "epoch": 0.9481151926892323,
      "flos": 18148606968960.0,
      "grad_norm": 3.2241615479621686,
      "language_loss": 0.7138468,
      "learning_rate": 2.811368341124232e-08,
      "loss": 0.7358157,
      "num_input_tokens_seen": 170400150,
      "step": 7885,
      "time_per_iteration": 2.7016994953155518
    },
    {
      "auxiliary_loss_clip": 0.01168743,
      "auxiliary_loss_mlp": 0.01028404,
      "balance_loss_clip": 1.0435605,
      "balance_loss_mlp": 1.02132869,
      "epoch": 0.9482354355798713,
      "flos": 22128012046080.0,
      "grad_norm": 2.9999219231083356,
      "language_loss": 0.68593037,
      "learning_rate": 2.7983681554131222e-08,
      "loss": 0.70790178,
      "num_input_tokens_seen": 170420410,
      "step": 7886,
      "time_per_iteration": 2.6470658779144287
    },
    {
      "auxiliary_loss_clip": 0.01169873,
      "auxiliary_loss_mlp": 0.01024729,
      "balance_loss_clip": 1.04343617,
      "balance_loss_mlp": 1.01721263,
      "epoch": 0.9483556784705104,
      "flos": 19063072344960.0,
      "grad_norm": 2.2690356226141426,
      "language_loss": 0.70247221,
      "learning_rate": 2.7853978850327365e-08,
      "loss": 0.72441822,
      "num_input_tokens_seen": 170439580,
      "step": 7887,
      "time_per_iteration": 3.537977457046509
    },
    {
      "auxiliary_loss_clip": 0.01172388,
      "auxiliary_loss_mlp": 0.01025006,
      "balance_loss_clip": 1.04605842,
      "balance_loss_mlp": 1.01768374,
      "epoch": 0.9484759213611496,
      "flos": 25777110631680.0,
      "grad_norm": 1.8813776069724537,
      "language_loss": 0.87139809,
      "learning_rate": 2.7724575319507225e-08,
      "loss": 0.89337206,
      "num_input_tokens_seen": 170459290,
      "step": 7888,
      "time_per_iteration": 2.685551166534424
    },
    {
      "auxiliary_loss_clip": 0.01166264,
      "auxiliary_loss_mlp": 0.01021684,
      "balance_loss_clip": 1.04274964,
      "balance_loss_mlp": 1.01465702,
      "epoch": 0.9485961642517886,
      "flos": 20667740532480.0,
      "grad_norm": 1.8071402466114304,
      "language_loss": 0.77333498,
      "learning_rate": 2.759547098130044e-08,
      "loss": 0.79521441,
      "num_input_tokens_seen": 170478020,
      "step": 7889,
      "time_per_iteration": 2.6746318340301514
    },
    {
      "auxiliary_loss_clip": 0.01162606,
      "auxiliary_loss_mlp": 0.0102345,
      "balance_loss_clip": 1.04606056,
      "balance_loss_mlp": 1.01626778,
      "epoch": 0.9487164071424277,
      "flos": 22674069578880.0,
      "grad_norm": 1.9263917187867678,
      "language_loss": 0.76581192,
      "learning_rate": 2.746666585529267e-08,
      "loss": 0.78767252,
      "num_input_tokens_seen": 170498295,
      "step": 7890,
      "time_per_iteration": 2.5719335079193115
    },
    {
      "auxiliary_loss_clip": 0.01159706,
      "auxiliary_loss_mlp": 0.01022477,
      "balance_loss_clip": 1.04298103,
      "balance_loss_mlp": 1.0149014,
      "epoch": 0.9488366500330668,
      "flos": 38726461716480.0,
      "grad_norm": 3.0397659016036527,
      "language_loss": 0.74745226,
      "learning_rate": 2.73381599610234e-08,
      "loss": 0.76927412,
      "num_input_tokens_seen": 170518695,
      "step": 7891,
      "time_per_iteration": 2.722548484802246
    },
    {
      "auxiliary_loss_clip": 0.0116144,
      "auxiliary_loss_mlp": 0.01024239,
      "balance_loss_clip": 1.04154181,
      "balance_loss_mlp": 1.01641917,
      "epoch": 0.9489568929237059,
      "flos": 27890920149120.0,
      "grad_norm": 36.46364668040626,
      "language_loss": 0.7142874,
      "learning_rate": 2.7209953317987033e-08,
      "loss": 0.73614424,
      "num_input_tokens_seen": 170539735,
      "step": 7892,
      "time_per_iteration": 2.71211838722229
    },
    {
      "auxiliary_loss_clip": 0.01170153,
      "auxiliary_loss_mlp": 0.01023655,
      "balance_loss_clip": 1.04709804,
      "balance_loss_mlp": 1.01628208,
      "epoch": 0.9490771358143449,
      "flos": 33580642291200.0,
      "grad_norm": 2.6472179886892464,
      "language_loss": 0.78330141,
      "learning_rate": 2.7082045945631793e-08,
      "loss": 0.8052395,
      "num_input_tokens_seen": 170561950,
      "step": 7893,
      "time_per_iteration": 2.7717089653015137
    },
    {
      "auxiliary_loss_clip": 0.01166538,
      "auxiliary_loss_mlp": 0.0102406,
      "balance_loss_clip": 1.04184127,
      "balance_loss_mlp": 1.01650822,
      "epoch": 0.9491973787049841,
      "flos": 14793796512000.0,
      "grad_norm": 2.1959666119201016,
      "language_loss": 0.69512373,
      "learning_rate": 2.6954437863361712e-08,
      "loss": 0.71702969,
      "num_input_tokens_seen": 170579865,
      "step": 7894,
      "time_per_iteration": 2.7156407833099365
    },
    {
      "auxiliary_loss_clip": 0.01181073,
      "auxiliary_loss_mlp": 0.0102167,
      "balance_loss_clip": 1.04035294,
      "balance_loss_mlp": 1.01488733,
      "epoch": 0.9493176215956232,
      "flos": 25332535998720.0,
      "grad_norm": 2.3718694274967076,
      "language_loss": 0.70831609,
      "learning_rate": 2.6827129090534862e-08,
      "loss": 0.73034358,
      "num_input_tokens_seen": 170600165,
      "step": 7895,
      "time_per_iteration": 2.722184896469116
    },
    {
      "auxiliary_loss_clip": 0.01166427,
      "auxiliary_loss_mlp": 0.01024686,
      "balance_loss_clip": 1.0434103,
      "balance_loss_mlp": 1.01742077,
      "epoch": 0.9494378644862622,
      "flos": 21029971236480.0,
      "grad_norm": 1.9220533357751746,
      "language_loss": 0.78114796,
      "learning_rate": 2.670011964646335e-08,
      "loss": 0.8030591,
      "num_input_tokens_seen": 170618845,
      "step": 7896,
      "time_per_iteration": 2.643321990966797
    },
    {
      "auxiliary_loss_clip": 0.0118422,
      "auxiliary_loss_mlp": 0.01024582,
      "balance_loss_clip": 1.03327394,
      "balance_loss_mlp": 1.01693797,
      "epoch": 0.9495581073769014,
      "flos": 15195134148480.0,
      "grad_norm": 2.208899221502293,
      "language_loss": 0.68019587,
      "learning_rate": 2.657340955041487e-08,
      "loss": 0.70228392,
      "num_input_tokens_seen": 170637620,
      "step": 7897,
      "time_per_iteration": 2.7781119346618652
    },
    {
      "auxiliary_loss_clip": 0.01170604,
      "auxiliary_loss_mlp": 0.01023346,
      "balance_loss_clip": 1.04627752,
      "balance_loss_mlp": 1.01550806,
      "epoch": 0.9496783502675404,
      "flos": 28616566705920.0,
      "grad_norm": 6.875845607181666,
      "language_loss": 0.71692753,
      "learning_rate": 2.6446998821611167e-08,
      "loss": 0.73886704,
      "num_input_tokens_seen": 170657815,
      "step": 7898,
      "time_per_iteration": 3.700801372528076
    },
    {
      "auxiliary_loss_clip": 0.01175983,
      "auxiliary_loss_mlp": 0.01025651,
      "balance_loss_clip": 1.04013872,
      "balance_loss_mlp": 1.01821852,
      "epoch": 0.9497985931581795,
      "flos": 14866874732160.0,
      "grad_norm": 2.8547151743016945,
      "language_loss": 0.71837473,
      "learning_rate": 2.6320887479228228e-08,
      "loss": 0.74039108,
      "num_input_tokens_seen": 170674415,
      "step": 7899,
      "time_per_iteration": 2.6788957118988037
    },
    {
      "auxiliary_loss_clip": 0.01174669,
      "auxiliary_loss_mlp": 0.0102823,
      "balance_loss_clip": 1.04440916,
      "balance_loss_mlp": 1.02083945,
      "epoch": 0.9499188360488187,
      "flos": 27193319136000.0,
      "grad_norm": 2.6447086920415126,
      "language_loss": 0.7248317,
      "learning_rate": 2.619507554239786e-08,
      "loss": 0.74686074,
      "num_input_tokens_seen": 170692975,
      "step": 7900,
      "time_per_iteration": 2.725785732269287
    },
    {
      "auxiliary_loss_clip": 0.0116856,
      "auxiliary_loss_mlp": 0.01026931,
      "balance_loss_clip": 1.04289794,
      "balance_loss_mlp": 1.01939142,
      "epoch": 0.9500390789394577,
      "flos": 24316479982080.0,
      "grad_norm": 2.0096888836919167,
      "language_loss": 0.70246023,
      "learning_rate": 2.606956303020502e-08,
      "loss": 0.72441518,
      "num_input_tokens_seen": 170713780,
      "step": 7901,
      "time_per_iteration": 2.7436485290527344
    },
    {
      "auxiliary_loss_clip": 0.01169275,
      "auxiliary_loss_mlp": 0.01023014,
      "balance_loss_clip": 1.04727745,
      "balance_loss_mlp": 1.01545656,
      "epoch": 0.9501593218300968,
      "flos": 14354752573440.0,
      "grad_norm": 1.6001186017049156,
      "language_loss": 0.84068674,
      "learning_rate": 2.5944349961690036e-08,
      "loss": 0.86260962,
      "num_input_tokens_seen": 170730800,
      "step": 7902,
      "time_per_iteration": 2.561689853668213
    },
    {
      "auxiliary_loss_clip": 0.01172089,
      "auxiliary_loss_mlp": 0.01023243,
      "balance_loss_clip": 1.04211009,
      "balance_loss_mlp": 1.01580405,
      "epoch": 0.9502795647207359,
      "flos": 38728113742080.0,
      "grad_norm": 1.648244305178225,
      "language_loss": 0.73407704,
      "learning_rate": 2.581943635584749e-08,
      "loss": 0.75603032,
      "num_input_tokens_seen": 170753630,
      "step": 7903,
      "time_per_iteration": 2.782604694366455
    },
    {
      "auxiliary_loss_clip": 0.01163202,
      "auxiliary_loss_mlp": 0.01020974,
      "balance_loss_clip": 1.04312193,
      "balance_loss_mlp": 1.01417327,
      "epoch": 0.950399807611375,
      "flos": 40808023799040.0,
      "grad_norm": 1.5879214913880944,
      "language_loss": 0.65492046,
      "learning_rate": 2.569482223162689e-08,
      "loss": 0.67676228,
      "num_input_tokens_seen": 170777605,
      "step": 7904,
      "time_per_iteration": 2.7774460315704346
    },
    {
      "auxiliary_loss_clip": 0.01169054,
      "auxiliary_loss_mlp": 0.01021815,
      "balance_loss_clip": 1.04426742,
      "balance_loss_mlp": 1.01397097,
      "epoch": 0.950520050502014,
      "flos": 23440403266560.0,
      "grad_norm": 4.074150648911905,
      "language_loss": 0.72405648,
      "learning_rate": 2.5570507607932e-08,
      "loss": 0.74596512,
      "num_input_tokens_seen": 170797520,
      "step": 7905,
      "time_per_iteration": 2.5997133255004883
    },
    {
      "auxiliary_loss_clip": 0.01174009,
      "auxiliary_loss_mlp": 0.01023204,
      "balance_loss_clip": 1.04575777,
      "balance_loss_mlp": 1.01568222,
      "epoch": 0.9506402933926532,
      "flos": 17783718658560.0,
      "grad_norm": 2.736954901669067,
      "language_loss": 0.64219743,
      "learning_rate": 2.54464925036213e-08,
      "loss": 0.66416955,
      "num_input_tokens_seen": 170814810,
      "step": 7906,
      "time_per_iteration": 3.445189952850342
    },
    {
      "auxiliary_loss_clip": 0.01169343,
      "auxiliary_loss_mlp": 0.01023989,
      "balance_loss_clip": 1.04658806,
      "balance_loss_mlp": 1.01600766,
      "epoch": 0.9507605362832923,
      "flos": 32561928668160.0,
      "grad_norm": 2.075668939582419,
      "language_loss": 0.60941398,
      "learning_rate": 2.532277693750773e-08,
      "loss": 0.6313473,
      "num_input_tokens_seen": 170835735,
      "step": 7907,
      "time_per_iteration": 2.7177345752716064
    },
    {
      "auxiliary_loss_clip": 0.01172127,
      "auxiliary_loss_mlp": 0.0102465,
      "balance_loss_clip": 1.0434587,
      "balance_loss_mlp": 1.01728332,
      "epoch": 0.9508807791739313,
      "flos": 19602054898560.0,
      "grad_norm": 2.009212039278828,
      "language_loss": 0.76067656,
      "learning_rate": 2.5199360928358948e-08,
      "loss": 0.78264433,
      "num_input_tokens_seen": 170852970,
      "step": 7908,
      "time_per_iteration": 2.658750057220459
    },
    {
      "auxiliary_loss_clip": 0.01158099,
      "auxiliary_loss_mlp": 0.01207236,
      "balance_loss_clip": 1.04252672,
      "balance_loss_mlp": 1.00075006,
      "epoch": 0.9510010220645704,
      "flos": 21471852349440.0,
      "grad_norm": 1.659051629682655,
      "language_loss": 0.87363362,
      "learning_rate": 2.507624449489665e-08,
      "loss": 0.89728701,
      "num_input_tokens_seen": 170871600,
      "step": 7909,
      "time_per_iteration": 2.6344523429870605
    },
    {
      "auxiliary_loss_clip": 0.01171488,
      "auxiliary_loss_mlp": 0.01025868,
      "balance_loss_clip": 1.04594624,
      "balance_loss_mlp": 1.01819694,
      "epoch": 0.9511212649552095,
      "flos": 18879999701760.0,
      "grad_norm": 2.2687078438104273,
      "language_loss": 0.64865679,
      "learning_rate": 2.495342765579811e-08,
      "loss": 0.67063034,
      "num_input_tokens_seen": 170890260,
      "step": 7910,
      "time_per_iteration": 3.538919687271118
    },
    {
      "auxiliary_loss_clip": 0.01172448,
      "auxiliary_loss_mlp": 0.01020482,
      "balance_loss_clip": 1.04307425,
      "balance_loss_mlp": 1.01365685,
      "epoch": 0.9512415078458486,
      "flos": 20810521094400.0,
      "grad_norm": 1.8610293113832261,
      "language_loss": 0.71258485,
      "learning_rate": 2.4830910429693984e-08,
      "loss": 0.73451418,
      "num_input_tokens_seen": 170910220,
      "step": 7911,
      "time_per_iteration": 2.7046804428100586
    },
    {
      "auxiliary_loss_clip": 0.0116454,
      "auxiliary_loss_mlp": 0.01026309,
      "balance_loss_clip": 1.04540706,
      "balance_loss_mlp": 1.01902843,
      "epoch": 0.9513617507364877,
      "flos": 18369565482240.0,
      "grad_norm": 2.0929376681312446,
      "language_loss": 0.79924607,
      "learning_rate": 2.470869283517052e-08,
      "loss": 0.82115459,
      "num_input_tokens_seen": 170928255,
      "step": 7912,
      "time_per_iteration": 2.585399627685547
    },
    {
      "auxiliary_loss_clip": 0.0116133,
      "auxiliary_loss_mlp": 0.01024862,
      "balance_loss_clip": 1.04286051,
      "balance_loss_mlp": 1.01771879,
      "epoch": 0.9514819936271268,
      "flos": 25010166412800.0,
      "grad_norm": 1.6309489448086518,
      "language_loss": 0.77337801,
      "learning_rate": 2.458677489076777e-08,
      "loss": 0.79523993,
      "num_input_tokens_seen": 170949265,
      "step": 7913,
      "time_per_iteration": 3.567399740219116
    },
    {
      "auxiliary_loss_clip": 0.01157808,
      "auxiliary_loss_mlp": 0.01026373,
      "balance_loss_clip": 1.04254508,
      "balance_loss_mlp": 1.01949477,
      "epoch": 0.9516022365177659,
      "flos": 18662129758080.0,
      "grad_norm": 1.7059992334804557,
      "language_loss": 0.83187354,
      "learning_rate": 2.446515661498072e-08,
      "loss": 0.8537153,
      "num_input_tokens_seen": 170968595,
      "step": 7914,
      "time_per_iteration": 2.611593246459961
    },
    {
      "auxiliary_loss_clip": 0.01174139,
      "auxiliary_loss_mlp": 0.01026524,
      "balance_loss_clip": 1.03875637,
      "balance_loss_mlp": 1.01936889,
      "epoch": 0.9517224794084049,
      "flos": 25372109808000.0,
      "grad_norm": 3.8339917888980475,
      "language_loss": 0.74506748,
      "learning_rate": 2.434383802625861e-08,
      "loss": 0.76707417,
      "num_input_tokens_seen": 170987550,
      "step": 7915,
      "time_per_iteration": 2.7444705963134766
    },
    {
      "auxiliary_loss_clip": 0.01172125,
      "auxiliary_loss_mlp": 0.01024334,
      "balance_loss_clip": 1.03974199,
      "balance_loss_mlp": 1.0171721,
      "epoch": 0.9518427222990441,
      "flos": 21470918595840.0,
      "grad_norm": 1.8836192835306216,
      "language_loss": 0.73865843,
      "learning_rate": 2.4222819143005168e-08,
      "loss": 0.76062298,
      "num_input_tokens_seen": 171007145,
      "step": 7916,
      "time_per_iteration": 2.633146047592163
    },
    {
      "auxiliary_loss_clip": 0.01164437,
      "auxiliary_loss_mlp": 0.01023266,
      "balance_loss_clip": 1.04755068,
      "balance_loss_mlp": 1.01587152,
      "epoch": 0.9519629651896832,
      "flos": 21033634423680.0,
      "grad_norm": 1.910004979645647,
      "language_loss": 0.81024921,
      "learning_rate": 2.4102099983579706e-08,
      "loss": 0.8321262,
      "num_input_tokens_seen": 171026295,
      "step": 7917,
      "time_per_iteration": 2.6443326473236084
    },
    {
      "auxiliary_loss_clip": 0.01168095,
      "auxiliary_loss_mlp": 0.01027127,
      "balance_loss_clip": 1.04486883,
      "balance_loss_mlp": 1.01905036,
      "epoch": 0.9520832080803222,
      "flos": 21689219502720.0,
      "grad_norm": 1.8033019979685794,
      "language_loss": 0.77363801,
      "learning_rate": 2.3981680566294236e-08,
      "loss": 0.79559028,
      "num_input_tokens_seen": 171045895,
      "step": 7918,
      "time_per_iteration": 2.6094439029693604
    },
    {
      "auxiliary_loss_clip": 0.01164074,
      "auxiliary_loss_mlp": 0.01024585,
      "balance_loss_clip": 1.04757488,
      "balance_loss_mlp": 1.01779628,
      "epoch": 0.9522034509709614,
      "flos": 23145289125120.0,
      "grad_norm": 1.8581725700797806,
      "language_loss": 0.73797673,
      "learning_rate": 2.3861560909416822e-08,
      "loss": 0.75986326,
      "num_input_tokens_seen": 171065445,
      "step": 7919,
      "time_per_iteration": 2.5983479022979736
    },
    {
      "auxiliary_loss_clip": 0.01179809,
      "auxiliary_loss_mlp": 0.01025907,
      "balance_loss_clip": 1.04471922,
      "balance_loss_mlp": 1.01867652,
      "epoch": 0.9523236938616004,
      "flos": 24679428958080.0,
      "grad_norm": 1.776675950687398,
      "language_loss": 0.82519031,
      "learning_rate": 2.3741741031169325e-08,
      "loss": 0.84724748,
      "num_input_tokens_seen": 171085015,
      "step": 7920,
      "time_per_iteration": 2.8335559368133545
    },
    {
      "auxiliary_loss_clip": 0.01171677,
      "auxiliary_loss_mlp": 0.01023334,
      "balance_loss_clip": 1.03943074,
      "balance_loss_mlp": 1.01647639,
      "epoch": 0.9524439367522395,
      "flos": 22672309812480.0,
      "grad_norm": 2.160229420480693,
      "language_loss": 0.72039473,
      "learning_rate": 2.3622220949728544e-08,
      "loss": 0.74234486,
      "num_input_tokens_seen": 171103900,
      "step": 7921,
      "time_per_iteration": 2.7090301513671875
    },
    {
      "auxiliary_loss_clip": 0.0116245,
      "auxiliary_loss_mlp": 0.01026324,
      "balance_loss_clip": 1.04495358,
      "balance_loss_mlp": 1.01816428,
      "epoch": 0.9525641796428787,
      "flos": 34055525024640.0,
      "grad_norm": 2.8043245807099764,
      "language_loss": 0.61479831,
      "learning_rate": 2.3503000683225526e-08,
      "loss": 0.63668603,
      "num_input_tokens_seen": 171121615,
      "step": 7922,
      "time_per_iteration": 2.7858338356018066
    },
    {
      "auxiliary_loss_clip": 0.01168088,
      "auxiliary_loss_mlp": 0.01024565,
      "balance_loss_clip": 1.04735017,
      "balance_loss_mlp": 1.017097,
      "epoch": 0.9526844225335177,
      "flos": 16727083251840.0,
      "grad_norm": 3.3360019898923237,
      "language_loss": 0.84603876,
      "learning_rate": 2.3384080249745585e-08,
      "loss": 0.86796522,
      "num_input_tokens_seen": 171139505,
      "step": 7923,
      "time_per_iteration": 2.683574676513672
    },
    {
      "auxiliary_loss_clip": 0.01180278,
      "auxiliary_loss_mlp": 0.01022555,
      "balance_loss_clip": 1.04165733,
      "balance_loss_mlp": 1.01582909,
      "epoch": 0.9528046654241568,
      "flos": 36939367330560.0,
      "grad_norm": 2.2407083845268416,
      "language_loss": 0.83389318,
      "learning_rate": 2.3265459667329178e-08,
      "loss": 0.85592151,
      "num_input_tokens_seen": 171158995,
      "step": 7924,
      "time_per_iteration": 2.845165729522705
    },
    {
      "auxiliary_loss_clip": 0.01173527,
      "auxiliary_loss_mlp": 0.01023199,
      "balance_loss_clip": 1.04480791,
      "balance_loss_mlp": 1.01550746,
      "epoch": 0.9529249083147959,
      "flos": 18255010032000.0,
      "grad_norm": 2.2278364500576666,
      "language_loss": 0.8602376,
      "learning_rate": 2.31471389539708e-08,
      "loss": 0.88220483,
      "num_input_tokens_seen": 171176120,
      "step": 7925,
      "time_per_iteration": 3.5848541259765625
    },
    {
      "auxiliary_loss_clip": 0.01169446,
      "auxiliary_loss_mlp": 0.01207233,
      "balance_loss_clip": 1.04636025,
      "balance_loss_mlp": 1.00074196,
      "epoch": 0.953045151205435,
      "flos": 28658438985600.0,
      "grad_norm": 2.213432007858598,
      "language_loss": 0.73433888,
      "learning_rate": 2.3029118127619872e-08,
      "loss": 0.75810564,
      "num_input_tokens_seen": 171195835,
      "step": 7926,
      "time_per_iteration": 2.68949818611145
    },
    {
      "auxiliary_loss_clip": 0.01163732,
      "auxiliary_loss_mlp": 0.01023297,
      "balance_loss_clip": 1.04313564,
      "balance_loss_mlp": 1.01576889,
      "epoch": 0.953165394096074,
      "flos": 21835232288640.0,
      "grad_norm": 2.404844390674175,
      "language_loss": 0.86964178,
      "learning_rate": 2.2911397206179628e-08,
      "loss": 0.89151204,
      "num_input_tokens_seen": 171212585,
      "step": 7927,
      "time_per_iteration": 2.6599538326263428
    },
    {
      "auxiliary_loss_clip": 0.01162609,
      "auxiliary_loss_mlp": 0.01023885,
      "balance_loss_clip": 1.04583371,
      "balance_loss_mlp": 1.01693213,
      "epoch": 0.9532856369867132,
      "flos": 19975059682560.0,
      "grad_norm": 3.0950731769263076,
      "language_loss": 0.6299662,
      "learning_rate": 2.279397620750845e-08,
      "loss": 0.65183115,
      "num_input_tokens_seen": 171231630,
      "step": 7928,
      "time_per_iteration": 2.662994146347046
    },
    {
      "auxiliary_loss_clip": 0.01167912,
      "auxiliary_loss_mlp": 0.01022916,
      "balance_loss_clip": 1.04239511,
      "balance_loss_mlp": 1.01628804,
      "epoch": 0.9534058798773523,
      "flos": 15049588239360.0,
      "grad_norm": 4.187079239377457,
      "language_loss": 0.78772509,
      "learning_rate": 2.2676855149419195e-08,
      "loss": 0.80963337,
      "num_input_tokens_seen": 171248800,
      "step": 7929,
      "time_per_iteration": 2.6202383041381836
    },
    {
      "auxiliary_loss_clip": 0.01166251,
      "auxiliary_loss_mlp": 0.0102492,
      "balance_loss_clip": 1.04689562,
      "balance_loss_mlp": 1.01752961,
      "epoch": 0.9535261227679913,
      "flos": 17602800831360.0,
      "grad_norm": 5.899637887767391,
      "language_loss": 0.75369072,
      "learning_rate": 2.2560034049678988e-08,
      "loss": 0.77560246,
      "num_input_tokens_seen": 171263150,
      "step": 7930,
      "time_per_iteration": 2.612734794616699
    },
    {
      "auxiliary_loss_clip": 0.01170927,
      "auxiliary_loss_mlp": 0.01025036,
      "balance_loss_clip": 1.0492487,
      "balance_loss_mlp": 1.01737976,
      "epoch": 0.9536463656586305,
      "flos": 23142954741120.0,
      "grad_norm": 2.1629962872902477,
      "language_loss": 0.75856471,
      "learning_rate": 2.2443512926008988e-08,
      "loss": 0.78052431,
      "num_input_tokens_seen": 171282480,
      "step": 7931,
      "time_per_iteration": 2.6505894660949707
    },
    {
      "auxiliary_loss_clip": 0.01173215,
      "auxiliary_loss_mlp": 0.01024857,
      "balance_loss_clip": 1.04125643,
      "balance_loss_mlp": 1.01803184,
      "epoch": 0.9537666085492695,
      "flos": 18625033987200.0,
      "grad_norm": 2.7531469642263167,
      "language_loss": 0.69705832,
      "learning_rate": 2.2327291796085946e-08,
      "loss": 0.71903896,
      "num_input_tokens_seen": 171300840,
      "step": 7932,
      "time_per_iteration": 2.625993490219116
    },
    {
      "auxiliary_loss_clip": 0.01166471,
      "auxiliary_loss_mlp": 0.01023767,
      "balance_loss_clip": 1.04589248,
      "balance_loss_mlp": 1.01604807,
      "epoch": 0.9538868514399086,
      "flos": 18989347680000.0,
      "grad_norm": 2.9630084238668886,
      "language_loss": 0.77237743,
      "learning_rate": 2.2211370677540197e-08,
      "loss": 0.79427981,
      "num_input_tokens_seen": 171317365,
      "step": 7933,
      "time_per_iteration": 3.5525174140930176
    },
    {
      "auxiliary_loss_clip": 0.0116688,
      "auxiliary_loss_mlp": 0.01025284,
      "balance_loss_clip": 1.04679179,
      "balance_loss_mlp": 1.01775336,
      "epoch": 0.9540070943305478,
      "flos": 16800556521600.0,
      "grad_norm": 2.637748734380461,
      "language_loss": 0.78109062,
      "learning_rate": 2.2095749587957012e-08,
      "loss": 0.80301231,
      "num_input_tokens_seen": 171335270,
      "step": 7934,
      "time_per_iteration": 2.6113362312316895
    },
    {
      "auxiliary_loss_clip": 0.01167186,
      "auxiliary_loss_mlp": 0.01023754,
      "balance_loss_clip": 1.04111958,
      "balance_loss_mlp": 1.0160892,
      "epoch": 0.9541273372211868,
      "flos": 20156911263360.0,
      "grad_norm": 1.9320022121312104,
      "language_loss": 0.69832283,
      "learning_rate": 2.1980428544876138e-08,
      "loss": 0.72023225,
      "num_input_tokens_seen": 171353910,
      "step": 7935,
      "time_per_iteration": 2.657102584838867
    },
    {
      "auxiliary_loss_clip": 0.01168564,
      "auxiliary_loss_mlp": 0.01021756,
      "balance_loss_clip": 1.03677821,
      "balance_loss_mlp": 1.01423705,
      "epoch": 0.9542475801118259,
      "flos": 26725511381760.0,
      "grad_norm": 5.1108000757011425,
      "language_loss": 0.74361199,
      "learning_rate": 2.1865407565791584e-08,
      "loss": 0.76551515,
      "num_input_tokens_seen": 171375480,
      "step": 7936,
      "time_per_iteration": 2.7105772495269775
    },
    {
      "auxiliary_loss_clip": 0.01172725,
      "auxiliary_loss_mlp": 0.01022303,
      "balance_loss_clip": 1.04314756,
      "balance_loss_mlp": 1.01458716,
      "epoch": 0.954367823002465,
      "flos": 23330911633920.0,
      "grad_norm": 1.9498468940280078,
      "language_loss": 0.77482617,
      "learning_rate": 2.175068666815183e-08,
      "loss": 0.79677641,
      "num_input_tokens_seen": 171396320,
      "step": 7937,
      "time_per_iteration": 3.6322717666625977
    },
    {
      "auxiliary_loss_clip": 0.01174953,
      "auxiliary_loss_mlp": 0.01028968,
      "balance_loss_clip": 1.04244542,
      "balance_loss_mlp": 1.02117717,
      "epoch": 0.9544880658931041,
      "flos": 14902713527040.0,
      "grad_norm": 2.3773192140356247,
      "language_loss": 0.7937234,
      "learning_rate": 2.163626586935985e-08,
      "loss": 0.81576264,
      "num_input_tokens_seen": 171412860,
      "step": 7938,
      "time_per_iteration": 2.6561739444732666
    },
    {
      "auxiliary_loss_clip": 0.01166067,
      "auxiliary_loss_mlp": 0.01027826,
      "balance_loss_clip": 1.04527116,
      "balance_loss_mlp": 1.02012479,
      "epoch": 0.9546083087837431,
      "flos": 29095902725760.0,
      "grad_norm": 2.9226193466860537,
      "language_loss": 0.63049555,
      "learning_rate": 2.1522145186773755e-08,
      "loss": 0.65243453,
      "num_input_tokens_seen": 171431780,
      "step": 7939,
      "time_per_iteration": 2.6242220401763916
    },
    {
      "auxiliary_loss_clip": 0.01170149,
      "auxiliary_loss_mlp": 0.01022971,
      "balance_loss_clip": 1.0442034,
      "balance_loss_mlp": 1.01587558,
      "epoch": 0.9547285516743822,
      "flos": 21142335957120.0,
      "grad_norm": 1.7159321019084188,
      "language_loss": 0.8524996,
      "learning_rate": 2.140832463770481e-08,
      "loss": 0.87443084,
      "num_input_tokens_seen": 171450975,
      "step": 7940,
      "time_per_iteration": 3.6359078884124756
    },
    {
      "auxiliary_loss_clip": 0.01175856,
      "auxiliary_loss_mlp": 0.01022824,
      "balance_loss_clip": 1.04458356,
      "balance_loss_mlp": 1.01535583,
      "epoch": 0.9548487945650214,
      "flos": 27490157130240.0,
      "grad_norm": 2.190683552117147,
      "language_loss": 0.76440674,
      "learning_rate": 2.129480423941987e-08,
      "loss": 0.78639352,
      "num_input_tokens_seen": 171467645,
      "step": 7941,
      "time_per_iteration": 2.7965683937072754
    },
    {
      "auxiliary_loss_clip": 0.01173286,
      "auxiliary_loss_mlp": 0.01022662,
      "balance_loss_clip": 1.04441547,
      "balance_loss_mlp": 1.01575947,
      "epoch": 0.9549690374556604,
      "flos": 22273198819200.0,
      "grad_norm": 1.614153990988731,
      "language_loss": 0.8006469,
      "learning_rate": 2.1181584009140052e-08,
      "loss": 0.82260633,
      "num_input_tokens_seen": 171487185,
      "step": 7942,
      "time_per_iteration": 2.724527597427368
    },
    {
      "auxiliary_loss_clip": 0.01181029,
      "auxiliary_loss_mlp": 0.01025097,
      "balance_loss_clip": 1.04386711,
      "balance_loss_mlp": 1.01844525,
      "epoch": 0.9550892803462995,
      "flos": 17595294888960.0,
      "grad_norm": 2.0460706623770952,
      "language_loss": 0.83920395,
      "learning_rate": 2.10686639640405e-08,
      "loss": 0.8612653,
      "num_input_tokens_seen": 171501275,
      "step": 7943,
      "time_per_iteration": 2.7773783206939697
    },
    {
      "auxiliary_loss_clip": 0.01172835,
      "auxiliary_loss_mlp": 0.01025707,
      "balance_loss_clip": 1.04714322,
      "balance_loss_mlp": 1.01801765,
      "epoch": 0.9552095232369386,
      "flos": 24353144789760.0,
      "grad_norm": 1.6800035205183903,
      "language_loss": 0.81303918,
      "learning_rate": 2.0956044121251294e-08,
      "loss": 0.8350246,
      "num_input_tokens_seen": 171520060,
      "step": 7944,
      "time_per_iteration": 2.624119758605957
    },
    {
      "auxiliary_loss_clip": 0.01172931,
      "auxiliary_loss_mlp": 0.01023525,
      "balance_loss_clip": 1.0445528,
      "balance_loss_mlp": 1.01578844,
      "epoch": 0.9553297661275777,
      "flos": 22746860490240.0,
      "grad_norm": 7.845397127089858,
      "language_loss": 0.8086288,
      "learning_rate": 2.084372449785654e-08,
      "loss": 0.83059335,
      "num_input_tokens_seen": 171539895,
      "step": 7945,
      "time_per_iteration": 2.749608278274536
    },
    {
      "auxiliary_loss_clip": 0.01167739,
      "auxiliary_loss_mlp": 0.01027349,
      "balance_loss_clip": 1.04290462,
      "balance_loss_mlp": 1.0197283,
      "epoch": 0.9554500090182168,
      "flos": 15413866018560.0,
      "grad_norm": 2.134993316060341,
      "language_loss": 0.69426185,
      "learning_rate": 2.0731705110895282e-08,
      "loss": 0.71621275,
      "num_input_tokens_seen": 171557385,
      "step": 7946,
      "time_per_iteration": 2.6347508430480957
    },
    {
      "auxiliary_loss_clip": 0.01173199,
      "auxiliary_loss_mlp": 0.0102521,
      "balance_loss_clip": 1.04837441,
      "balance_loss_mlp": 1.01718092,
      "epoch": 0.9555702519088559,
      "flos": 23513517400320.0,
      "grad_norm": 3.617392841151501,
      "language_loss": 0.86953545,
      "learning_rate": 2.0619985977360587e-08,
      "loss": 0.89151955,
      "num_input_tokens_seen": 171575705,
      "step": 7947,
      "time_per_iteration": 2.775055170059204
    },
    {
      "auxiliary_loss_clip": 0.01173813,
      "auxiliary_loss_mlp": 0.01025112,
      "balance_loss_clip": 1.0365634,
      "balance_loss_mlp": 1.0177803,
      "epoch": 0.955690494799495,
      "flos": 22962072827520.0,
      "grad_norm": 1.9810721995913134,
      "language_loss": 0.76953334,
      "learning_rate": 2.0508567114200237e-08,
      "loss": 0.79152256,
      "num_input_tokens_seen": 171595620,
      "step": 7948,
      "time_per_iteration": 2.757662296295166
    },
    {
      "auxiliary_loss_clip": 0.01173209,
      "auxiliary_loss_mlp": 0.01021786,
      "balance_loss_clip": 1.04417479,
      "balance_loss_mlp": 1.01493776,
      "epoch": 0.955810737690134,
      "flos": 26031250333440.0,
      "grad_norm": 3.9977651083193497,
      "language_loss": 0.78718364,
      "learning_rate": 2.0397448538316485e-08,
      "loss": 0.80913353,
      "num_input_tokens_seen": 171616660,
      "step": 7949,
      "time_per_iteration": 2.721393585205078
    },
    {
      "auxiliary_loss_clip": 0.01169547,
      "auxiliary_loss_mlp": 0.01022725,
      "balance_loss_clip": 1.04309785,
      "balance_loss_mlp": 1.01588213,
      "epoch": 0.9559309805807732,
      "flos": 20849951249280.0,
      "grad_norm": 2.727703162747947,
      "language_loss": 0.6679672,
      "learning_rate": 2.028663026656563e-08,
      "loss": 0.68988997,
      "num_input_tokens_seen": 171635515,
      "step": 7950,
      "time_per_iteration": 2.700870990753174
    },
    {
      "auxiliary_loss_clip": 0.01165655,
      "auxiliary_loss_mlp": 0.01208196,
      "balance_loss_clip": 1.04776096,
      "balance_loss_mlp": 1.0007329,
      "epoch": 0.9560512234714122,
      "flos": 21578219498880.0,
      "grad_norm": 2.820270597202469,
      "language_loss": 0.7164073,
      "learning_rate": 2.0176112315758885e-08,
      "loss": 0.74014586,
      "num_input_tokens_seen": 171653305,
      "step": 7951,
      "time_per_iteration": 3.7109408378601074
    },
    {
      "auxiliary_loss_clip": 0.01183561,
      "auxiliary_loss_mlp": 0.01027827,
      "balance_loss_clip": 1.04179931,
      "balance_loss_mlp": 1.020275,
      "epoch": 0.9561714663620513,
      "flos": 17450144029440.0,
      "grad_norm": 2.3482719509810455,
      "language_loss": 0.68996084,
      "learning_rate": 2.0065894702661957e-08,
      "loss": 0.71207464,
      "num_input_tokens_seen": 171669980,
      "step": 7952,
      "time_per_iteration": 2.7174205780029297
    },
    {
      "auxiliary_loss_clip": 0.01164859,
      "auxiliary_loss_mlp": 0.01207751,
      "balance_loss_clip": 1.03986228,
      "balance_loss_mlp": 1.00070775,
      "epoch": 0.9562917092526905,
      "flos": 26098510550400.0,
      "grad_norm": 3.43264817561849,
      "language_loss": 0.77999139,
      "learning_rate": 1.9955977443994577e-08,
      "loss": 0.80371749,
      "num_input_tokens_seen": 171689970,
      "step": 7953,
      "time_per_iteration": 2.6670188903808594
    },
    {
      "auxiliary_loss_clip": 0.01173641,
      "auxiliary_loss_mlp": 0.01027196,
      "balance_loss_clip": 1.04593658,
      "balance_loss_mlp": 1.0194118,
      "epoch": 0.9564119521433295,
      "flos": 24096742531200.0,
      "grad_norm": 3.997177821710861,
      "language_loss": 0.61979681,
      "learning_rate": 1.9846360556430965e-08,
      "loss": 0.64180517,
      "num_input_tokens_seen": 171708270,
      "step": 7954,
      "time_per_iteration": 2.679420232772827
    },
    {
      "auxiliary_loss_clip": 0.01163496,
      "auxiliary_loss_mlp": 0.01023127,
      "balance_loss_clip": 1.04561603,
      "balance_loss_mlp": 1.01580203,
      "epoch": 0.9565321950339686,
      "flos": 32008903896960.0,
      "grad_norm": 2.942441927206911,
      "language_loss": 0.61515462,
      "learning_rate": 1.973704405660004e-08,
      "loss": 0.63702083,
      "num_input_tokens_seen": 171729385,
      "step": 7955,
      "time_per_iteration": 2.7014541625976562
    },
    {
      "auxiliary_loss_clip": 0.01175474,
      "auxiliary_loss_mlp": 0.01024656,
      "balance_loss_clip": 1.03971744,
      "balance_loss_mlp": 1.01794755,
      "epoch": 0.9566524379246077,
      "flos": 23588642695680.0,
      "grad_norm": 1.5168083277998,
      "language_loss": 0.78265774,
      "learning_rate": 1.9628027961085203e-08,
      "loss": 0.80465901,
      "num_input_tokens_seen": 171752615,
      "step": 7956,
      "time_per_iteration": 2.824124813079834
    },
    {
      "auxiliary_loss_clip": 0.0116083,
      "auxiliary_loss_mlp": 0.01021632,
      "balance_loss_clip": 1.03813481,
      "balance_loss_mlp": 1.01461661,
      "epoch": 0.9567726808152468,
      "flos": 38067716240640.0,
      "grad_norm": 1.6575109819255813,
      "language_loss": 0.83785045,
      "learning_rate": 1.9519312286423894e-08,
      "loss": 0.85967505,
      "num_input_tokens_seen": 171775810,
      "step": 7957,
      "time_per_iteration": 2.8331849575042725
    },
    {
      "auxiliary_loss_clip": 0.0116782,
      "auxiliary_loss_mlp": 0.01021049,
      "balance_loss_clip": 1.04729688,
      "balance_loss_mlp": 1.01356864,
      "epoch": 0.9568929237058859,
      "flos": 22744059229440.0,
      "grad_norm": 2.3323623877776636,
      "language_loss": 0.77850312,
      "learning_rate": 1.9410897049108255e-08,
      "loss": 0.80039179,
      "num_input_tokens_seen": 171795090,
      "step": 7958,
      "time_per_iteration": 2.6396820545196533
    },
    {
      "auxiliary_loss_clip": 0.01174378,
      "auxiliary_loss_mlp": 0.01024669,
      "balance_loss_clip": 1.05204844,
      "balance_loss_mlp": 1.01671219,
      "epoch": 0.957013166596525,
      "flos": 23841633162240.0,
      "grad_norm": 1.96061653467107,
      "language_loss": 0.91518188,
      "learning_rate": 1.9302782265584905e-08,
      "loss": 0.93717235,
      "num_input_tokens_seen": 171815755,
      "step": 7959,
      "time_per_iteration": 2.5668671131134033
    },
    {
      "auxiliary_loss_clip": 0.01162974,
      "auxiliary_loss_mlp": 0.01021467,
      "balance_loss_clip": 1.04053617,
      "balance_loss_mlp": 1.01433849,
      "epoch": 0.9571334094871641,
      "flos": 17639286071040.0,
      "grad_norm": 6.240177472111303,
      "language_loss": 0.87023211,
      "learning_rate": 1.9194967952254282e-08,
      "loss": 0.89207649,
      "num_input_tokens_seen": 171834330,
      "step": 7960,
      "time_per_iteration": 3.5591819286346436
    },
    {
      "auxiliary_loss_clip": 0.0116933,
      "auxiliary_loss_mlp": 0.01024373,
      "balance_loss_clip": 1.04694533,
      "balance_loss_mlp": 1.01676106,
      "epoch": 0.9572536523778031,
      "flos": 15369623441280.0,
      "grad_norm": 2.9934965636547988,
      "language_loss": 0.80879843,
      "learning_rate": 1.9087454125472635e-08,
      "loss": 0.83073556,
      "num_input_tokens_seen": 171848805,
      "step": 7961,
      "time_per_iteration": 2.5473978519439697
    },
    {
      "auxiliary_loss_clip": 0.01166823,
      "auxiliary_loss_mlp": 0.01023305,
      "balance_loss_clip": 1.04734361,
      "balance_loss_mlp": 1.01554418,
      "epoch": 0.9573738952684423,
      "flos": 24969838417920.0,
      "grad_norm": 2.639513317119972,
      "language_loss": 0.78782284,
      "learning_rate": 1.8980240801548696e-08,
      "loss": 0.80972409,
      "num_input_tokens_seen": 171867995,
      "step": 7962,
      "time_per_iteration": 2.606240749359131
    },
    {
      "auxiliary_loss_clip": 0.01170561,
      "auxiliary_loss_mlp": 0.01020352,
      "balance_loss_clip": 1.04732621,
      "balance_loss_mlp": 1.01329505,
      "epoch": 0.9574941381590814,
      "flos": 25769461034880.0,
      "grad_norm": 2.101049538729273,
      "language_loss": 0.73994493,
      "learning_rate": 1.8873327996747458e-08,
      "loss": 0.76185405,
      "num_input_tokens_seen": 171886495,
      "step": 7963,
      "time_per_iteration": 2.6508336067199707
    },
    {
      "auxiliary_loss_clip": 0.01170551,
      "auxiliary_loss_mlp": 0.01023659,
      "balance_loss_clip": 1.04450798,
      "balance_loss_mlp": 1.01669765,
      "epoch": 0.9576143810497204,
      "flos": 32307178435200.0,
      "grad_norm": 1.9145594418866372,
      "language_loss": 0.65920895,
      "learning_rate": 1.8766715727287053e-08,
      "loss": 0.68115103,
      "num_input_tokens_seen": 171908200,
      "step": 7964,
      "time_per_iteration": 3.4861953258514404
    },
    {
      "auxiliary_loss_clip": 0.0117069,
      "auxiliary_loss_mlp": 0.01207762,
      "balance_loss_clip": 1.04437804,
      "balance_loss_mlp": 1.00076437,
      "epoch": 0.9577346239403596,
      "flos": 27745733376000.0,
      "grad_norm": 2.023668822404429,
      "language_loss": 0.79362613,
      "learning_rate": 1.8660404009340546e-08,
      "loss": 0.81741059,
      "num_input_tokens_seen": 171928650,
      "step": 7965,
      "time_per_iteration": 2.6836581230163574
    },
    {
      "auxiliary_loss_clip": 0.01070703,
      "auxiliary_loss_mlp": 0.01001011,
      "balance_loss_clip": 1.01017284,
      "balance_loss_mlp": 1.00025749,
      "epoch": 0.9578548668309986,
      "flos": 57468313710720.0,
      "grad_norm": 0.8697702359560517,
      "language_loss": 0.59472263,
      "learning_rate": 1.8554392859035485e-08,
      "loss": 0.61543977,
      "num_input_tokens_seen": 171986400,
      "step": 7966,
      "time_per_iteration": 3.149973154067993
    },
    {
      "auxiliary_loss_clip": 0.01184753,
      "auxiliary_loss_mlp": 0.01024628,
      "balance_loss_clip": 1.03812671,
      "balance_loss_mlp": 1.01739192,
      "epoch": 0.9579751097216377,
      "flos": 19756040503680.0,
      "grad_norm": 2.013668650808982,
      "language_loss": 0.78969586,
      "learning_rate": 1.8448682292453444e-08,
      "loss": 0.81178969,
      "num_input_tokens_seen": 172005475,
      "step": 7967,
      "time_per_iteration": 3.677576780319214
    },
    {
      "auxiliary_loss_clip": 0.01165518,
      "auxiliary_loss_mlp": 0.01025444,
      "balance_loss_clip": 1.04735625,
      "balance_loss_mlp": 1.01811838,
      "epoch": 0.9580953526122769,
      "flos": 18041270152320.0,
      "grad_norm": 1.7137658149144561,
      "language_loss": 0.66291761,
      "learning_rate": 1.8343272325631154e-08,
      "loss": 0.68482721,
      "num_input_tokens_seen": 172024420,
      "step": 7968,
      "time_per_iteration": 2.6362318992614746
    },
    {
      "auxiliary_loss_clip": 0.01183742,
      "auxiliary_loss_mlp": 0.01207869,
      "balance_loss_clip": 1.03669727,
      "balance_loss_mlp": 1.00078869,
      "epoch": 0.9582155955029159,
      "flos": 24270154416000.0,
      "grad_norm": 2.441645247893296,
      "language_loss": 0.78481805,
      "learning_rate": 1.8238162974558492e-08,
      "loss": 0.80873418,
      "num_input_tokens_seen": 172038350,
      "step": 7969,
      "time_per_iteration": 2.7492311000823975
    },
    {
      "auxiliary_loss_clip": 0.01166905,
      "auxiliary_loss_mlp": 0.0102837,
      "balance_loss_clip": 1.04431152,
      "balance_loss_mlp": 1.02082396,
      "epoch": 0.958335838393555,
      "flos": 22783309816320.0,
      "grad_norm": 1.9315412667201461,
      "language_loss": 0.75031883,
      "learning_rate": 1.8133354255181144e-08,
      "loss": 0.77227151,
      "num_input_tokens_seen": 172058665,
      "step": 7970,
      "time_per_iteration": 2.6996190547943115
    },
    {
      "auxiliary_loss_clip": 0.01161918,
      "auxiliary_loss_mlp": 0.0102513,
      "balance_loss_clip": 1.04314721,
      "balance_loss_mlp": 1.01774502,
      "epoch": 0.958456081284194,
      "flos": 16911484698240.0,
      "grad_norm": 2.1485117791955175,
      "language_loss": 0.75189626,
      "learning_rate": 1.802884618339795e-08,
      "loss": 0.77376676,
      "num_input_tokens_seen": 172077470,
      "step": 7971,
      "time_per_iteration": 2.6236932277679443
    },
    {
      "auxiliary_loss_clip": 0.01173656,
      "auxiliary_loss_mlp": 0.01022148,
      "balance_loss_clip": 1.05023098,
      "balance_loss_mlp": 1.01441145,
      "epoch": 0.9585763241748332,
      "flos": 19974951941760.0,
      "grad_norm": 2.4378553094180146,
      "language_loss": 0.81455845,
      "learning_rate": 1.7924638775062894e-08,
      "loss": 0.8365165,
      "num_input_tokens_seen": 172096590,
      "step": 7972,
      "time_per_iteration": 2.6713099479675293
    },
    {
      "auxiliary_loss_clip": 0.01167108,
      "auxiliary_loss_mlp": 0.01025751,
      "balance_loss_clip": 1.04307747,
      "balance_loss_mlp": 1.01850307,
      "epoch": 0.9586965670654722,
      "flos": 21395649646080.0,
      "grad_norm": 1.9116303743422247,
      "language_loss": 0.818313,
      "learning_rate": 1.7820732045984444e-08,
      "loss": 0.84024155,
      "num_input_tokens_seen": 172116735,
      "step": 7973,
      "time_per_iteration": 2.69559383392334
    },
    {
      "auxiliary_loss_clip": 0.01166697,
      "auxiliary_loss_mlp": 0.01024768,
      "balance_loss_clip": 1.04442859,
      "balance_loss_mlp": 1.01721632,
      "epoch": 0.9588168099561113,
      "flos": 21435115714560.0,
      "grad_norm": 1.8110326323941204,
      "language_loss": 0.73997426,
      "learning_rate": 1.7717126011924655e-08,
      "loss": 0.76188898,
      "num_input_tokens_seen": 172138320,
      "step": 7974,
      "time_per_iteration": 2.6476871967315674
    },
    {
      "auxiliary_loss_clip": 0.01166893,
      "auxiliary_loss_mlp": 0.01025069,
      "balance_loss_clip": 1.03556824,
      "balance_loss_mlp": 1.01769304,
      "epoch": 0.9589370528467505,
      "flos": 11763761852160.0,
      "grad_norm": 2.4706748287250386,
      "language_loss": 0.76726431,
      "learning_rate": 1.7613820688600957e-08,
      "loss": 0.78918397,
      "num_input_tokens_seen": 172154225,
      "step": 7975,
      "time_per_iteration": 2.7098796367645264
    },
    {
      "auxiliary_loss_clip": 0.01176819,
      "auxiliary_loss_mlp": 0.01024358,
      "balance_loss_clip": 1.04254413,
      "balance_loss_mlp": 1.01729751,
      "epoch": 0.9590572957373895,
      "flos": 23441516588160.0,
      "grad_norm": 2.1976092653788024,
      "language_loss": 0.78726202,
      "learning_rate": 1.7510816091684588e-08,
      "loss": 0.80927378,
      "num_input_tokens_seen": 172174150,
      "step": 7976,
      "time_per_iteration": 2.6314706802368164
    },
    {
      "auxiliary_loss_clip": 0.01173462,
      "auxiliary_loss_mlp": 0.01027243,
      "balance_loss_clip": 1.04486132,
      "balance_loss_mlp": 1.01958966,
      "epoch": 0.9591775386280286,
      "flos": 22528272274560.0,
      "grad_norm": 3.2409250247889476,
      "language_loss": 0.78548717,
      "learning_rate": 1.740811223680083e-08,
      "loss": 0.80749416,
      "num_input_tokens_seen": 172191005,
      "step": 7977,
      "time_per_iteration": 2.6421239376068115
    },
    {
      "auxiliary_loss_clip": 0.01165883,
      "auxiliary_loss_mlp": 0.01024644,
      "balance_loss_clip": 1.04664385,
      "balance_loss_mlp": 1.01705325,
      "epoch": 0.9592977815186677,
      "flos": 18186959715840.0,
      "grad_norm": 2.470319513781341,
      "language_loss": 0.7415061,
      "learning_rate": 1.7305709139530334e-08,
      "loss": 0.76341146,
      "num_input_tokens_seen": 172209785,
      "step": 7978,
      "time_per_iteration": 3.5256736278533936
    },
    {
      "auxiliary_loss_clip": 0.01161942,
      "auxiliary_loss_mlp": 0.01021176,
      "balance_loss_clip": 1.0435133,
      "balance_loss_mlp": 1.0139339,
      "epoch": 0.9594180244093068,
      "flos": 16537797555840.0,
      "grad_norm": 2.519603736976007,
      "language_loss": 0.74668711,
      "learning_rate": 1.7203606815407334e-08,
      "loss": 0.76851827,
      "num_input_tokens_seen": 172224380,
      "step": 7979,
      "time_per_iteration": 2.624713897705078
    },
    {
      "auxiliary_loss_clip": 0.01178908,
      "auxiliary_loss_mlp": 0.01023704,
      "balance_loss_clip": 1.04830718,
      "balance_loss_mlp": 1.0162828,
      "epoch": 0.9595382672999458,
      "flos": 20554334317440.0,
      "grad_norm": 1.7827436282117464,
      "language_loss": 0.79370373,
      "learning_rate": 1.7101805279920557e-08,
      "loss": 0.81572986,
      "num_input_tokens_seen": 172242540,
      "step": 7980,
      "time_per_iteration": 2.602473497390747
    },
    {
      "auxiliary_loss_clip": 0.01166229,
      "auxiliary_loss_mlp": 0.01023264,
      "balance_loss_clip": 1.0473932,
      "balance_loss_mlp": 1.01535738,
      "epoch": 0.959658510190585,
      "flos": 22638266697600.0,
      "grad_norm": 2.533872486176575,
      "language_loss": 0.81117916,
      "learning_rate": 1.7000304548513643e-08,
      "loss": 0.83307409,
      "num_input_tokens_seen": 172262645,
      "step": 7981,
      "time_per_iteration": 2.6160271167755127
    },
    {
      "auxiliary_loss_clip": 0.01172571,
      "auxiliary_loss_mlp": 0.01026786,
      "balance_loss_clip": 1.0422461,
      "balance_loss_mlp": 1.01916218,
      "epoch": 0.9597787530812241,
      "flos": 19135252725120.0,
      "grad_norm": 7.769365611014474,
      "language_loss": 0.83061981,
      "learning_rate": 1.6899104636583394e-08,
      "loss": 0.85261345,
      "num_input_tokens_seen": 172280695,
      "step": 7982,
      "time_per_iteration": 2.6556198596954346
    },
    {
      "auxiliary_loss_clip": 0.01070859,
      "auxiliary_loss_mlp": 0.01000546,
      "balance_loss_clip": 1.01029158,
      "balance_loss_mlp": 0.99982756,
      "epoch": 0.9598989959718631,
      "flos": 60098124055680.0,
      "grad_norm": 0.7465614562003953,
      "language_loss": 0.61841333,
      "learning_rate": 1.6798205559482638e-08,
      "loss": 0.63912737,
      "num_input_tokens_seen": 172343075,
      "step": 7983,
      "time_per_iteration": 3.274888515472412
    },
    {
      "auxiliary_loss_clip": 0.0117589,
      "auxiliary_loss_mlp": 0.01025073,
      "balance_loss_clip": 1.04517293,
      "balance_loss_mlp": 1.0174675,
      "epoch": 0.9600192388625023,
      "flos": 20886795624960.0,
      "grad_norm": 7.328693144985866,
      "language_loss": 0.76746881,
      "learning_rate": 1.669760733251713e-08,
      "loss": 0.78947848,
      "num_input_tokens_seen": 172361950,
      "step": 7984,
      "time_per_iteration": 2.66829252243042
    },
    {
      "auxiliary_loss_clip": 0.01188657,
      "auxiliary_loss_mlp": 0.0102492,
      "balance_loss_clip": 1.04081404,
      "balance_loss_mlp": 1.01826179,
      "epoch": 0.9601394817531413,
      "flos": 20445740524800.0,
      "grad_norm": 1.607753119142967,
      "language_loss": 0.82128656,
      "learning_rate": 1.659730997094755e-08,
      "loss": 0.84342229,
      "num_input_tokens_seen": 172380440,
      "step": 7985,
      "time_per_iteration": 2.6607048511505127
    },
    {
      "auxiliary_loss_clip": 0.01160969,
      "auxiliary_loss_mlp": 0.010233,
      "balance_loss_clip": 1.0429399,
      "balance_loss_mlp": 1.01625776,
      "epoch": 0.9602597246437804,
      "flos": 21507152440320.0,
      "grad_norm": 1.7435561872272705,
      "language_loss": 0.62540007,
      "learning_rate": 1.6497313489989283e-08,
      "loss": 0.64724272,
      "num_input_tokens_seen": 172400265,
      "step": 7986,
      "time_per_iteration": 2.6377573013305664
    },
    {
      "auxiliary_loss_clip": 0.01175279,
      "auxiliary_loss_mlp": 0.01027275,
      "balance_loss_clip": 1.03512239,
      "balance_loss_mlp": 1.01975894,
      "epoch": 0.9603799675344196,
      "flos": 29935099152000.0,
      "grad_norm": 4.964701003363229,
      "language_loss": 0.70415545,
      "learning_rate": 1.639761790481131e-08,
      "loss": 0.72618103,
      "num_input_tokens_seen": 172421145,
      "step": 7987,
      "time_per_iteration": 3.637779712677002
    },
    {
      "auxiliary_loss_clip": 0.01170753,
      "auxiliary_loss_mlp": 0.01024791,
      "balance_loss_clip": 1.04573941,
      "balance_loss_mlp": 1.01756084,
      "epoch": 0.9605002104250586,
      "flos": 28001525103360.0,
      "grad_norm": 2.260934136495835,
      "language_loss": 0.79306293,
      "learning_rate": 1.6298223230537754e-08,
      "loss": 0.81501836,
      "num_input_tokens_seen": 172438945,
      "step": 7988,
      "time_per_iteration": 2.652622938156128
    },
    {
      "auxiliary_loss_clip": 0.01166993,
      "auxiliary_loss_mlp": 0.01207908,
      "balance_loss_clip": 1.04200649,
      "balance_loss_mlp": 1.00073838,
      "epoch": 0.9606204533156977,
      "flos": 35590490870400.0,
      "grad_norm": 1.9494239490922833,
      "language_loss": 0.69895303,
      "learning_rate": 1.619912948224611e-08,
      "loss": 0.72270203,
      "num_input_tokens_seen": 172460150,
      "step": 7989,
      "time_per_iteration": 2.818206548690796
    },
    {
      "auxiliary_loss_clip": 0.01166723,
      "auxiliary_loss_mlp": 0.01027909,
      "balance_loss_clip": 1.04216635,
      "balance_loss_mlp": 1.02012134,
      "epoch": 0.9607406962063368,
      "flos": 26574614346240.0,
      "grad_norm": 6.412285900339188,
      "language_loss": 0.61346531,
      "learning_rate": 1.6100336674969682e-08,
      "loss": 0.63541162,
      "num_input_tokens_seen": 172478990,
      "step": 7990,
      "time_per_iteration": 2.8060662746429443
    },
    {
      "auxiliary_loss_clip": 0.01181159,
      "auxiliary_loss_mlp": 0.01032548,
      "balance_loss_clip": 1.04149079,
      "balance_loss_mlp": 1.02486229,
      "epoch": 0.9608609390969759,
      "flos": 25331781813120.0,
      "grad_norm": 1.9107788552743874,
      "language_loss": 0.7680161,
      "learning_rate": 1.600184482369449e-08,
      "loss": 0.79015321,
      "num_input_tokens_seen": 172498905,
      "step": 7991,
      "time_per_iteration": 3.6298532485961914
    },
    {
      "auxiliary_loss_clip": 0.01179874,
      "auxiliary_loss_mlp": 0.01027403,
      "balance_loss_clip": 1.04353237,
      "balance_loss_mlp": 1.01911235,
      "epoch": 0.960981181987615,
      "flos": 21069114082560.0,
      "grad_norm": 2.7255702787170413,
      "language_loss": 0.89319396,
      "learning_rate": 1.5903653943362126e-08,
      "loss": 0.91526675,
      "num_input_tokens_seen": 172517900,
      "step": 7992,
      "time_per_iteration": 2.7063140869140625
    },
    {
      "auxiliary_loss_clip": 0.01171622,
      "auxiliary_loss_mlp": 0.01021886,
      "balance_loss_clip": 1.04439414,
      "balance_loss_mlp": 1.01490903,
      "epoch": 0.9611014248782541,
      "flos": 17823256554240.0,
      "grad_norm": 2.6472536255581334,
      "language_loss": 0.77115047,
      "learning_rate": 1.580576404886802e-08,
      "loss": 0.79308558,
      "num_input_tokens_seen": 172536430,
      "step": 7993,
      "time_per_iteration": 3.531402111053467
    },
    {
      "auxiliary_loss_clip": 0.01168185,
      "auxiliary_loss_mlp": 0.01023175,
      "balance_loss_clip": 1.04461765,
      "balance_loss_mlp": 1.01637125,
      "epoch": 0.9612216677688932,
      "flos": 19354631040000.0,
      "grad_norm": 14.103192154718498,
      "language_loss": 0.79943371,
      "learning_rate": 1.570817515506162e-08,
      "loss": 0.82134724,
      "num_input_tokens_seen": 172555120,
      "step": 7994,
      "time_per_iteration": 2.631817579269409
    },
    {
      "auxiliary_loss_clip": 0.01163901,
      "auxiliary_loss_mlp": 0.01021754,
      "balance_loss_clip": 1.0471983,
      "balance_loss_mlp": 1.0150423,
      "epoch": 0.9613419106595322,
      "flos": 15808739207040.0,
      "grad_norm": 1.9372947591479412,
      "language_loss": 0.81341887,
      "learning_rate": 1.561088727674753e-08,
      "loss": 0.83527541,
      "num_input_tokens_seen": 172569330,
      "step": 7995,
      "time_per_iteration": 2.558856248855591
    },
    {
      "auxiliary_loss_clip": 0.01192046,
      "auxiliary_loss_mlp": 0.0103186,
      "balance_loss_clip": 1.04335177,
      "balance_loss_mlp": 1.02352452,
      "epoch": 0.9614621535501714,
      "flos": 25702488126720.0,
      "grad_norm": 2.3760676582804208,
      "language_loss": 0.71051109,
      "learning_rate": 1.551390042868417e-08,
      "loss": 0.73275012,
      "num_input_tokens_seen": 172591100,
      "step": 7996,
      "time_per_iteration": 2.7351863384246826
    },
    {
      "auxiliary_loss_clip": 0.0116983,
      "auxiliary_loss_mlp": 0.01021636,
      "balance_loss_clip": 1.04677355,
      "balance_loss_mlp": 1.01376247,
      "epoch": 0.9615823964408104,
      "flos": 17819054663040.0,
      "grad_norm": 1.8947703798302777,
      "language_loss": 0.70909107,
      "learning_rate": 1.5417214625584207e-08,
      "loss": 0.73100573,
      "num_input_tokens_seen": 172608755,
      "step": 7997,
      "time_per_iteration": 2.6197636127471924
    },
    {
      "auxiliary_loss_clip": 0.01162179,
      "auxiliary_loss_mlp": 0.01022176,
      "balance_loss_clip": 1.04253674,
      "balance_loss_mlp": 1.01473129,
      "epoch": 0.9617026393314495,
      "flos": 20190020624640.0,
      "grad_norm": 7.662813084569592,
      "language_loss": 0.85154176,
      "learning_rate": 1.5320829882114806e-08,
      "loss": 0.87338531,
      "num_input_tokens_seen": 172626830,
      "step": 7998,
      "time_per_iteration": 2.6051113605499268
    },
    {
      "auxiliary_loss_clip": 0.01165126,
      "auxiliary_loss_mlp": 0.01024677,
      "balance_loss_clip": 1.04517257,
      "balance_loss_mlp": 1.01738763,
      "epoch": 0.9618228822220887,
      "flos": 20267013427200.0,
      "grad_norm": 2.8704582498453592,
      "language_loss": 0.790012,
      "learning_rate": 1.5224746212897378e-08,
      "loss": 0.81191003,
      "num_input_tokens_seen": 172646125,
      "step": 7999,
      "time_per_iteration": 2.603692054748535
    },
    {
      "auxiliary_loss_clip": 0.01162813,
      "auxiliary_loss_mlp": 0.01024069,
      "balance_loss_clip": 1.04565239,
      "balance_loss_mlp": 1.01688647,
      "epoch": 0.9619431251127277,
      "flos": 21031300039680.0,
      "grad_norm": 1.6169942892018676,
      "language_loss": 0.7721765,
      "learning_rate": 1.512896363250804e-08,
      "loss": 0.79404533,
      "num_input_tokens_seen": 172666235,
      "step": 8000,
      "time_per_iteration": 2.6608474254608154
    },
    {
      "auxiliary_loss_clip": 0.01170373,
      "auxiliary_loss_mlp": 0.01028775,
      "balance_loss_clip": 1.04494929,
      "balance_loss_mlp": 1.02135086,
      "epoch": 0.9620633680033668,
      "flos": 22382654538240.0,
      "grad_norm": 2.0959755594330334,
      "language_loss": 0.75858724,
      "learning_rate": 1.503348215547673e-08,
      "loss": 0.78057873,
      "num_input_tokens_seen": 172687325,
      "step": 8001,
      "time_per_iteration": 2.757321834564209
    },
    {
      "auxiliary_loss_clip": 0.0116843,
      "auxiliary_loss_mlp": 0.01023061,
      "balance_loss_clip": 1.04383016,
      "balance_loss_mlp": 1.01586068,
      "epoch": 0.962183610894006,
      "flos": 18471730740480.0,
      "grad_norm": 2.58745721883756,
      "language_loss": 0.80962759,
      "learning_rate": 1.4938301796288078e-08,
      "loss": 0.83154255,
      "num_input_tokens_seen": 172703895,
      "step": 8002,
      "time_per_iteration": 2.654238224029541
    },
    {
      "auxiliary_loss_clip": 0.01166031,
      "auxiliary_loss_mlp": 0.01022443,
      "balance_loss_clip": 1.04697299,
      "balance_loss_mlp": 1.01453984,
      "epoch": 0.962303853784645,
      "flos": 18435245500800.0,
      "grad_norm": 2.6962830511183284,
      "language_loss": 0.81741667,
      "learning_rate": 1.4843422569380537e-08,
      "loss": 0.83930141,
      "num_input_tokens_seen": 172720650,
      "step": 8003,
      "time_per_iteration": 2.545008897781372
    },
    {
      "auxiliary_loss_clip": 0.01171973,
      "auxiliary_loss_mlp": 0.01021177,
      "balance_loss_clip": 1.03869343,
      "balance_loss_mlp": 1.01396525,
      "epoch": 0.9624240966752841,
      "flos": 26391074826240.0,
      "grad_norm": 2.084310324959368,
      "language_loss": 0.82815248,
      "learning_rate": 1.4748844489147483e-08,
      "loss": 0.85008401,
      "num_input_tokens_seen": 172737640,
      "step": 8004,
      "time_per_iteration": 2.7078158855438232
    },
    {
      "auxiliary_loss_clip": 0.01170143,
      "auxiliary_loss_mlp": 0.01020809,
      "balance_loss_clip": 1.04193592,
      "balance_loss_mlp": 1.01418996,
      "epoch": 0.9625443395659231,
      "flos": 14647675985280.0,
      "grad_norm": 3.0034032250456413,
      "language_loss": 0.7102859,
      "learning_rate": 1.4654567569936326e-08,
      "loss": 0.73219544,
      "num_input_tokens_seen": 172755215,
      "step": 8005,
      "time_per_iteration": 3.496706962585449
    },
    {
      "auxiliary_loss_clip": 0.01170777,
      "auxiliary_loss_mlp": 0.01029058,
      "balance_loss_clip": 1.03949642,
      "balance_loss_mlp": 1.02167034,
      "epoch": 0.9626645824565623,
      "flos": 18367626147840.0,
      "grad_norm": 2.201104955598812,
      "language_loss": 0.83040273,
      "learning_rate": 1.456059182604874e-08,
      "loss": 0.85240108,
      "num_input_tokens_seen": 172774020,
      "step": 8006,
      "time_per_iteration": 2.622069835662842
    },
    {
      "auxiliary_loss_clip": 0.01168887,
      "auxiliary_loss_mlp": 0.01026963,
      "balance_loss_clip": 1.04948854,
      "balance_loss_mlp": 1.01926184,
      "epoch": 0.9627848253472013,
      "flos": 16580424021120.0,
      "grad_norm": 1.9352795158135971,
      "language_loss": 0.76540434,
      "learning_rate": 1.4466917271740653e-08,
      "loss": 0.78736287,
      "num_input_tokens_seen": 172792220,
      "step": 8007,
      "time_per_iteration": 2.6469924449920654
    },
    {
      "auxiliary_loss_clip": 0.01167696,
      "auxiliary_loss_mlp": 0.01024775,
      "balance_loss_clip": 1.04284215,
      "balance_loss_mlp": 1.0170269,
      "epoch": 0.9629050682378404,
      "flos": 20886867452160.0,
      "grad_norm": 2.045089463037287,
      "language_loss": 0.67725241,
      "learning_rate": 1.4373543921222697e-08,
      "loss": 0.69917715,
      "num_input_tokens_seen": 172811805,
      "step": 8008,
      "time_per_iteration": 2.6028144359588623
    },
    {
      "auxiliary_loss_clip": 0.01168306,
      "auxiliary_loss_mlp": 0.0102673,
      "balance_loss_clip": 1.04445899,
      "balance_loss_mlp": 1.01922846,
      "epoch": 0.9630253111284796,
      "flos": 17019252478080.0,
      "grad_norm": 2.0244096457459477,
      "language_loss": 0.78291845,
      "learning_rate": 1.428047178865932e-08,
      "loss": 0.80486882,
      "num_input_tokens_seen": 172828595,
      "step": 8009,
      "time_per_iteration": 2.606405019760132
    },
    {
      "auxiliary_loss_clip": 0.01169841,
      "auxiliary_loss_mlp": 0.0102416,
      "balance_loss_clip": 1.04181492,
      "balance_loss_mlp": 1.01658106,
      "epoch": 0.9631455540191186,
      "flos": 20338942412160.0,
      "grad_norm": 2.688364798722209,
      "language_loss": 0.74567604,
      "learning_rate": 1.4187700888169451e-08,
      "loss": 0.76761603,
      "num_input_tokens_seen": 172847770,
      "step": 8010,
      "time_per_iteration": 2.6631412506103516
    },
    {
      "auxiliary_loss_clip": 0.01069678,
      "auxiliary_loss_mlp": 0.01000617,
      "balance_loss_clip": 1.01114821,
      "balance_loss_mlp": 0.99985695,
      "epoch": 0.9632657969097577,
      "flos": 65956700033280.0,
      "grad_norm": 0.7504991636350754,
      "language_loss": 0.56913161,
      "learning_rate": 1.40952312338265e-08,
      "loss": 0.58983451,
      "num_input_tokens_seen": 172912415,
      "step": 8011,
      "time_per_iteration": 3.26259708404541
    },
    {
      "auxiliary_loss_clip": 0.01172873,
      "auxiliary_loss_mlp": 0.01025314,
      "balance_loss_clip": 1.04031217,
      "balance_loss_mlp": 1.0177505,
      "epoch": 0.9633860398003968,
      "flos": 44419523823360.0,
      "grad_norm": 3.1417335264892348,
      "language_loss": 0.6867702,
      "learning_rate": 1.4003062839657909e-08,
      "loss": 0.70875216,
      "num_input_tokens_seen": 172934895,
      "step": 8012,
      "time_per_iteration": 2.9208109378814697
    },
    {
      "auxiliary_loss_clip": 0.01176421,
      "auxiliary_loss_mlp": 0.01019051,
      "balance_loss_clip": 1.04322553,
      "balance_loss_mlp": 1.01217818,
      "epoch": 0.9635062826910359,
      "flos": 24827704300800.0,
      "grad_norm": 1.8680082383957692,
      "language_loss": 0.79889119,
      "learning_rate": 1.391119571964583e-08,
      "loss": 0.8208459,
      "num_input_tokens_seen": 172955835,
      "step": 8013,
      "time_per_iteration": 2.837456226348877
    },
    {
      "auxiliary_loss_clip": 0.01168443,
      "auxiliary_loss_mlp": 0.0102649,
      "balance_loss_clip": 1.04732382,
      "balance_loss_mlp": 1.01891994,
      "epoch": 0.9636265255816749,
      "flos": 15961360095360.0,
      "grad_norm": 2.2211322954978256,
      "language_loss": 0.72698867,
      "learning_rate": 1.3819629887726225e-08,
      "loss": 0.74893796,
      "num_input_tokens_seen": 172973925,
      "step": 8014,
      "time_per_iteration": 3.6017489433288574
    },
    {
      "auxiliary_loss_clip": 0.01177308,
      "auxiliary_loss_mlp": 0.01023015,
      "balance_loss_clip": 1.04816604,
      "balance_loss_mlp": 1.01552916,
      "epoch": 0.9637467684723141,
      "flos": 22601781457920.0,
      "grad_norm": 4.872302378063715,
      "language_loss": 0.76549417,
      "learning_rate": 1.3728365357789317e-08,
      "loss": 0.7874974,
      "num_input_tokens_seen": 172993290,
      "step": 8015,
      "time_per_iteration": 2.649498701095581
    },
    {
      "auxiliary_loss_clip": 0.01171188,
      "auxiliary_loss_mlp": 0.01022802,
      "balance_loss_clip": 1.03841317,
      "balance_loss_mlp": 1.01455295,
      "epoch": 0.9638670113629532,
      "flos": 17565812801280.0,
      "grad_norm": 7.224635930181017,
      "language_loss": 0.77093452,
      "learning_rate": 1.3637402143680254e-08,
      "loss": 0.79287446,
      "num_input_tokens_seen": 173008190,
      "step": 8016,
      "time_per_iteration": 2.7241063117980957
    },
    {
      "auxiliary_loss_clip": 0.01078513,
      "auxiliary_loss_mlp": 0.01001192,
      "balance_loss_clip": 1.012766,
      "balance_loss_mlp": 1.00047386,
      "epoch": 0.9639872542535922,
      "flos": 55072139379840.0,
      "grad_norm": 0.7266203705003811,
      "language_loss": 0.54973447,
      "learning_rate": 1.3546740259197998e-08,
      "loss": 0.57053149,
      "num_input_tokens_seen": 173061000,
      "step": 8017,
      "time_per_iteration": 4.00602388381958
    },
    {
      "auxiliary_loss_clip": 0.01170622,
      "auxiliary_loss_mlp": 0.01027956,
      "balance_loss_clip": 1.04364562,
      "balance_loss_mlp": 1.02016592,
      "epoch": 0.9641074971442314,
      "flos": 24134484746880.0,
      "grad_norm": 2.5444553359875157,
      "language_loss": 0.70562744,
      "learning_rate": 1.3456379718095989e-08,
      "loss": 0.72761321,
      "num_input_tokens_seen": 173081415,
      "step": 8018,
      "time_per_iteration": 2.6435723304748535
    },
    {
      "auxiliary_loss_clip": 0.01076347,
      "auxiliary_loss_mlp": 0.01001983,
      "balance_loss_clip": 1.01204884,
      "balance_loss_mlp": 1.00124955,
      "epoch": 0.9642277400348704,
      "flos": 66747416077440.0,
      "grad_norm": 0.8408242921517329,
      "language_loss": 0.62043583,
      "learning_rate": 1.3366320534081487e-08,
      "loss": 0.64121914,
      "num_input_tokens_seen": 173144095,
      "step": 8019,
      "time_per_iteration": 3.2078380584716797
    },
    {
      "auxiliary_loss_clip": 0.01167849,
      "auxiliary_loss_mlp": 0.01021568,
      "balance_loss_clip": 1.04483914,
      "balance_loss_mlp": 1.01425481,
      "epoch": 0.9643479829255095,
      "flos": 30920272450560.0,
      "grad_norm": 2.5567431715635887,
      "language_loss": 0.75560105,
      "learning_rate": 1.3276562720816675e-08,
      "loss": 0.77749527,
      "num_input_tokens_seen": 173165605,
      "step": 8020,
      "time_per_iteration": 3.6681113243103027
    },
    {
      "auxiliary_loss_clip": 0.01166772,
      "auxiliary_loss_mlp": 0.01026165,
      "balance_loss_clip": 1.04674196,
      "balance_loss_mlp": 1.01845801,
      "epoch": 0.9644682258161487,
      "flos": 20048245643520.0,
      "grad_norm": 2.535745822022282,
      "language_loss": 0.82489705,
      "learning_rate": 1.3187106291917549e-08,
      "loss": 0.84682637,
      "num_input_tokens_seen": 173182595,
      "step": 8021,
      "time_per_iteration": 2.6010584831237793
    },
    {
      "auxiliary_loss_clip": 0.01161915,
      "auxiliary_loss_mlp": 0.01019881,
      "balance_loss_clip": 1.04407978,
      "balance_loss_mlp": 1.01333308,
      "epoch": 0.9645884687067877,
      "flos": 21178713456000.0,
      "grad_norm": 3.16418227878228,
      "language_loss": 0.70879656,
      "learning_rate": 1.309795126095503e-08,
      "loss": 0.73061454,
      "num_input_tokens_seen": 173200895,
      "step": 8022,
      "time_per_iteration": 2.5879950523376465
    },
    {
      "auxiliary_loss_clip": 0.01174415,
      "auxiliary_loss_mlp": 0.01023334,
      "balance_loss_clip": 1.03565216,
      "balance_loss_mlp": 1.01591587,
      "epoch": 0.9647087115974268,
      "flos": 18945967029120.0,
      "grad_norm": 5.690024894984531,
      "language_loss": 0.80710089,
      "learning_rate": 1.3009097641453192e-08,
      "loss": 0.82907844,
      "num_input_tokens_seen": 173218745,
      "step": 8023,
      "time_per_iteration": 2.7751240730285645
    },
    {
      "auxiliary_loss_clip": 0.0116976,
      "auxiliary_loss_mlp": 0.01019698,
      "balance_loss_clip": 1.04479182,
      "balance_loss_mlp": 1.01249182,
      "epoch": 0.9648289544880659,
      "flos": 16545088016640.0,
      "grad_norm": 8.546321848353504,
      "language_loss": 0.76160598,
      "learning_rate": 1.2920545446891474e-08,
      "loss": 0.78350055,
      "num_input_tokens_seen": 173235465,
      "step": 8024,
      "time_per_iteration": 2.6488614082336426
    },
    {
      "auxiliary_loss_clip": 0.01173254,
      "auxiliary_loss_mlp": 0.01031455,
      "balance_loss_clip": 1.04700172,
      "balance_loss_mlp": 1.02387881,
      "epoch": 0.964949197378705,
      "flos": 24057527857920.0,
      "grad_norm": 2.217737470124313,
      "language_loss": 0.70731062,
      "learning_rate": 1.2832294690703127e-08,
      "loss": 0.72935772,
      "num_input_tokens_seen": 173254440,
      "step": 8025,
      "time_per_iteration": 2.6422486305236816
    },
    {
      "auxiliary_loss_clip": 0.01167082,
      "auxiliary_loss_mlp": 0.0102505,
      "balance_loss_clip": 1.04583883,
      "balance_loss_mlp": 1.017537,
      "epoch": 0.965069440269344,
      "flos": 23365565280000.0,
      "grad_norm": 1.9067930952486585,
      "language_loss": 0.77629584,
      "learning_rate": 1.2744345386275668e-08,
      "loss": 0.79821718,
      "num_input_tokens_seen": 173273980,
      "step": 8026,
      "time_per_iteration": 2.6542298793792725
    },
    {
      "auxiliary_loss_clip": 0.0118011,
      "auxiliary_loss_mlp": 0.01026632,
      "balance_loss_clip": 1.04879999,
      "balance_loss_mlp": 1.01898527,
      "epoch": 0.9651896831599832,
      "flos": 25374875155200.0,
      "grad_norm": 1.5907417691125671,
      "language_loss": 0.78844726,
      "learning_rate": 1.265669754695109e-08,
      "loss": 0.81051469,
      "num_input_tokens_seen": 173293550,
      "step": 8027,
      "time_per_iteration": 2.6475627422332764
    },
    {
      "auxiliary_loss_clip": 0.01178521,
      "auxiliary_loss_mlp": 0.01027466,
      "balance_loss_clip": 1.03787839,
      "balance_loss_mlp": 1.01980639,
      "epoch": 0.9653099260506223,
      "flos": 22272875596800.0,
      "grad_norm": 2.4018567767447467,
      "language_loss": 0.82182693,
      "learning_rate": 1.2569351186025201e-08,
      "loss": 0.84388679,
      "num_input_tokens_seen": 173312005,
      "step": 8028,
      "time_per_iteration": 2.7526509761810303
    },
    {
      "auxiliary_loss_clip": 0.01161187,
      "auxiliary_loss_mlp": 0.01020468,
      "balance_loss_clip": 1.0397892,
      "balance_loss_mlp": 1.01352453,
      "epoch": 0.9654301689412613,
      "flos": 26760847386240.0,
      "grad_norm": 3.0017193726755864,
      "language_loss": 0.75562203,
      "learning_rate": 1.2482306316748737e-08,
      "loss": 0.77743858,
      "num_input_tokens_seen": 173332450,
      "step": 8029,
      "time_per_iteration": 2.692566394805908
    },
    {
      "auxiliary_loss_clip": 0.01173077,
      "auxiliary_loss_mlp": 0.01025145,
      "balance_loss_clip": 1.04511428,
      "balance_loss_mlp": 1.01846623,
      "epoch": 0.9655504118319005,
      "flos": 17412689122560.0,
      "grad_norm": 2.6514357951992893,
      "language_loss": 0.78593302,
      "learning_rate": 1.2395562952326021e-08,
      "loss": 0.80791533,
      "num_input_tokens_seen": 173349610,
      "step": 8030,
      "time_per_iteration": 2.557713747024536
    },
    {
      "auxiliary_loss_clip": 0.01182164,
      "auxiliary_loss_mlp": 0.01030194,
      "balance_loss_clip": 1.04725122,
      "balance_loss_mlp": 1.02158093,
      "epoch": 0.9656706547225395,
      "flos": 22126970551680.0,
      "grad_norm": 3.615512598622793,
      "language_loss": 0.81482667,
      "learning_rate": 1.2309121105916309e-08,
      "loss": 0.83695024,
      "num_input_tokens_seen": 173367900,
      "step": 8031,
      "time_per_iteration": 3.567423105239868
    },
    {
      "auxiliary_loss_clip": 0.01172056,
      "auxiliary_loss_mlp": 0.01021064,
      "balance_loss_clip": 1.0464325,
      "balance_loss_mlp": 1.01398289,
      "epoch": 0.9657908976131786,
      "flos": 37049289926400.0,
      "grad_norm": 2.067612427742068,
      "language_loss": 0.69301695,
      "learning_rate": 1.222298079063222e-08,
      "loss": 0.71494812,
      "num_input_tokens_seen": 173389040,
      "step": 8032,
      "time_per_iteration": 2.6915841102600098
    },
    {
      "auxiliary_loss_clip": 0.0116836,
      "auxiliary_loss_mlp": 0.01024093,
      "balance_loss_clip": 1.04638183,
      "balance_loss_mlp": 1.0169704,
      "epoch": 0.9659111405038178,
      "flos": 24389809597440.0,
      "grad_norm": 2.0831260852684954,
      "language_loss": 0.72793889,
      "learning_rate": 1.2137142019541524e-08,
      "loss": 0.74986339,
      "num_input_tokens_seen": 173407595,
      "step": 8033,
      "time_per_iteration": 2.62493896484375
    },
    {
      "auxiliary_loss_clip": 0.01177777,
      "auxiliary_loss_mlp": 0.01025167,
      "balance_loss_clip": 1.04420662,
      "balance_loss_mlp": 1.01812756,
      "epoch": 0.9660313833944568,
      "flos": 25009412227200.0,
      "grad_norm": 2.5131158024666993,
      "language_loss": 0.73127663,
      "learning_rate": 1.2051604805666027e-08,
      "loss": 0.75330603,
      "num_input_tokens_seen": 173424720,
      "step": 8034,
      "time_per_iteration": 2.728079080581665
    },
    {
      "auxiliary_loss_clip": 0.01166412,
      "auxiliary_loss_mlp": 0.01207305,
      "balance_loss_clip": 1.04722452,
      "balance_loss_mlp": 1.00075102,
      "epoch": 0.9661516262850959,
      "flos": 11801575895040.0,
      "grad_norm": 2.3445551302134597,
      "language_loss": 0.78323871,
      "learning_rate": 1.196636916198135e-08,
      "loss": 0.80697584,
      "num_input_tokens_seen": 173442260,
      "step": 8035,
      "time_per_iteration": 2.6139628887176514
    },
    {
      "auxiliary_loss_clip": 0.01168327,
      "auxiliary_loss_mlp": 0.01021314,
      "balance_loss_clip": 1.04775047,
      "balance_loss_mlp": 1.01407182,
      "epoch": 0.9662718691757349,
      "flos": 20047778766720.0,
      "grad_norm": 1.9722753992065223,
      "language_loss": 0.77064759,
      "learning_rate": 1.1881435101418036e-08,
      "loss": 0.79254401,
      "num_input_tokens_seen": 173461675,
      "step": 8036,
      "time_per_iteration": 2.585114002227783
    },
    {
      "auxiliary_loss_clip": 0.01076085,
      "auxiliary_loss_mlp": 0.01001655,
      "balance_loss_clip": 1.01137018,
      "balance_loss_mlp": 1.0009073,
      "epoch": 0.9663921120663741,
      "flos": 68027703517440.0,
      "grad_norm": 0.7497114824596006,
      "language_loss": 0.65474635,
      "learning_rate": 1.1796802636860003e-08,
      "loss": 0.67552376,
      "num_input_tokens_seen": 173530205,
      "step": 8037,
      "time_per_iteration": 3.22802734375
    },
    {
      "auxiliary_loss_clip": 0.01165719,
      "auxiliary_loss_mlp": 0.0102468,
      "balance_loss_clip": 1.04583621,
      "balance_loss_mlp": 1.01719666,
      "epoch": 0.9665123549570132,
      "flos": 26322916769280.0,
      "grad_norm": 1.926811851929446,
      "language_loss": 0.73825085,
      "learning_rate": 1.1712471781146316e-08,
      "loss": 0.76015484,
      "num_input_tokens_seen": 173549540,
      "step": 8038,
      "time_per_iteration": 2.5757384300231934
    },
    {
      "auxiliary_loss_clip": 0.01163591,
      "auxiliary_loss_mlp": 0.01022308,
      "balance_loss_clip": 1.04459643,
      "balance_loss_mlp": 1.01498234,
      "epoch": 0.9666325978476522,
      "flos": 43941121557120.0,
      "grad_norm": 3.537707792694049,
      "language_loss": 0.67137039,
      "learning_rate": 1.1628442547069628e-08,
      "loss": 0.69322944,
      "num_input_tokens_seen": 173571740,
      "step": 8039,
      "time_per_iteration": 2.778156280517578
    },
    {
      "auxiliary_loss_clip": 0.01172318,
      "auxiliary_loss_mlp": 0.01208008,
      "balance_loss_clip": 1.04495597,
      "balance_loss_mlp": 1.00079322,
      "epoch": 0.9667528407382914,
      "flos": 21543422198400.0,
      "grad_norm": 2.2304506598790663,
      "language_loss": 0.77265894,
      "learning_rate": 1.1544714947377521e-08,
      "loss": 0.79646218,
      "num_input_tokens_seen": 173589425,
      "step": 8040,
      "time_per_iteration": 2.5859527587890625
    },
    {
      "auxiliary_loss_clip": 0.01168653,
      "auxiliary_loss_mlp": 0.0102691,
      "balance_loss_clip": 1.04783416,
      "balance_loss_mlp": 1.01836276,
      "epoch": 0.9668730836289304,
      "flos": 23878585278720.0,
      "grad_norm": 2.513731450641782,
      "language_loss": 0.69769084,
      "learning_rate": 1.1461288994770945e-08,
      "loss": 0.71964645,
      "num_input_tokens_seen": 173608500,
      "step": 8041,
      "time_per_iteration": 3.5705888271331787
    },
    {
      "auxiliary_loss_clip": 0.01168101,
      "auxiliary_loss_mlp": 0.0102655,
      "balance_loss_clip": 1.04607177,
      "balance_loss_mlp": 1.01883125,
      "epoch": 0.9669933265195695,
      "flos": 28293011971200.0,
      "grad_norm": 1.9700815305071673,
      "language_loss": 0.77238381,
      "learning_rate": 1.1378164701906002e-08,
      "loss": 0.7943303,
      "num_input_tokens_seen": 173630265,
      "step": 8042,
      "time_per_iteration": 2.6569888591766357
    },
    {
      "auxiliary_loss_clip": 0.01169713,
      "auxiliary_loss_mlp": 0.01026725,
      "balance_loss_clip": 1.04844069,
      "balance_loss_mlp": 1.0190661,
      "epoch": 0.9671135694102087,
      "flos": 22454763091200.0,
      "grad_norm": 2.5130913483305055,
      "language_loss": 0.66921866,
      "learning_rate": 1.1295342081392156e-08,
      "loss": 0.69118309,
      "num_input_tokens_seen": 173649625,
      "step": 8043,
      "time_per_iteration": 3.4674012660980225
    },
    {
      "auxiliary_loss_clip": 0.01172303,
      "auxiliary_loss_mlp": 0.01024002,
      "balance_loss_clip": 1.04321754,
      "balance_loss_mlp": 1.01640224,
      "epoch": 0.9672338123008477,
      "flos": 20155941596160.0,
      "grad_norm": 1.9468854756585607,
      "language_loss": 0.69315398,
      "learning_rate": 1.1212821145793804e-08,
      "loss": 0.7151171,
      "num_input_tokens_seen": 173669240,
      "step": 8044,
      "time_per_iteration": 2.622223138809204
    },
    {
      "auxiliary_loss_clip": 0.01172063,
      "auxiliary_loss_mlp": 0.01026152,
      "balance_loss_clip": 1.04251719,
      "balance_loss_mlp": 1.01861167,
      "epoch": 0.9673540551914868,
      "flos": 16977487939200.0,
      "grad_norm": 4.5123172243902685,
      "language_loss": 0.78705716,
      "learning_rate": 1.1130601907629156e-08,
      "loss": 0.80903935,
      "num_input_tokens_seen": 173686970,
      "step": 8045,
      "time_per_iteration": 2.724748373031616
    },
    {
      "auxiliary_loss_clip": 0.01071062,
      "auxiliary_loss_mlp": 0.01002925,
      "balance_loss_clip": 1.00999236,
      "balance_loss_mlp": 1.00222135,
      "epoch": 0.9674742980821259,
      "flos": 61892903952000.0,
      "grad_norm": 0.8120407750245905,
      "language_loss": 0.64663208,
      "learning_rate": 1.1048684379370899e-08,
      "loss": 0.66737187,
      "num_input_tokens_seen": 173747655,
      "step": 8046,
      "time_per_iteration": 3.109365224838257
    },
    {
      "auxiliary_loss_clip": 0.01163193,
      "auxiliary_loss_mlp": 0.01021994,
      "balance_loss_clip": 1.04485416,
      "balance_loss_mlp": 1.01536584,
      "epoch": 0.967594540972765,
      "flos": 18697824898560.0,
      "grad_norm": 2.1121066098260157,
      "language_loss": 0.7505247,
      "learning_rate": 1.0967068573445759e-08,
      "loss": 0.77237654,
      "num_input_tokens_seen": 173765140,
      "step": 8047,
      "time_per_iteration": 3.5388405323028564
    },
    {
      "auxiliary_loss_clip": 0.0116801,
      "auxiliary_loss_mlp": 0.01024136,
      "balance_loss_clip": 1.04261804,
      "balance_loss_mlp": 1.01675987,
      "epoch": 0.967714783863404,
      "flos": 20777411733120.0,
      "grad_norm": 2.7071901083354084,
      "language_loss": 0.65952861,
      "learning_rate": 1.0885754502234945e-08,
      "loss": 0.68145013,
      "num_input_tokens_seen": 173784800,
      "step": 8048,
      "time_per_iteration": 2.601787567138672
    },
    {
      "auxiliary_loss_clip": 0.01171773,
      "auxiliary_loss_mlp": 0.01021451,
      "balance_loss_clip": 1.04299021,
      "balance_loss_mlp": 1.01414371,
      "epoch": 0.9678350267540432,
      "flos": 23185473465600.0,
      "grad_norm": 2.5137855912336424,
      "language_loss": 0.78228587,
      "learning_rate": 1.08047421780737e-08,
      "loss": 0.80421811,
      "num_input_tokens_seen": 173803990,
      "step": 8049,
      "time_per_iteration": 2.6055924892425537
    },
    {
      "auxiliary_loss_clip": 0.01178804,
      "auxiliary_loss_mlp": 0.01207433,
      "balance_loss_clip": 1.04544032,
      "balance_loss_mlp": 1.00077009,
      "epoch": 0.9679552696446823,
      "flos": 21726063878400.0,
      "grad_norm": 2.4660783739597965,
      "language_loss": 0.7401098,
      "learning_rate": 1.0724031613251305e-08,
      "loss": 0.76397216,
      "num_input_tokens_seen": 173821890,
      "step": 8050,
      "time_per_iteration": 2.6311304569244385
    },
    {
      "auxiliary_loss_clip": 0.01176249,
      "auxiliary_loss_mlp": 0.01025811,
      "balance_loss_clip": 1.04749215,
      "balance_loss_mlp": 1.01784754,
      "epoch": 0.9680755125353213,
      "flos": 26869046129280.0,
      "grad_norm": 2.2469609419531653,
      "language_loss": 0.66308582,
      "learning_rate": 1.0643622820011744e-08,
      "loss": 0.6851064,
      "num_input_tokens_seen": 173842945,
      "step": 8051,
      "time_per_iteration": 2.6259119510650635
    },
    {
      "auxiliary_loss_clip": 0.01168806,
      "auxiliary_loss_mlp": 0.01028379,
      "balance_loss_clip": 1.04658914,
      "balance_loss_mlp": 1.02029622,
      "epoch": 0.9681957554259605,
      "flos": 28325008010880.0,
      "grad_norm": 3.603409318250973,
      "language_loss": 0.68233812,
      "learning_rate": 1.0563515810552814e-08,
      "loss": 0.70430994,
      "num_input_tokens_seen": 173859915,
      "step": 8052,
      "time_per_iteration": 2.5258264541625977
    },
    {
      "auxiliary_loss_clip": 0.01168748,
      "auxiliary_loss_mlp": 0.01025222,
      "balance_loss_clip": 1.04985881,
      "balance_loss_mlp": 1.01832914,
      "epoch": 0.9683159983165995,
      "flos": 20557674282240.0,
      "grad_norm": 2.1727459912258924,
      "language_loss": 0.73290199,
      "learning_rate": 1.0483710597026795e-08,
      "loss": 0.75484169,
      "num_input_tokens_seen": 173879775,
      "step": 8053,
      "time_per_iteration": 2.5389904975891113
    },
    {
      "auxiliary_loss_clip": 0.01172536,
      "auxiliary_loss_mlp": 0.01028482,
      "balance_loss_clip": 1.04191971,
      "balance_loss_mlp": 1.02104974,
      "epoch": 0.9684362412072386,
      "flos": 24207958016640.0,
      "grad_norm": 2.168247326097946,
      "language_loss": 0.74132746,
      "learning_rate": 1.0404207191540227e-08,
      "loss": 0.76333761,
      "num_input_tokens_seen": 173900230,
      "step": 8054,
      "time_per_iteration": 2.607381820678711
    },
    {
      "auxiliary_loss_clip": 0.0116569,
      "auxiliary_loss_mlp": 0.01023738,
      "balance_loss_clip": 1.0467248,
      "balance_loss_mlp": 1.0164454,
      "epoch": 0.9685564840978778,
      "flos": 22346241125760.0,
      "grad_norm": 2.202764059349207,
      "language_loss": 0.75071687,
      "learning_rate": 1.0325005606153236e-08,
      "loss": 0.7726112,
      "num_input_tokens_seen": 173919690,
      "step": 8055,
      "time_per_iteration": 2.5890934467315674
    },
    {
      "auxiliary_loss_clip": 0.01179865,
      "auxiliary_loss_mlp": 0.01025577,
      "balance_loss_clip": 1.04094267,
      "balance_loss_mlp": 1.01859713,
      "epoch": 0.9686767269885168,
      "flos": 14386389477120.0,
      "grad_norm": 2.62778296691951,
      "language_loss": 0.79463893,
      "learning_rate": 1.0246105852881104e-08,
      "loss": 0.81669331,
      "num_input_tokens_seen": 173934790,
      "step": 8056,
      "time_per_iteration": 2.6546554565429688
    },
    {
      "auxiliary_loss_clip": 0.01167874,
      "auxiliary_loss_mlp": 0.01021245,
      "balance_loss_clip": 1.04735327,
      "balance_loss_mlp": 1.01382101,
      "epoch": 0.9687969698791559,
      "flos": 21287630471040.0,
      "grad_norm": 2.1036236344637143,
      "language_loss": 0.78928053,
      "learning_rate": 1.0167507943692476e-08,
      "loss": 0.81117177,
      "num_input_tokens_seen": 173953875,
      "step": 8057,
      "time_per_iteration": 2.5910134315490723
    },
    {
      "auxiliary_loss_clip": 0.01165834,
      "auxiliary_loss_mlp": 0.01034023,
      "balance_loss_clip": 1.04651308,
      "balance_loss_mlp": 1.02604795,
      "epoch": 0.968917212769795,
      "flos": 19828328624640.0,
      "grad_norm": 3.5804145278342814,
      "language_loss": 0.71700585,
      "learning_rate": 1.008921189051093e-08,
      "loss": 0.73900443,
      "num_input_tokens_seen": 173971220,
      "step": 8058,
      "time_per_iteration": 3.496574878692627
    },
    {
      "auxiliary_loss_clip": 0.0116772,
      "auxiliary_loss_mlp": 0.01025161,
      "balance_loss_clip": 1.04789138,
      "balance_loss_mlp": 1.01752234,
      "epoch": 0.9690374556604341,
      "flos": 21681749473920.0,
      "grad_norm": 2.1138412121167645,
      "language_loss": 0.77557194,
      "learning_rate": 1.0011217705213848e-08,
      "loss": 0.79750073,
      "num_input_tokens_seen": 173989095,
      "step": 8059,
      "time_per_iteration": 2.6248061656951904
    },
    {
      "auxiliary_loss_clip": 0.01167007,
      "auxiliary_loss_mlp": 0.01023471,
      "balance_loss_clip": 1.04691458,
      "balance_loss_mlp": 1.0167954,
      "epoch": 0.9691576985510731,
      "flos": 32635437851520.0,
      "grad_norm": 1.6744417770381868,
      "language_loss": 0.74933368,
      "learning_rate": 9.933525399632658e-09,
      "loss": 0.77123845,
      "num_input_tokens_seen": 174007330,
      "step": 8060,
      "time_per_iteration": 2.7078659534454346
    },
    {
      "auxiliary_loss_clip": 0.01167575,
      "auxiliary_loss_mlp": 0.01023352,
      "balance_loss_clip": 1.04372573,
      "balance_loss_mlp": 1.01530552,
      "epoch": 0.9692779414417123,
      "flos": 35663174040960.0,
      "grad_norm": 1.784523092820475,
      "language_loss": 0.65162289,
      "learning_rate": 9.856134985553488e-09,
      "loss": 0.67353219,
      "num_input_tokens_seen": 174027055,
      "step": 8061,
      "time_per_iteration": 2.6906182765960693
    },
    {
      "auxiliary_loss_clip": 0.01166554,
      "auxiliary_loss_mlp": 0.01022655,
      "balance_loss_clip": 1.04727268,
      "balance_loss_mlp": 1.01528537,
      "epoch": 0.9693981843323514,
      "flos": 28366952117760.0,
      "grad_norm": 1.6751439562785446,
      "language_loss": 0.73637849,
      "learning_rate": 9.77904647471628e-09,
      "loss": 0.75827056,
      "num_input_tokens_seen": 174050235,
      "step": 8062,
      "time_per_iteration": 2.665285348892212
    },
    {
      "auxiliary_loss_clip": 0.01169254,
      "auxiliary_loss_mlp": 0.01025286,
      "balance_loss_clip": 1.03921604,
      "balance_loss_mlp": 1.01797819,
      "epoch": 0.9695184272229904,
      "flos": 23622865378560.0,
      "grad_norm": 1.5454964138268674,
      "language_loss": 0.74065983,
      "learning_rate": 9.702259878815454e-09,
      "loss": 0.76260531,
      "num_input_tokens_seen": 174070560,
      "step": 8063,
      "time_per_iteration": 2.805687427520752
    },
    {
      "auxiliary_loss_clip": 0.01172972,
      "auxiliary_loss_mlp": 0.01026893,
      "balance_loss_clip": 1.04805994,
      "balance_loss_mlp": 1.01892984,
      "epoch": 0.9696386701136296,
      "flos": 23294677789440.0,
      "grad_norm": 2.0629983292967062,
      "language_loss": 0.7403245,
      "learning_rate": 9.625775209499254e-09,
      "loss": 0.76232314,
      "num_input_tokens_seen": 174090565,
      "step": 8064,
      "time_per_iteration": 2.6563096046447754
    },
    {
      "auxiliary_loss_clip": 0.01166744,
      "auxiliary_loss_mlp": 0.01025704,
      "balance_loss_clip": 1.03871059,
      "balance_loss_mlp": 1.01846194,
      "epoch": 0.9697589130042686,
      "flos": 15121876360320.0,
      "grad_norm": 3.4585423372002615,
      "language_loss": 0.74480247,
      "learning_rate": 9.549592478370172e-09,
      "loss": 0.76672703,
      "num_input_tokens_seen": 174108745,
      "step": 8065,
      "time_per_iteration": 2.60957407951355
    },
    {
      "auxiliary_loss_clip": 0.01169185,
      "auxiliary_loss_mlp": 0.0102231,
      "balance_loss_clip": 1.04436612,
      "balance_loss_mlp": 1.01519322,
      "epoch": 0.9698791558949077,
      "flos": 18879532824960.0,
      "grad_norm": 1.8601665894557669,
      "language_loss": 0.79203582,
      "learning_rate": 9.473711696985632e-09,
      "loss": 0.81395078,
      "num_input_tokens_seen": 174128075,
      "step": 8066,
      "time_per_iteration": 2.600390672683716
    },
    {
      "auxiliary_loss_clip": 0.01169832,
      "auxiliary_loss_mlp": 0.01025019,
      "balance_loss_clip": 1.04299653,
      "balance_loss_mlp": 1.01757407,
      "epoch": 0.9699993987855468,
      "flos": 17931455297280.0,
      "grad_norm": 2.4480585427125305,
      "language_loss": 0.75527596,
      "learning_rate": 9.398132876856201e-09,
      "loss": 0.77722442,
      "num_input_tokens_seen": 174147040,
      "step": 8067,
      "time_per_iteration": 3.474245071411133
    },
    {
      "auxiliary_loss_clip": 0.01090515,
      "auxiliary_loss_mlp": 0.01000744,
      "balance_loss_clip": 1.01230872,
      "balance_loss_mlp": 1.00003457,
      "epoch": 0.9701196416761859,
      "flos": 67182186297600.0,
      "grad_norm": 0.7746489280186845,
      "language_loss": 0.60758114,
      "learning_rate": 9.322856029447379e-09,
      "loss": 0.62849379,
      "num_input_tokens_seen": 174208225,
      "step": 8068,
      "time_per_iteration": 3.1277098655700684
    },
    {
      "auxiliary_loss_clip": 0.011636,
      "auxiliary_loss_mlp": 0.01023156,
      "balance_loss_clip": 1.0470022,
      "balance_loss_mlp": 1.01599109,
      "epoch": 0.970239884566825,
      "flos": 24277804012800.0,
      "grad_norm": 2.749731342784404,
      "language_loss": 0.80831885,
      "learning_rate": 9.247881166178695e-09,
      "loss": 0.83018643,
      "num_input_tokens_seen": 174226935,
      "step": 8069,
      "time_per_iteration": 2.6816015243530273
    },
    {
      "auxiliary_loss_clip": 0.01183915,
      "auxiliary_loss_mlp": 0.01026002,
      "balance_loss_clip": 1.04514289,
      "balance_loss_mlp": 1.01864648,
      "epoch": 0.970360127457464,
      "flos": 25301689194240.0,
      "grad_norm": 2.1548329830071924,
      "language_loss": 0.76762366,
      "learning_rate": 9.173208298423274e-09,
      "loss": 0.7897228,
      "num_input_tokens_seen": 174248140,
      "step": 8070,
      "time_per_iteration": 3.6969974040985107
    },
    {
      "auxiliary_loss_clip": 0.01173055,
      "auxiliary_loss_mlp": 0.01207938,
      "balance_loss_clip": 1.04109669,
      "balance_loss_mlp": 1.00070941,
      "epoch": 0.9704803703481032,
      "flos": 29572473398400.0,
      "grad_norm": 1.6758599416657014,
      "language_loss": 0.76309204,
      "learning_rate": 9.09883743750961e-09,
      "loss": 0.78690195,
      "num_input_tokens_seen": 174271030,
      "step": 8071,
      "time_per_iteration": 2.7744297981262207
    },
    {
      "auxiliary_loss_clip": 0.01170133,
      "auxiliary_loss_mlp": 0.01021522,
      "balance_loss_clip": 1.04414845,
      "balance_loss_mlp": 1.01441741,
      "epoch": 0.9706006132387422,
      "flos": 17380046638080.0,
      "grad_norm": 1.7195368194280451,
      "language_loss": 0.83949983,
      "learning_rate": 9.024768594719124e-09,
      "loss": 0.86141634,
      "num_input_tokens_seen": 174289410,
      "step": 8072,
      "time_per_iteration": 2.6334755420684814
    },
    {
      "auxiliary_loss_clip": 0.01174209,
      "auxiliary_loss_mlp": 0.01021669,
      "balance_loss_clip": 1.04449964,
      "balance_loss_mlp": 1.01460004,
      "epoch": 0.9707208561293813,
      "flos": 18186421011840.0,
      "grad_norm": 8.169671726435443,
      "language_loss": 0.72824216,
      "learning_rate": 8.95100178128816e-09,
      "loss": 0.75020099,
      "num_input_tokens_seen": 174308550,
      "step": 8073,
      "time_per_iteration": 3.528825044631958
    },
    {
      "auxiliary_loss_clip": 0.0117237,
      "auxiliary_loss_mlp": 0.01025049,
      "balance_loss_clip": 1.04445589,
      "balance_loss_mlp": 1.01705587,
      "epoch": 0.9708410990200205,
      "flos": 31248388212480.0,
      "grad_norm": 2.0255871124259883,
      "language_loss": 0.70332634,
      "learning_rate": 8.877537008407321e-09,
      "loss": 0.72530055,
      "num_input_tokens_seen": 174328600,
      "step": 8074,
      "time_per_iteration": 2.6747212409973145
    },
    {
      "auxiliary_loss_clip": 0.01174889,
      "auxiliary_loss_mlp": 0.01025351,
      "balance_loss_clip": 1.04473031,
      "balance_loss_mlp": 1.01796031,
      "epoch": 0.9709613419106595,
      "flos": 30554450386560.0,
      "grad_norm": 1.5485142066314452,
      "language_loss": 0.68699372,
      "learning_rate": 8.804374287221028e-09,
      "loss": 0.70899618,
      "num_input_tokens_seen": 174349835,
      "step": 8075,
      "time_per_iteration": 2.7035181522369385
    },
    {
      "auxiliary_loss_clip": 0.01165113,
      "auxiliary_loss_mlp": 0.01023641,
      "balance_loss_clip": 1.0370779,
      "balance_loss_mlp": 1.01590705,
      "epoch": 0.9710815848012986,
      "flos": 23730166281600.0,
      "grad_norm": 1.813958362504971,
      "language_loss": 0.84759402,
      "learning_rate": 8.731513628827958e-09,
      "loss": 0.86948156,
      "num_input_tokens_seen": 174369200,
      "step": 8076,
      "time_per_iteration": 2.669689893722534
    },
    {
      "auxiliary_loss_clip": 0.01170297,
      "auxiliary_loss_mlp": 0.01023242,
      "balance_loss_clip": 1.0458107,
      "balance_loss_mlp": 1.01578903,
      "epoch": 0.9712018276919377,
      "flos": 23761875012480.0,
      "grad_norm": 4.6930708031437245,
      "language_loss": 0.82776487,
      "learning_rate": 8.658955044280825e-09,
      "loss": 0.84970021,
      "num_input_tokens_seen": 174388125,
      "step": 8077,
      "time_per_iteration": 2.5742573738098145
    },
    {
      "auxiliary_loss_clip": 0.01165761,
      "auxiliary_loss_mlp": 0.01020956,
      "balance_loss_clip": 1.04579425,
      "balance_loss_mlp": 1.01341307,
      "epoch": 0.9713220705825768,
      "flos": 23330983461120.0,
      "grad_norm": 1.533723726668541,
      "language_loss": 0.77495599,
      "learning_rate": 8.586698544587268e-09,
      "loss": 0.79682314,
      "num_input_tokens_seen": 174409735,
      "step": 8078,
      "time_per_iteration": 2.560530662536621
    },
    {
      "auxiliary_loss_clip": 0.01165371,
      "auxiliary_loss_mlp": 0.01026513,
      "balance_loss_clip": 1.04296637,
      "balance_loss_mlp": 1.01839507,
      "epoch": 0.9714423134732159,
      "flos": 22200946611840.0,
      "grad_norm": 2.0331509075073932,
      "language_loss": 0.73786503,
      "learning_rate": 8.514744140707853e-09,
      "loss": 0.75978386,
      "num_input_tokens_seen": 174428875,
      "step": 8079,
      "time_per_iteration": 2.655242681503296
    },
    {
      "auxiliary_loss_clip": 0.01164115,
      "auxiliary_loss_mlp": 0.01021228,
      "balance_loss_clip": 1.0457896,
      "balance_loss_mlp": 1.01427484,
      "epoch": 0.971562556363855,
      "flos": 20229917656320.0,
      "grad_norm": 2.1361359815679366,
      "language_loss": 0.76178777,
      "learning_rate": 8.443091843558515e-09,
      "loss": 0.78364122,
      "num_input_tokens_seen": 174447960,
      "step": 8080,
      "time_per_iteration": 2.5629682540893555
    },
    {
      "auxiliary_loss_clip": 0.01166986,
      "auxiliary_loss_mlp": 0.01021549,
      "balance_loss_clip": 1.04314613,
      "balance_loss_mlp": 1.01362181,
      "epoch": 0.9716827992544941,
      "flos": 24970197553920.0,
      "grad_norm": 2.556323756303274,
      "language_loss": 0.64305484,
      "learning_rate": 8.37174166400878e-09,
      "loss": 0.66494018,
      "num_input_tokens_seen": 174463535,
      "step": 8081,
      "time_per_iteration": 2.6279664039611816
    },
    {
      "auxiliary_loss_clip": 0.01165806,
      "auxiliary_loss_mlp": 0.01023728,
      "balance_loss_clip": 1.04809725,
      "balance_loss_mlp": 1.01656091,
      "epoch": 0.9718030421451331,
      "flos": 24681476033280.0,
      "grad_norm": 2.153782613567578,
      "language_loss": 0.84886289,
      "learning_rate": 8.300693612881992e-09,
      "loss": 0.87075824,
      "num_input_tokens_seen": 174483600,
      "step": 8082,
      "time_per_iteration": 2.6589691638946533
    },
    {
      "auxiliary_loss_clip": 0.01167904,
      "auxiliary_loss_mlp": 0.01207363,
      "balance_loss_clip": 1.04625952,
      "balance_loss_mlp": 1.0007242,
      "epoch": 0.9719232850357723,
      "flos": 22090700793600.0,
      "grad_norm": 2.0800175258973916,
      "language_loss": 0.81223321,
      "learning_rate": 8.22994770095664e-09,
      "loss": 0.83598596,
      "num_input_tokens_seen": 174502175,
      "step": 8083,
      "time_per_iteration": 2.64067006111145
    },
    {
      "auxiliary_loss_clip": 0.01171635,
      "auxiliary_loss_mlp": 0.01027875,
      "balance_loss_clip": 1.04740787,
      "balance_loss_mlp": 1.01997161,
      "epoch": 0.9720435279264114,
      "flos": 23656908493440.0,
      "grad_norm": 2.092064495107422,
      "language_loss": 0.7511903,
      "learning_rate": 8.159503938964585e-09,
      "loss": 0.77318543,
      "num_input_tokens_seen": 174519495,
      "step": 8084,
      "time_per_iteration": 2.6214041709899902
    },
    {
      "auxiliary_loss_clip": 0.01168425,
      "auxiliary_loss_mlp": 0.01017017,
      "balance_loss_clip": 1.04337811,
      "balance_loss_mlp": 1.01000166,
      "epoch": 0.9721637708170504,
      "flos": 28365910623360.0,
      "grad_norm": 1.8239159481518528,
      "language_loss": 0.70849568,
      "learning_rate": 8.089362337592164e-09,
      "loss": 0.73035014,
      "num_input_tokens_seen": 174543120,
      "step": 8085,
      "time_per_iteration": 3.6718244552612305
    },
    {
      "auxiliary_loss_clip": 0.01168391,
      "auxiliary_loss_mlp": 0.0102669,
      "balance_loss_clip": 1.04342818,
      "balance_loss_mlp": 1.01962686,
      "epoch": 0.9722840137076896,
      "flos": 29130807767040.0,
      "grad_norm": 2.219005441151757,
      "language_loss": 0.72126144,
      "learning_rate": 8.019522907479536e-09,
      "loss": 0.74321234,
      "num_input_tokens_seen": 174563480,
      "step": 8086,
      "time_per_iteration": 2.66762638092041
    },
    {
      "auxiliary_loss_clip": 0.01172575,
      "auxiliary_loss_mlp": 0.01026776,
      "balance_loss_clip": 1.04746342,
      "balance_loss_mlp": 1.01896834,
      "epoch": 0.9724042565983286,
      "flos": 19243954258560.0,
      "grad_norm": 2.0677384052293166,
      "language_loss": 0.77297747,
      "learning_rate": 7.949985659221558e-09,
      "loss": 0.79497099,
      "num_input_tokens_seen": 174580745,
      "step": 8087,
      "time_per_iteration": 2.702080011367798
    },
    {
      "auxiliary_loss_clip": 0.01173988,
      "auxiliary_loss_mlp": 0.0102771,
      "balance_loss_clip": 1.04457319,
      "balance_loss_mlp": 1.02095687,
      "epoch": 0.9725244994889677,
      "flos": 23039676161280.0,
      "grad_norm": 1.999261780280519,
      "language_loss": 0.79052913,
      "learning_rate": 7.880750603366904e-09,
      "loss": 0.81254613,
      "num_input_tokens_seen": 174599615,
      "step": 8088,
      "time_per_iteration": 2.6307780742645264
    },
    {
      "auxiliary_loss_clip": 0.01181815,
      "auxiliary_loss_mlp": 0.01026759,
      "balance_loss_clip": 1.04143417,
      "balance_loss_mlp": 1.01839662,
      "epoch": 0.9726447423796069,
      "flos": 23367468700800.0,
      "grad_norm": 4.611413365536493,
      "language_loss": 0.79726535,
      "learning_rate": 7.811817750418282e-09,
      "loss": 0.81935114,
      "num_input_tokens_seen": 174618375,
      "step": 8089,
      "time_per_iteration": 2.6177828311920166
    },
    {
      "auxiliary_loss_clip": 0.01170396,
      "auxiliary_loss_mlp": 0.01022,
      "balance_loss_clip": 1.04332519,
      "balance_loss_mlp": 1.01410818,
      "epoch": 0.9727649852702459,
      "flos": 26541648639360.0,
      "grad_norm": 1.892306678065801,
      "language_loss": 0.80301666,
      "learning_rate": 7.743187110833105e-09,
      "loss": 0.82494068,
      "num_input_tokens_seen": 174641135,
      "step": 8090,
      "time_per_iteration": 2.7901501655578613
    },
    {
      "auxiliary_loss_clip": 0.01173168,
      "auxiliary_loss_mlp": 0.01016886,
      "balance_loss_clip": 1.04326487,
      "balance_loss_mlp": 1.00997829,
      "epoch": 0.972885228160885,
      "flos": 20522338277760.0,
      "grad_norm": 2.2679842930328014,
      "language_loss": 0.81064725,
      "learning_rate": 7.674858695022602e-09,
      "loss": 0.83254778,
      "num_input_tokens_seen": 174659490,
      "step": 8091,
      "time_per_iteration": 2.6358187198638916
    },
    {
      "auxiliary_loss_clip": 0.01169276,
      "auxiliary_loss_mlp": 0.01025432,
      "balance_loss_clip": 1.04914474,
      "balance_loss_mlp": 1.01807129,
      "epoch": 0.9730054710515241,
      "flos": 17566064196480.0,
      "grad_norm": 2.4406673703397663,
      "language_loss": 0.75508893,
      "learning_rate": 7.606832513351591e-09,
      "loss": 0.77703595,
      "num_input_tokens_seen": 174677440,
      "step": 8092,
      "time_per_iteration": 2.559046506881714
    },
    {
      "auxiliary_loss_clip": 0.01063023,
      "auxiliary_loss_mlp": 0.01198636,
      "balance_loss_clip": 1.01039493,
      "balance_loss_mlp": 0.99988997,
      "epoch": 0.9731257139421632,
      "flos": 68972010117120.0,
      "grad_norm": 0.83030515595399,
      "language_loss": 0.63907325,
      "learning_rate": 7.539108576140264e-09,
      "loss": 0.66168988,
      "num_input_tokens_seen": 174741550,
      "step": 8093,
      "time_per_iteration": 3.1807801723480225
    },
    {
      "auxiliary_loss_clip": 0.01175441,
      "auxiliary_loss_mlp": 0.0101761,
      "balance_loss_clip": 1.04107773,
      "balance_loss_mlp": 1.01134896,
      "epoch": 0.9732459568328022,
      "flos": 18478841633280.0,
      "grad_norm": 2.0004519944557413,
      "language_loss": 0.70384657,
      "learning_rate": 7.471686893661732e-09,
      "loss": 0.72577703,
      "num_input_tokens_seen": 174759845,
      "step": 8094,
      "time_per_iteration": 3.6308553218841553
    },
    {
      "auxiliary_loss_clip": 0.01168287,
      "auxiliary_loss_mlp": 0.01024573,
      "balance_loss_clip": 1.04386687,
      "balance_loss_mlp": 1.01735246,
      "epoch": 0.9733661997234414,
      "flos": 20883886623360.0,
      "grad_norm": 1.946669939217944,
      "language_loss": 0.64369643,
      "learning_rate": 7.4045674761442636e-09,
      "loss": 0.66562504,
      "num_input_tokens_seen": 174777175,
      "step": 8095,
      "time_per_iteration": 2.6791555881500244
    },
    {
      "auxiliary_loss_clip": 0.01165239,
      "auxiliary_loss_mlp": 0.01207383,
      "balance_loss_clip": 1.04672635,
      "balance_loss_mlp": 1.00063479,
      "epoch": 0.9734864426140805,
      "flos": 23766795175680.0,
      "grad_norm": 1.9584324310947887,
      "language_loss": 0.74329573,
      "learning_rate": 7.337750333769488e-09,
      "loss": 0.76702201,
      "num_input_tokens_seen": 174796980,
      "step": 8096,
      "time_per_iteration": 2.643014669418335
    },
    {
      "auxiliary_loss_clip": 0.01176844,
      "auxiliary_loss_mlp": 0.01023791,
      "balance_loss_clip": 1.04116726,
      "balance_loss_mlp": 1.01563704,
      "epoch": 0.9736066855047195,
      "flos": 35042422176000.0,
      "grad_norm": 2.3513508869333144,
      "language_loss": 0.72991657,
      "learning_rate": 7.2712354766737425e-09,
      "loss": 0.75192297,
      "num_input_tokens_seen": 174817310,
      "step": 8097,
      "time_per_iteration": 3.687931776046753
    },
    {
      "auxiliary_loss_clip": 0.01166171,
      "auxiliary_loss_mlp": 0.01024958,
      "balance_loss_clip": 1.04430044,
      "balance_loss_mlp": 1.01685762,
      "epoch": 0.9737269283953586,
      "flos": 20410620001920.0,
      "grad_norm": 1.5859447141534846,
      "language_loss": 0.81194746,
      "learning_rate": 7.2050229149469565e-09,
      "loss": 0.83385873,
      "num_input_tokens_seen": 174837320,
      "step": 8098,
      "time_per_iteration": 2.6143062114715576
    },
    {
      "auxiliary_loss_clip": 0.01177601,
      "auxiliary_loss_mlp": 0.01025601,
      "balance_loss_clip": 1.0416255,
      "balance_loss_mlp": 1.01807344,
      "epoch": 0.9738471712859977,
      "flos": 28911680847360.0,
      "grad_norm": 1.8208588076038528,
      "language_loss": 0.63339299,
      "learning_rate": 7.139112658633984e-09,
      "loss": 0.65542501,
      "num_input_tokens_seen": 174857470,
      "step": 8099,
      "time_per_iteration": 2.7154479026794434
    },
    {
      "auxiliary_loss_clip": 0.01172411,
      "auxiliary_loss_mlp": 0.01022488,
      "balance_loss_clip": 1.0440836,
      "balance_loss_mlp": 1.01548409,
      "epoch": 0.9739674141766368,
      "flos": 27782326356480.0,
      "grad_norm": 1.9677287383267783,
      "language_loss": 0.70330459,
      "learning_rate": 7.073504717733048e-09,
      "loss": 0.72525358,
      "num_input_tokens_seen": 174877035,
      "step": 8100,
      "time_per_iteration": 3.786377191543579
    },
    {
      "auxiliary_loss_clip": 0.01087231,
      "auxiliary_loss_mlp": 0.0100235,
      "balance_loss_clip": 1.01274848,
      "balance_loss_mlp": 1.00165272,
      "epoch": 0.9740876570672758,
      "flos": 68863057188480.0,
      "grad_norm": 0.7309262406110357,
      "language_loss": 0.571329,
      "learning_rate": 7.008199102196855e-09,
      "loss": 0.59222472,
      "num_input_tokens_seen": 174938460,
      "step": 8101,
      "time_per_iteration": 3.20055890083313
    },
    {
      "auxiliary_loss_clip": 0.01069057,
      "auxiliary_loss_mlp": 0.00999961,
      "balance_loss_clip": 1.01147151,
      "balance_loss_mlp": 0.9992246,
      "epoch": 0.974207899957915,
      "flos": 58236622646400.0,
      "grad_norm": 0.794412935239271,
      "language_loss": 0.58972979,
      "learning_rate": 6.9431958219321464e-09,
      "loss": 0.61041999,
      "num_input_tokens_seen": 174994625,
      "step": 8102,
      "time_per_iteration": 3.134103775024414
    },
    {
      "auxiliary_loss_clip": 0.01172601,
      "auxiliary_loss_mlp": 0.01027019,
      "balance_loss_clip": 1.04457664,
      "balance_loss_mlp": 1.01924324,
      "epoch": 0.9743281428485541,
      "flos": 22600057605120.0,
      "grad_norm": 1.838629963318565,
      "language_loss": 0.77697766,
      "learning_rate": 6.878494886800146e-09,
      "loss": 0.79897392,
      "num_input_tokens_seen": 175015400,
      "step": 8103,
      "time_per_iteration": 2.689694881439209
    },
    {
      "auxiliary_loss_clip": 0.01173053,
      "auxiliary_loss_mlp": 0.01024495,
      "balance_loss_clip": 1.04635239,
      "balance_loss_mlp": 1.01716661,
      "epoch": 0.9744483857391931,
      "flos": 20008815488640.0,
      "grad_norm": 1.9248791882961231,
      "language_loss": 0.76389825,
      "learning_rate": 6.814096306615669e-09,
      "loss": 0.78587377,
      "num_input_tokens_seen": 175033540,
      "step": 8104,
      "time_per_iteration": 2.604443311691284
    },
    {
      "auxiliary_loss_clip": 0.01176892,
      "auxiliary_loss_mlp": 0.01025539,
      "balance_loss_clip": 1.04334021,
      "balance_loss_mlp": 1.01804924,
      "epoch": 0.9745686286298323,
      "flos": 17675268520320.0,
      "grad_norm": 2.685087896727446,
      "language_loss": 0.65901554,
      "learning_rate": 6.750000091148011e-09,
      "loss": 0.68103981,
      "num_input_tokens_seen": 175050835,
      "step": 8105,
      "time_per_iteration": 2.7133665084838867
    },
    {
      "auxiliary_loss_clip": 0.01167641,
      "auxiliary_loss_mlp": 0.01023632,
      "balance_loss_clip": 1.04856348,
      "balance_loss_mlp": 1.01554942,
      "epoch": 0.9746888715204713,
      "flos": 29460252332160.0,
      "grad_norm": 2.283917224719123,
      "language_loss": 0.72424543,
      "learning_rate": 6.686206250120729e-09,
      "loss": 0.74615812,
      "num_input_tokens_seen": 175072330,
      "step": 8106,
      "time_per_iteration": 2.648153305053711
    },
    {
      "auxiliary_loss_clip": 0.01178767,
      "auxiliary_loss_mlp": 0.01020798,
      "balance_loss_clip": 1.0411495,
      "balance_loss_mlp": 1.01374626,
      "epoch": 0.9748091144111104,
      "flos": 18479308510080.0,
      "grad_norm": 1.8755481791552258,
      "language_loss": 0.74981004,
      "learning_rate": 6.622714793210749e-09,
      "loss": 0.7718057,
      "num_input_tokens_seen": 175091250,
      "step": 8107,
      "time_per_iteration": 2.634965419769287
    },
    {
      "auxiliary_loss_clip": 0.01166757,
      "auxiliary_loss_mlp": 0.01019087,
      "balance_loss_clip": 1.0470531,
      "balance_loss_mlp": 1.01199675,
      "epoch": 0.9749293573017496,
      "flos": 20665154753280.0,
      "grad_norm": 1.9285729917341599,
      "language_loss": 0.78976309,
      "learning_rate": 6.559525730050364e-09,
      "loss": 0.81162149,
      "num_input_tokens_seen": 175111350,
      "step": 8108,
      "time_per_iteration": 2.633223533630371
    },
    {
      "auxiliary_loss_clip": 0.01178006,
      "auxiliary_loss_mlp": 0.0102271,
      "balance_loss_clip": 1.04462421,
      "balance_loss_mlp": 1.0154798,
      "epoch": 0.9750496001923886,
      "flos": 18478590238080.0,
      "grad_norm": 2.0060853753947163,
      "language_loss": 0.75881499,
      "learning_rate": 6.496639070224574e-09,
      "loss": 0.78082216,
      "num_input_tokens_seen": 175129835,
      "step": 8109,
      "time_per_iteration": 2.6994123458862305
    },
    {
      "auxiliary_loss_clip": 0.01171672,
      "auxiliary_loss_mlp": 0.01020334,
      "balance_loss_clip": 1.04662776,
      "balance_loss_mlp": 1.01393867,
      "epoch": 0.9751698430830277,
      "flos": 19572967860480.0,
      "grad_norm": 2.5620549884072905,
      "language_loss": 0.83538526,
      "learning_rate": 6.4340548232739714e-09,
      "loss": 0.85730529,
      "num_input_tokens_seen": 175146035,
      "step": 8110,
      "time_per_iteration": 2.575545310974121
    },
    {
      "auxiliary_loss_clip": 0.01178633,
      "auxiliary_loss_mlp": 0.01024638,
      "balance_loss_clip": 1.04339921,
      "balance_loss_mlp": 1.01731825,
      "epoch": 0.9752900859736668,
      "flos": 23550325862400.0,
      "grad_norm": 1.9010551989952742,
      "language_loss": 0.79232109,
      "learning_rate": 6.371772998692071e-09,
      "loss": 0.81435382,
      "num_input_tokens_seen": 175165290,
      "step": 8111,
      "time_per_iteration": 2.7264442443847656
    },
    {
      "auxiliary_loss_clip": 0.01179301,
      "auxiliary_loss_mlp": 0.01018696,
      "balance_loss_clip": 1.041924,
      "balance_loss_mlp": 1.01134634,
      "epoch": 0.9754103288643059,
      "flos": 20303211358080.0,
      "grad_norm": 6.922137086161398,
      "language_loss": 0.64700162,
      "learning_rate": 6.309793605927094e-09,
      "loss": 0.66898167,
      "num_input_tokens_seen": 175183610,
      "step": 8112,
      "time_per_iteration": 3.6381747722625732
    },
    {
      "auxiliary_loss_clip": 0.01175203,
      "auxiliary_loss_mlp": 0.01020334,
      "balance_loss_clip": 1.04304361,
      "balance_loss_mlp": 1.01296139,
      "epoch": 0.975530571754945,
      "flos": 19350680544000.0,
      "grad_norm": 2.11390294569159,
      "language_loss": 0.80385727,
      "learning_rate": 6.248116654381297e-09,
      "loss": 0.8258127,
      "num_input_tokens_seen": 175202080,
      "step": 8113,
      "time_per_iteration": 2.6727147102355957
    },
    {
      "auxiliary_loss_clip": 0.01172423,
      "auxiliary_loss_mlp": 0.01022354,
      "balance_loss_clip": 1.04190028,
      "balance_loss_mlp": 1.01537466,
      "epoch": 0.9756508146455841,
      "flos": 23583399310080.0,
      "grad_norm": 1.8041739301513873,
      "language_loss": 0.72930545,
      "learning_rate": 6.186742153410751e-09,
      "loss": 0.75125325,
      "num_input_tokens_seen": 175221575,
      "step": 8114,
      "time_per_iteration": 2.6359500885009766
    },
    {
      "auxiliary_loss_clip": 0.01170161,
      "auxiliary_loss_mlp": 0.01026907,
      "balance_loss_clip": 1.04485679,
      "balance_loss_mlp": 1.01885438,
      "epoch": 0.9757710575362232,
      "flos": 22966921163520.0,
      "grad_norm": 2.368002316980022,
      "language_loss": 0.87696147,
      "learning_rate": 6.125670112326453e-09,
      "loss": 0.89893216,
      "num_input_tokens_seen": 175240835,
      "step": 8115,
      "time_per_iteration": 2.6889429092407227
    },
    {
      "auxiliary_loss_clip": 0.01169056,
      "auxiliary_loss_mlp": 0.0102284,
      "balance_loss_clip": 1.04324675,
      "balance_loss_mlp": 1.0154494,
      "epoch": 0.9758913004268622,
      "flos": 27966009530880.0,
      "grad_norm": 1.733898363444791,
      "language_loss": 0.70307982,
      "learning_rate": 6.064900540392548e-09,
      "loss": 0.72499871,
      "num_input_tokens_seen": 175262930,
      "step": 8116,
      "time_per_iteration": 2.6623969078063965
    },
    {
      "auxiliary_loss_clip": 0.01166125,
      "auxiliary_loss_mlp": 0.01020593,
      "balance_loss_clip": 1.04489994,
      "balance_loss_mlp": 1.01355433,
      "epoch": 0.9760115433175014,
      "flos": 22200156512640.0,
      "grad_norm": 2.169911370532606,
      "language_loss": 0.78721118,
      "learning_rate": 6.0044334468278835e-09,
      "loss": 0.80907834,
      "num_input_tokens_seen": 175282275,
      "step": 8117,
      "time_per_iteration": 2.6609160900115967
    },
    {
      "auxiliary_loss_clip": 0.01180627,
      "auxiliary_loss_mlp": 0.01026613,
      "balance_loss_clip": 1.04296529,
      "balance_loss_mlp": 1.01880503,
      "epoch": 0.9761317862081405,
      "flos": 26250736389120.0,
      "grad_norm": 1.971010513656392,
      "language_loss": 0.7161305,
      "learning_rate": 5.944268840805345e-09,
      "loss": 0.73820287,
      "num_input_tokens_seen": 175303020,
      "step": 8118,
      "time_per_iteration": 2.7159528732299805
    },
    {
      "auxiliary_loss_clip": 0.01164334,
      "auxiliary_loss_mlp": 0.01022464,
      "balance_loss_clip": 1.03896284,
      "balance_loss_mlp": 1.0155673,
      "epoch": 0.9762520290987795,
      "flos": 26575440359040.0,
      "grad_norm": 4.141999539205227,
      "language_loss": 0.63938695,
      "learning_rate": 5.88440673145163e-09,
      "loss": 0.66125488,
      "num_input_tokens_seen": 175324070,
      "step": 8119,
      "time_per_iteration": 2.6549088954925537
    },
    {
      "auxiliary_loss_clip": 0.01168237,
      "auxiliary_loss_mlp": 0.01026682,
      "balance_loss_clip": 1.04734707,
      "balance_loss_mlp": 1.01916015,
      "epoch": 0.9763722719894187,
      "flos": 18005036307840.0,
      "grad_norm": 2.098047732650717,
      "language_loss": 0.82406354,
      "learning_rate": 5.824847127848142e-09,
      "loss": 0.84601277,
      "num_input_tokens_seen": 175342595,
      "step": 8120,
      "time_per_iteration": 2.573728322982788
    },
    {
      "auxiliary_loss_clip": 0.01178553,
      "auxiliary_loss_mlp": 0.01021407,
      "balance_loss_clip": 1.04253697,
      "balance_loss_mlp": 1.01398599,
      "epoch": 0.9764925148800577,
      "flos": 22455660931200.0,
      "grad_norm": 1.9352685135832595,
      "language_loss": 0.78812909,
      "learning_rate": 5.765590039029433e-09,
      "loss": 0.81012875,
      "num_input_tokens_seen": 175361915,
      "step": 8121,
      "time_per_iteration": 3.592874526977539
    },
    {
      "auxiliary_loss_clip": 0.01165939,
      "auxiliary_loss_mlp": 0.01023516,
      "balance_loss_clip": 1.04830682,
      "balance_loss_mlp": 1.01597047,
      "epoch": 0.9766127577706968,
      "flos": 36757084786560.0,
      "grad_norm": 1.5947398422787122,
      "language_loss": 0.7088306,
      "learning_rate": 5.706635473985422e-09,
      "loss": 0.73072517,
      "num_input_tokens_seen": 175385785,
      "step": 8122,
      "time_per_iteration": 2.645730972290039
    },
    {
      "auxiliary_loss_clip": 0.01165145,
      "auxiliary_loss_mlp": 0.01024709,
      "balance_loss_clip": 1.04387546,
      "balance_loss_mlp": 1.01752687,
      "epoch": 0.976733000661336,
      "flos": 22309971367680.0,
      "grad_norm": 2.0381030212259894,
      "language_loss": 0.85240197,
      "learning_rate": 5.6479834416591764e-09,
      "loss": 0.87430048,
      "num_input_tokens_seen": 175405145,
      "step": 8123,
      "time_per_iteration": 2.5679895877838135
    },
    {
      "auxiliary_loss_clip": 0.01166807,
      "auxiliary_loss_mlp": 0.01207842,
      "balance_loss_clip": 1.04632831,
      "balance_loss_mlp": 1.00055552,
      "epoch": 0.976853243551975,
      "flos": 25810938264960.0,
      "grad_norm": 1.7828727469164811,
      "language_loss": 0.68036437,
      "learning_rate": 5.589633950947803e-09,
      "loss": 0.70411086,
      "num_input_tokens_seen": 175422645,
      "step": 8124,
      "time_per_iteration": 3.543928861618042
    },
    {
      "auxiliary_loss_clip": 0.01172524,
      "auxiliary_loss_mlp": 0.01029288,
      "balance_loss_clip": 1.04528117,
      "balance_loss_mlp": 1.0213244,
      "epoch": 0.9769734864426141,
      "flos": 21397445326080.0,
      "grad_norm": 2.1547872531844425,
      "language_loss": 0.70111114,
      "learning_rate": 5.5315870107035535e-09,
      "loss": 0.72312927,
      "num_input_tokens_seen": 175440695,
      "step": 8125,
      "time_per_iteration": 2.626502513885498
    },
    {
      "auxiliary_loss_clip": 0.01166705,
      "auxiliary_loss_mlp": 0.01023087,
      "balance_loss_clip": 1.04623508,
      "balance_loss_mlp": 1.01566052,
      "epoch": 0.9770937293332532,
      "flos": 13990977584640.0,
      "grad_norm": 1.9457468489458138,
      "language_loss": 0.78970599,
      "learning_rate": 5.473842629731607e-09,
      "loss": 0.81160396,
      "num_input_tokens_seen": 175459195,
      "step": 8126,
      "time_per_iteration": 2.5735843181610107
    },
    {
      "auxiliary_loss_clip": 0.01181067,
      "auxiliary_loss_mlp": 0.01207895,
      "balance_loss_clip": 1.04537463,
      "balance_loss_mlp": 1.00074756,
      "epoch": 0.9772139722238923,
      "flos": 17931994001280.0,
      "grad_norm": 2.7462072953681114,
      "language_loss": 0.78354025,
      "learning_rate": 5.416400816792066e-09,
      "loss": 0.80742991,
      "num_input_tokens_seen": 175476710,
      "step": 8127,
      "time_per_iteration": 3.5598790645599365
    },
    {
      "auxiliary_loss_clip": 0.01164861,
      "auxiliary_loss_mlp": 0.01022013,
      "balance_loss_clip": 1.04642296,
      "balance_loss_mlp": 1.01491976,
      "epoch": 0.9773342151145313,
      "flos": 20446171488000.0,
      "grad_norm": 2.8828419872571676,
      "language_loss": 0.78284186,
      "learning_rate": 5.359261580598407e-09,
      "loss": 0.80471063,
      "num_input_tokens_seen": 175492550,
      "step": 8128,
      "time_per_iteration": 2.510317087173462
    },
    {
      "auxiliary_loss_clip": 0.01173,
      "auxiliary_loss_mlp": 0.01021844,
      "balance_loss_clip": 1.04715562,
      "balance_loss_mlp": 1.0139823,
      "epoch": 0.9774544580051704,
      "flos": 11837306949120.0,
      "grad_norm": 3.8987380098759363,
      "language_loss": 0.78042698,
      "learning_rate": 5.302424929819027e-09,
      "loss": 0.80237544,
      "num_input_tokens_seen": 175506560,
      "step": 8129,
      "time_per_iteration": 2.512707233428955
    },
    {
      "auxiliary_loss_clip": 0.01172413,
      "auxiliary_loss_mlp": 0.01026142,
      "balance_loss_clip": 1.04324567,
      "balance_loss_mlp": 1.01878953,
      "epoch": 0.9775747008958096,
      "flos": 13479932833920.0,
      "grad_norm": 2.439758041380862,
      "language_loss": 0.72845352,
      "learning_rate": 5.24589087307592e-09,
      "loss": 0.75043905,
      "num_input_tokens_seen": 175524180,
      "step": 8130,
      "time_per_iteration": 2.460805654525757
    },
    {
      "auxiliary_loss_clip": 0.01167945,
      "auxiliary_loss_mlp": 0.01024951,
      "balance_loss_clip": 1.04752326,
      "balance_loss_mlp": 1.0177654,
      "epoch": 0.9776949437864486,
      "flos": 59532314042880.0,
      "grad_norm": 1.8650780159202947,
      "language_loss": 0.65113521,
      "learning_rate": 5.189659418944891e-09,
      "loss": 0.67306417,
      "num_input_tokens_seen": 175554355,
      "step": 8131,
      "time_per_iteration": 2.8858039379119873
    },
    {
      "auxiliary_loss_clip": 0.01166216,
      "auxiliary_loss_mlp": 0.01026095,
      "balance_loss_clip": 1.0491159,
      "balance_loss_mlp": 1.01897562,
      "epoch": 0.9778151866770877,
      "flos": 21178605715200.0,
      "grad_norm": 2.743222319292985,
      "language_loss": 0.78559422,
      "learning_rate": 5.133730575956674e-09,
      "loss": 0.80751735,
      "num_input_tokens_seen": 175574025,
      "step": 8132,
      "time_per_iteration": 2.593172550201416
    },
    {
      "auxiliary_loss_clip": 0.01174444,
      "auxiliary_loss_mlp": 0.01019444,
      "balance_loss_clip": 1.04476976,
      "balance_loss_mlp": 1.01202941,
      "epoch": 0.9779354295677268,
      "flos": 20886795624960.0,
      "grad_norm": 2.1653392377907887,
      "language_loss": 0.71972883,
      "learning_rate": 5.0781043525953696e-09,
      "loss": 0.74166775,
      "num_input_tokens_seen": 175592090,
      "step": 8133,
      "time_per_iteration": 2.6175537109375
    },
    {
      "auxiliary_loss_clip": 0.01164292,
      "auxiliary_loss_mlp": 0.01024101,
      "balance_loss_clip": 1.04546928,
      "balance_loss_mlp": 1.01673949,
      "epoch": 0.9780556724583659,
      "flos": 23440618748160.0,
      "grad_norm": 2.2401541477394855,
      "language_loss": 0.74027282,
      "learning_rate": 5.0227807572995605e-09,
      "loss": 0.76215672,
      "num_input_tokens_seen": 175614065,
      "step": 8134,
      "time_per_iteration": 2.649050712585449
    },
    {
      "auxiliary_loss_clip": 0.01172745,
      "auxiliary_loss_mlp": 0.01023347,
      "balance_loss_clip": 1.04197025,
      "balance_loss_mlp": 1.01611066,
      "epoch": 0.9781759153490049,
      "flos": 20923244951040.0,
      "grad_norm": 2.7051083232911224,
      "language_loss": 0.67641246,
      "learning_rate": 4.967759798461646e-09,
      "loss": 0.69837344,
      "num_input_tokens_seen": 175632410,
      "step": 8135,
      "time_per_iteration": 2.805954694747925
    },
    {
      "auxiliary_loss_clip": 0.01164359,
      "auxiliary_loss_mlp": 0.01023982,
      "balance_loss_clip": 1.04777622,
      "balance_loss_mlp": 1.0168469,
      "epoch": 0.9782961582396441,
      "flos": 28293191539200.0,
      "grad_norm": 2.5666538202498663,
      "language_loss": 0.75086868,
      "learning_rate": 4.913041484428282e-09,
      "loss": 0.77275199,
      "num_input_tokens_seen": 175652885,
      "step": 8136,
      "time_per_iteration": 2.6472177505493164
    },
    {
      "auxiliary_loss_clip": 0.01170074,
      "auxiliary_loss_mlp": 0.01021123,
      "balance_loss_clip": 1.04677999,
      "balance_loss_mlp": 1.01414597,
      "epoch": 0.9784164011302832,
      "flos": 25552955808000.0,
      "grad_norm": 1.7375156217401497,
      "language_loss": 0.74203891,
      "learning_rate": 4.858625823500384e-09,
      "loss": 0.76395082,
      "num_input_tokens_seen": 175670585,
      "step": 8137,
      "time_per_iteration": 2.600827932357788
    },
    {
      "auxiliary_loss_clip": 0.0117136,
      "auxiliary_loss_mlp": 0.01025772,
      "balance_loss_clip": 1.04551446,
      "balance_loss_mlp": 1.01786828,
      "epoch": 0.9785366440209222,
      "flos": 29965945956480.0,
      "grad_norm": 2.102403189787409,
      "language_loss": 0.73621404,
      "learning_rate": 4.80451282393246e-09,
      "loss": 0.75818533,
      "num_input_tokens_seen": 175690570,
      "step": 8138,
      "time_per_iteration": 2.6838393211364746
    },
    {
      "auxiliary_loss_clip": 0.01170048,
      "auxiliary_loss_mlp": 0.01019552,
      "balance_loss_clip": 1.04338264,
      "balance_loss_mlp": 1.01251888,
      "epoch": 0.9786568869115614,
      "flos": 32343591847680.0,
      "grad_norm": 2.250545788233995,
      "language_loss": 0.67650557,
      "learning_rate": 4.750702493933722e-09,
      "loss": 0.69840157,
      "num_input_tokens_seen": 175710455,
      "step": 8139,
      "time_per_iteration": 3.6063008308410645
    },
    {
      "auxiliary_loss_clip": 0.01168852,
      "auxiliary_loss_mlp": 0.01207334,
      "balance_loss_clip": 1.04411209,
      "balance_loss_mlp": 1.00062251,
      "epoch": 0.9787771298022004,
      "flos": 23331414424320.0,
      "grad_norm": 1.9425235137635963,
      "language_loss": 0.85311747,
      "learning_rate": 4.697194841666974e-09,
      "loss": 0.87687933,
      "num_input_tokens_seen": 175729380,
      "step": 8140,
      "time_per_iteration": 2.7071778774261475
    },
    {
      "auxiliary_loss_clip": 0.01171858,
      "auxiliary_loss_mlp": 0.01027959,
      "balance_loss_clip": 1.04619145,
      "balance_loss_mlp": 1.01947689,
      "epoch": 0.9788973726928395,
      "flos": 21468548298240.0,
      "grad_norm": 2.6551583462059556,
      "language_loss": 0.82048178,
      "learning_rate": 4.6439898752492764e-09,
      "loss": 0.84247994,
      "num_input_tokens_seen": 175749520,
      "step": 8141,
      "time_per_iteration": 2.6502199172973633
    },
    {
      "auxiliary_loss_clip": 0.010707,
      "auxiliary_loss_mlp": 0.01198759,
      "balance_loss_clip": 1.01112127,
      "balance_loss_mlp": 0.99987215,
      "epoch": 0.9790176155834787,
      "flos": 68897459439360.0,
      "grad_norm": 0.7473065481133181,
      "language_loss": 0.6360321,
      "learning_rate": 4.591087602751731e-09,
      "loss": 0.65872669,
      "num_input_tokens_seen": 175811380,
      "step": 8142,
      "time_per_iteration": 3.2054479122161865
    },
    {
      "auxiliary_loss_clip": 0.01168813,
      "auxiliary_loss_mlp": 0.01023207,
      "balance_loss_clip": 1.04579973,
      "balance_loss_mlp": 1.0159409,
      "epoch": 0.9791378584741177,
      "flos": 21430877909760.0,
      "grad_norm": 1.6179333220002075,
      "language_loss": 0.72153145,
      "learning_rate": 4.538488032199916e-09,
      "loss": 0.7434516,
      "num_input_tokens_seen": 175829480,
      "step": 8143,
      "time_per_iteration": 2.60068941116333
    },
    {
      "auxiliary_loss_clip": 0.01172085,
      "auxiliary_loss_mlp": 0.01026237,
      "balance_loss_clip": 1.04421294,
      "balance_loss_mlp": 1.01881659,
      "epoch": 0.9792581013647568,
      "flos": 20153032594560.0,
      "grad_norm": 2.111552245486251,
      "language_loss": 0.68737793,
      "learning_rate": 4.486191171572784e-09,
      "loss": 0.7093612,
      "num_input_tokens_seen": 175846750,
      "step": 8144,
      "time_per_iteration": 2.65147066116333
    },
    {
      "auxiliary_loss_clip": 0.01171939,
      "auxiliary_loss_mlp": 0.01021335,
      "balance_loss_clip": 1.04833031,
      "balance_loss_mlp": 1.01421189,
      "epoch": 0.9793783442553959,
      "flos": 23728191033600.0,
      "grad_norm": 3.1952988413895933,
      "language_loss": 0.77411968,
      "learning_rate": 4.434197028803766e-09,
      "loss": 0.79605234,
      "num_input_tokens_seen": 175865975,
      "step": 8145,
      "time_per_iteration": 2.777418613433838
    },
    {
      "auxiliary_loss_clip": 0.01181328,
      "auxiliary_loss_mlp": 0.01030027,
      "balance_loss_clip": 1.04328144,
      "balance_loss_mlp": 1.02126479,
      "epoch": 0.979498587146035,
      "flos": 23038742407680.0,
      "grad_norm": 2.0811126129017468,
      "language_loss": 0.82224029,
      "learning_rate": 4.3825056117805514e-09,
      "loss": 0.8443538,
      "num_input_tokens_seen": 175881860,
      "step": 8146,
      "time_per_iteration": 2.70021653175354
    },
    {
      "auxiliary_loss_clip": 0.01166858,
      "auxiliary_loss_mlp": 0.0101931,
      "balance_loss_clip": 1.0462954,
      "balance_loss_mlp": 1.01193142,
      "epoch": 0.979618830036674,
      "flos": 14318841951360.0,
      "grad_norm": 4.067614233623715,
      "language_loss": 0.79302424,
      "learning_rate": 4.331116928344425e-09,
      "loss": 0.81488597,
      "num_input_tokens_seen": 175898175,
      "step": 8147,
      "time_per_iteration": 2.599310874938965
    },
    {
      "auxiliary_loss_clip": 0.01178489,
      "auxiliary_loss_mlp": 0.01207699,
      "balance_loss_clip": 1.04488182,
      "balance_loss_mlp": 1.00065124,
      "epoch": 0.9797390729273132,
      "flos": 16727514215040.0,
      "grad_norm": 2.0260316595418946,
      "language_loss": 0.62638938,
      "learning_rate": 4.28003098629115e-09,
      "loss": 0.65025127,
      "num_input_tokens_seen": 175914310,
      "step": 8148,
      "time_per_iteration": 3.640761613845825
    },
    {
      "auxiliary_loss_clip": 0.01171063,
      "auxiliary_loss_mlp": 0.0102172,
      "balance_loss_clip": 1.03746176,
      "balance_loss_mlp": 1.01414382,
      "epoch": 0.9798593158179523,
      "flos": 24532661986560.0,
      "grad_norm": 1.8493625897232344,
      "language_loss": 0.78562778,
      "learning_rate": 4.229247793370305e-09,
      "loss": 0.80755568,
      "num_input_tokens_seen": 175933435,
      "step": 8149,
      "time_per_iteration": 2.709808349609375
    },
    {
      "auxiliary_loss_clip": 0.01168828,
      "auxiliary_loss_mlp": 0.01023945,
      "balance_loss_clip": 1.04806447,
      "balance_loss_mlp": 1.01630712,
      "epoch": 0.9799795587085913,
      "flos": 27308808339840.0,
      "grad_norm": 2.16171964597173,
      "language_loss": 0.70745772,
      "learning_rate": 4.178767357285951e-09,
      "loss": 0.7293855,
      "num_input_tokens_seen": 175955065,
      "step": 8150,
      "time_per_iteration": 3.573368549346924
    },
    {
      "auxiliary_loss_clip": 0.01170247,
      "auxiliary_loss_mlp": 0.01208051,
      "balance_loss_clip": 1.04664695,
      "balance_loss_mlp": 1.00088334,
      "epoch": 0.9800998015992305,
      "flos": 26286575184000.0,
      "grad_norm": 2.2643678412938666,
      "language_loss": 0.71753395,
      "learning_rate": 4.128589685695516e-09,
      "loss": 0.74131691,
      "num_input_tokens_seen": 175975490,
      "step": 8151,
      "time_per_iteration": 2.714303493499756
    },
    {
      "auxiliary_loss_clip": 0.01166904,
      "auxiliary_loss_mlp": 0.01024362,
      "balance_loss_clip": 1.04816985,
      "balance_loss_mlp": 1.01746285,
      "epoch": 0.9802200444898695,
      "flos": 16723635546240.0,
      "grad_norm": 2.2022799832628777,
      "language_loss": 0.84374082,
      "learning_rate": 4.078714786211135e-09,
      "loss": 0.86565351,
      "num_input_tokens_seen": 175991340,
      "step": 8152,
      "time_per_iteration": 2.7265753746032715
    },
    {
      "auxiliary_loss_clip": 0.01166252,
      "auxiliary_loss_mlp": 0.01023879,
      "balance_loss_clip": 1.04580569,
      "balance_loss_mlp": 1.0167594,
      "epoch": 0.9803402873805086,
      "flos": 24900459298560.0,
      "grad_norm": 2.0212223433091676,
      "language_loss": 0.76956546,
      "learning_rate": 4.029142666398977e-09,
      "loss": 0.79146671,
      "num_input_tokens_seen": 176011505,
      "step": 8153,
      "time_per_iteration": 3.7678322792053223
    },
    {
      "auxiliary_loss_clip": 0.01164212,
      "auxiliary_loss_mlp": 0.01025247,
      "balance_loss_clip": 1.04741907,
      "balance_loss_mlp": 1.0183152,
      "epoch": 0.9804605302711478,
      "flos": 22564937082240.0,
      "grad_norm": 2.267190977109989,
      "language_loss": 0.80140042,
      "learning_rate": 3.979873333778805e-09,
      "loss": 0.823295,
      "num_input_tokens_seen": 176029680,
      "step": 8154,
      "time_per_iteration": 2.6683425903320312
    },
    {
      "auxiliary_loss_clip": 0.01175702,
      "auxiliary_loss_mlp": 0.01024346,
      "balance_loss_clip": 1.0459944,
      "balance_loss_mlp": 1.01689577,
      "epoch": 0.9805807731617868,
      "flos": 38905368382080.0,
      "grad_norm": 2.2021845129736795,
      "language_loss": 0.74480367,
      "learning_rate": 3.930906795824862e-09,
      "loss": 0.76680422,
      "num_input_tokens_seen": 176050355,
      "step": 8155,
      "time_per_iteration": 2.7622578144073486
    },
    {
      "auxiliary_loss_clip": 0.01166301,
      "auxiliary_loss_mlp": 0.01019463,
      "balance_loss_clip": 1.04510474,
      "balance_loss_mlp": 1.01222956,
      "epoch": 0.9807010160524259,
      "flos": 17821999578240.0,
      "grad_norm": 2.5633515677355745,
      "language_loss": 0.77317953,
      "learning_rate": 3.882243059965207e-09,
      "loss": 0.79503715,
      "num_input_tokens_seen": 176068070,
      "step": 8156,
      "time_per_iteration": 2.622654676437378
    },
    {
      "auxiliary_loss_clip": 0.01161917,
      "auxiliary_loss_mlp": 0.01023168,
      "balance_loss_clip": 1.04333448,
      "balance_loss_mlp": 1.01574731,
      "epoch": 0.980821258943065,
      "flos": 13552975140480.0,
      "grad_norm": 2.514211681746124,
      "language_loss": 0.65802038,
      "learning_rate": 3.833882133582156e-09,
      "loss": 0.6798712,
      "num_input_tokens_seen": 176083730,
      "step": 8157,
      "time_per_iteration": 2.5726654529571533
    },
    {
      "auxiliary_loss_clip": 0.01170063,
      "auxiliary_loss_mlp": 0.01019603,
      "balance_loss_clip": 1.0448246,
      "balance_loss_mlp": 1.01269758,
      "epoch": 0.9809415018337041,
      "flos": 21689794120320.0,
      "grad_norm": 1.6587786244724387,
      "language_loss": 0.78408194,
      "learning_rate": 3.785824024012285e-09,
      "loss": 0.8059786,
      "num_input_tokens_seen": 176102730,
      "step": 8158,
      "time_per_iteration": 2.694612741470337
    },
    {
      "auxiliary_loss_clip": 0.01163433,
      "auxiliary_loss_mlp": 0.01024171,
      "balance_loss_clip": 1.04628885,
      "balance_loss_mlp": 1.01682222,
      "epoch": 0.9810617447243432,
      "flos": 23294857357440.0,
      "grad_norm": 1.5733045702596946,
      "language_loss": 0.78580618,
      "learning_rate": 3.738068738545541e-09,
      "loss": 0.80768228,
      "num_input_tokens_seen": 176121815,
      "step": 8159,
      "time_per_iteration": 2.6882822513580322
    },
    {
      "auxiliary_loss_clip": 0.01174312,
      "auxiliary_loss_mlp": 0.01026182,
      "balance_loss_clip": 1.04722309,
      "balance_loss_mlp": 1.01874316,
      "epoch": 0.9811819876149822,
      "flos": 18332038748160.0,
      "grad_norm": 2.9648771748217135,
      "language_loss": 0.7878114,
      "learning_rate": 3.6906162844265733e-09,
      "loss": 0.80981636,
      "num_input_tokens_seen": 176138900,
      "step": 8160,
      "time_per_iteration": 2.6679203510284424
    },
    {
      "auxiliary_loss_clip": 0.01166778,
      "auxiliary_loss_mlp": 0.01027593,
      "balance_loss_clip": 1.0426482,
      "balance_loss_mlp": 1.01975524,
      "epoch": 0.9813022305056214,
      "flos": 22601961025920.0,
      "grad_norm": 2.3888400699545143,
      "language_loss": 0.70780885,
      "learning_rate": 3.643466668853845e-09,
      "loss": 0.72975254,
      "num_input_tokens_seen": 176156925,
      "step": 8161,
      "time_per_iteration": 2.718736410140991
    },
    {
      "auxiliary_loss_clip": 0.0117212,
      "auxiliary_loss_mlp": 0.01019556,
      "balance_loss_clip": 1.0434159,
      "balance_loss_mlp": 1.01198649,
      "epoch": 0.9814224733962604,
      "flos": 25413335642880.0,
      "grad_norm": 6.09936636041352,
      "language_loss": 0.75059688,
      "learning_rate": 3.59661989898008e-09,
      "loss": 0.77251363,
      "num_input_tokens_seen": 176177980,
      "step": 8162,
      "time_per_iteration": 2.70973801612854
    },
    {
      "auxiliary_loss_clip": 0.01164848,
      "auxiliary_loss_mlp": 0.0102224,
      "balance_loss_clip": 1.04163361,
      "balance_loss_mlp": 1.01495671,
      "epoch": 0.9815427162868995,
      "flos": 25007185584000.0,
      "grad_norm": 1.8053996575177702,
      "language_loss": 0.76857829,
      "learning_rate": 3.5500759819115934e-09,
      "loss": 0.7904492,
      "num_input_tokens_seen": 176198345,
      "step": 8163,
      "time_per_iteration": 2.6840710639953613
    },
    {
      "auxiliary_loss_clip": 0.01170319,
      "auxiliary_loss_mlp": 0.01020568,
      "balance_loss_clip": 1.05004787,
      "balance_loss_mlp": 1.01309967,
      "epoch": 0.9816629591775387,
      "flos": 20662604887680.0,
      "grad_norm": 1.8721539545284038,
      "language_loss": 0.81150746,
      "learning_rate": 3.5038349247094034e-09,
      "loss": 0.83341634,
      "num_input_tokens_seen": 176215605,
      "step": 8164,
      "time_per_iteration": 2.6074302196502686
    },
    {
      "auxiliary_loss_clip": 0.01170623,
      "auxiliary_loss_mlp": 0.01024432,
      "balance_loss_clip": 1.04197967,
      "balance_loss_mlp": 1.0169574,
      "epoch": 0.9817832020681777,
      "flos": 17712220636800.0,
      "grad_norm": 3.053808272429565,
      "language_loss": 0.77493167,
      "learning_rate": 3.4578967343878994e-09,
      "loss": 0.79688221,
      "num_input_tokens_seen": 176231810,
      "step": 8165,
      "time_per_iteration": 2.6411569118499756
    },
    {
      "auxiliary_loss_clip": 0.01171096,
      "auxiliary_loss_mlp": 0.01022276,
      "balance_loss_clip": 1.04492486,
      "balance_loss_mlp": 1.01503432,
      "epoch": 0.9819034449588168,
      "flos": 22530032040960.0,
      "grad_norm": 2.3263375374940933,
      "language_loss": 0.81250966,
      "learning_rate": 3.4122614179161733e-09,
      "loss": 0.83444339,
      "num_input_tokens_seen": 176251770,
      "step": 8166,
      "time_per_iteration": 3.6679136753082275
    },
    {
      "auxiliary_loss_clip": 0.01161441,
      "auxiliary_loss_mlp": 0.010231,
      "balance_loss_clip": 1.03978968,
      "balance_loss_mlp": 1.01615,
      "epoch": 0.9820236878494559,
      "flos": 20011221699840.0,
      "grad_norm": 1.9050846446171033,
      "language_loss": 0.78589976,
      "learning_rate": 3.36692898221691e-09,
      "loss": 0.8077451,
      "num_input_tokens_seen": 176270135,
      "step": 8167,
      "time_per_iteration": 2.697545289993286
    },
    {
      "auxiliary_loss_clip": 0.01169379,
      "auxiliary_loss_mlp": 0.01024401,
      "balance_loss_clip": 1.04507232,
      "balance_loss_mlp": 1.01739764,
      "epoch": 0.982143930740095,
      "flos": 18807316531200.0,
      "grad_norm": 1.7311144935699323,
      "language_loss": 0.7363764,
      "learning_rate": 3.3218994341668305e-09,
      "loss": 0.75831425,
      "num_input_tokens_seen": 176289065,
      "step": 8168,
      "time_per_iteration": 2.731841564178467
    },
    {
      "auxiliary_loss_clip": 0.01164375,
      "auxiliary_loss_mlp": 0.01022251,
      "balance_loss_clip": 1.04803443,
      "balance_loss_mlp": 1.01539087,
      "epoch": 0.982264173630734,
      "flos": 26578026138240.0,
      "grad_norm": 2.2521098217453566,
      "language_loss": 0.75431389,
      "learning_rate": 3.2771727805971373e-09,
      "loss": 0.77618015,
      "num_input_tokens_seen": 176310450,
      "step": 8169,
      "time_per_iteration": 2.6839139461517334
    },
    {
      "auxiliary_loss_clip": 0.01170895,
      "auxiliary_loss_mlp": 0.01021962,
      "balance_loss_clip": 1.03825736,
      "balance_loss_mlp": 1.01466632,
      "epoch": 0.9823844165213732,
      "flos": 22014462176640.0,
      "grad_norm": 2.969461271370885,
      "language_loss": 0.77527916,
      "learning_rate": 3.232749028292847e-09,
      "loss": 0.79720777,
      "num_input_tokens_seen": 176327415,
      "step": 8170,
      "time_per_iteration": 2.7190985679626465
    },
    {
      "auxiliary_loss_clip": 0.01167546,
      "auxiliary_loss_mlp": 0.0102681,
      "balance_loss_clip": 1.04622054,
      "balance_loss_mlp": 1.01924896,
      "epoch": 0.9825046594120123,
      "flos": 21908166854400.0,
      "grad_norm": 2.5539021039303655,
      "language_loss": 0.88673514,
      "learning_rate": 3.188628183992792e-09,
      "loss": 0.90867865,
      "num_input_tokens_seen": 176347680,
      "step": 8171,
      "time_per_iteration": 2.692028522491455
    },
    {
      "auxiliary_loss_clip": 0.01070527,
      "auxiliary_loss_mlp": 0.0100086,
      "balance_loss_clip": 1.01031256,
      "balance_loss_mlp": 1.00009143,
      "epoch": 0.9826249023026513,
      "flos": 59494610718720.0,
      "grad_norm": 0.7383768816899511,
      "language_loss": 0.62502307,
      "learning_rate": 3.1448102543902844e-09,
      "loss": 0.64573699,
      "num_input_tokens_seen": 176411595,
      "step": 8172,
      "time_per_iteration": 3.136498212814331
    },
    {
      "auxiliary_loss_clip": 0.01164107,
      "auxiliary_loss_mlp": 0.01026535,
      "balance_loss_clip": 1.04393506,
      "balance_loss_mlp": 1.01903975,
      "epoch": 0.9827451451932905,
      "flos": 16071031296000.0,
      "grad_norm": 1.9720384485882667,
      "language_loss": 0.67885137,
      "learning_rate": 3.1012952461324515e-09,
      "loss": 0.70075774,
      "num_input_tokens_seen": 176430570,
      "step": 8173,
      "time_per_iteration": 2.7406070232391357
    },
    {
      "auxiliary_loss_clip": 0.01166943,
      "auxiliary_loss_mlp": 0.01027699,
      "balance_loss_clip": 1.04788661,
      "balance_loss_mlp": 1.02014756,
      "epoch": 0.9828653880839295,
      "flos": 20262775622400.0,
      "grad_norm": 2.1730962092659274,
      "language_loss": 0.73829859,
      "learning_rate": 3.0580831658204575e-09,
      "loss": 0.76024497,
      "num_input_tokens_seen": 176448150,
      "step": 8174,
      "time_per_iteration": 3.6899476051330566
    },
    {
      "auxiliary_loss_clip": 0.01167674,
      "auxiliary_loss_mlp": 0.01020912,
      "balance_loss_clip": 1.04695153,
      "balance_loss_mlp": 1.01418602,
      "epoch": 0.9829856309745686,
      "flos": 21616141282560.0,
      "grad_norm": 1.57865089980539,
      "language_loss": 0.77931321,
      "learning_rate": 3.015174020009281e-09,
      "loss": 0.80119908,
      "num_input_tokens_seen": 176467475,
      "step": 8175,
      "time_per_iteration": 2.7079970836639404
    },
    {
      "auxiliary_loss_clip": 0.01177778,
      "auxiliary_loss_mlp": 0.01023171,
      "balance_loss_clip": 1.04300165,
      "balance_loss_mlp": 1.01636171,
      "epoch": 0.9831058738652078,
      "flos": 23764209396480.0,
      "grad_norm": 1.8596208183682639,
      "language_loss": 0.74871707,
      "learning_rate": 2.9725678152086043e-09,
      "loss": 0.77072656,
      "num_input_tokens_seen": 176486045,
      "step": 8176,
      "time_per_iteration": 2.8009495735168457
    },
    {
      "auxiliary_loss_clip": 0.01158141,
      "auxiliary_loss_mlp": 0.01023791,
      "balance_loss_clip": 1.04176831,
      "balance_loss_mlp": 1.01642084,
      "epoch": 0.9832261167558468,
      "flos": 11320911072000.0,
      "grad_norm": 2.460341839917179,
      "language_loss": 0.83043563,
      "learning_rate": 2.930264557881257e-09,
      "loss": 0.85225493,
      "num_input_tokens_seen": 176501230,
      "step": 8177,
      "time_per_iteration": 3.582143545150757
    },
    {
      "auxiliary_loss_clip": 0.01062518,
      "auxiliary_loss_mlp": 0.01002345,
      "balance_loss_clip": 1.01004291,
      "balance_loss_mlp": 1.00160027,
      "epoch": 0.9833463596464859,
      "flos": 60000304343040.0,
      "grad_norm": 0.834684558186543,
      "language_loss": 0.58153224,
      "learning_rate": 2.8882642544452163e-09,
      "loss": 0.60218084,
      "num_input_tokens_seen": 176565955,
      "step": 8178,
      "time_per_iteration": 3.284637689590454
    },
    {
      "auxiliary_loss_clip": 0.01161063,
      "auxiliary_loss_mlp": 0.01024634,
      "balance_loss_clip": 1.04301751,
      "balance_loss_mlp": 1.01707017,
      "epoch": 0.983466602537125,
      "flos": 13626699805440.0,
      "grad_norm": 2.126762694899309,
      "language_loss": 0.74335444,
      "learning_rate": 2.8465669112716083e-09,
      "loss": 0.76521134,
      "num_input_tokens_seen": 176583480,
      "step": 8179,
      "time_per_iteration": 2.655444622039795
    },
    {
      "auxiliary_loss_clip": 0.01169167,
      "auxiliary_loss_mlp": 0.01207709,
      "balance_loss_clip": 1.04369926,
      "balance_loss_mlp": 1.00085187,
      "epoch": 0.9835868454277641,
      "flos": 22926844563840.0,
      "grad_norm": 2.107512603646858,
      "language_loss": 0.76598918,
      "learning_rate": 2.8051725346858177e-09,
      "loss": 0.78975791,
      "num_input_tokens_seen": 176603740,
      "step": 8180,
      "time_per_iteration": 3.6627514362335205
    },
    {
      "auxiliary_loss_clip": 0.01167375,
      "auxiliary_loss_mlp": 0.01026887,
      "balance_loss_clip": 1.04579842,
      "balance_loss_mlp": 1.01953137,
      "epoch": 0.9837070883184031,
      "flos": 27673409341440.0,
      "grad_norm": 2.501676690058931,
      "language_loss": 0.7111733,
      "learning_rate": 2.7640811309674883e-09,
      "loss": 0.73311591,
      "num_input_tokens_seen": 176623240,
      "step": 8181,
      "time_per_iteration": 2.750009059906006
    },
    {
      "auxiliary_loss_clip": 0.01163034,
      "auxiliary_loss_mlp": 0.01018876,
      "balance_loss_clip": 1.04081559,
      "balance_loss_mlp": 1.01159823,
      "epoch": 0.9838273312090423,
      "flos": 29241951425280.0,
      "grad_norm": 1.8492657382550555,
      "language_loss": 0.80993593,
      "learning_rate": 2.7232927063498557e-09,
      "loss": 0.83175504,
      "num_input_tokens_seen": 176643615,
      "step": 8182,
      "time_per_iteration": 2.729024648666382
    },
    {
      "auxiliary_loss_clip": 0.01173041,
      "auxiliary_loss_mlp": 0.01022337,
      "balance_loss_clip": 1.04763055,
      "balance_loss_mlp": 1.01526761,
      "epoch": 0.9839475740996814,
      "flos": 40110207304320.0,
      "grad_norm": 2.0694732311263904,
      "language_loss": 0.69206685,
      "learning_rate": 2.682807267020859e-09,
      "loss": 0.71402061,
      "num_input_tokens_seen": 176666375,
      "step": 8183,
      "time_per_iteration": 2.8408257961273193
    },
    {
      "auxiliary_loss_clip": 0.01169747,
      "auxiliary_loss_mlp": 0.01021504,
      "balance_loss_clip": 1.04639101,
      "balance_loss_mlp": 1.01377964,
      "epoch": 0.9840678169903204,
      "flos": 24169389788160.0,
      "grad_norm": 5.143104936086935,
      "language_loss": 0.62477273,
      "learning_rate": 2.642624819121808e-09,
      "loss": 0.64668524,
      "num_input_tokens_seen": 176686525,
      "step": 8184,
      "time_per_iteration": 2.7122530937194824
    },
    {
      "auxiliary_loss_clip": 0.01169572,
      "auxiliary_loss_mlp": 0.0102635,
      "balance_loss_clip": 1.04588103,
      "balance_loss_mlp": 1.0191735,
      "epoch": 0.9841880598809596,
      "flos": 14684484447360.0,
      "grad_norm": 2.0085471519775275,
      "language_loss": 0.61899859,
      "learning_rate": 2.6027453687487154e-09,
      "loss": 0.64095777,
      "num_input_tokens_seen": 176703615,
      "step": 8185,
      "time_per_iteration": 2.590712070465088
    },
    {
      "auxiliary_loss_clip": 0.01171918,
      "auxiliary_loss_mlp": 0.01021381,
      "balance_loss_clip": 1.04471183,
      "balance_loss_mlp": 1.01412153,
      "epoch": 0.9843083027715986,
      "flos": 22344768668160.0,
      "grad_norm": 2.180351339476803,
      "language_loss": 0.53606683,
      "learning_rate": 2.5631689219509643e-09,
      "loss": 0.55799985,
      "num_input_tokens_seen": 176722295,
      "step": 8186,
      "time_per_iteration": 2.6824557781219482
    },
    {
      "auxiliary_loss_clip": 0.01172836,
      "auxiliary_loss_mlp": 0.01020852,
      "balance_loss_clip": 1.04555154,
      "balance_loss_mlp": 1.01393771,
      "epoch": 0.9844285456622377,
      "flos": 21800111765760.0,
      "grad_norm": 1.8029283244943142,
      "language_loss": 0.83339071,
      "learning_rate": 2.523895484732197e-09,
      "loss": 0.85532761,
      "num_input_tokens_seen": 176741750,
      "step": 8187,
      "time_per_iteration": 2.685810089111328
    },
    {
      "auxiliary_loss_clip": 0.01175339,
      "auxiliary_loss_mlp": 0.01023134,
      "balance_loss_clip": 1.04631996,
      "balance_loss_mlp": 1.01508427,
      "epoch": 0.9845487885528769,
      "flos": 18035380321920.0,
      "grad_norm": 2.3478706843774795,
      "language_loss": 0.75012493,
      "learning_rate": 2.4849250630505357e-09,
      "loss": 0.77210963,
      "num_input_tokens_seen": 176759995,
      "step": 8188,
      "time_per_iteration": 2.626492977142334
    },
    {
      "auxiliary_loss_clip": 0.01170344,
      "auxiliary_loss_mlp": 0.01025365,
      "balance_loss_clip": 1.03737843,
      "balance_loss_mlp": 1.01808095,
      "epoch": 0.9846690314435159,
      "flos": 25228610974080.0,
      "grad_norm": 1.7219913156024471,
      "language_loss": 0.73513901,
      "learning_rate": 2.4462576628172528e-09,
      "loss": 0.75709605,
      "num_input_tokens_seen": 176778625,
      "step": 8189,
      "time_per_iteration": 2.8477280139923096
    },
    {
      "auxiliary_loss_clip": 0.01164885,
      "auxiliary_loss_mlp": 0.01028741,
      "balance_loss_clip": 1.0456562,
      "balance_loss_mlp": 1.02074766,
      "epoch": 0.984789274334155,
      "flos": 18552171248640.0,
      "grad_norm": 2.5120507035290003,
      "language_loss": 0.74211526,
      "learning_rate": 2.407893289898766e-09,
      "loss": 0.76405156,
      "num_input_tokens_seen": 176797655,
      "step": 8190,
      "time_per_iteration": 2.6913743019104004
    },
    {
      "auxiliary_loss_clip": 0.01166018,
      "auxiliary_loss_mlp": 0.01019286,
      "balance_loss_clip": 1.03964019,
      "balance_loss_mlp": 1.01184714,
      "epoch": 0.984909517224794,
      "flos": 27345437233920.0,
      "grad_norm": 1.7716526680387106,
      "language_loss": 0.84066987,
      "learning_rate": 2.3698319501144202e-09,
      "loss": 0.86252284,
      "num_input_tokens_seen": 176818640,
      "step": 8191,
      "time_per_iteration": 2.711500406265259
    },
    {
      "auxiliary_loss_clip": 0.01175595,
      "auxiliary_loss_mlp": 0.0102708,
      "balance_loss_clip": 1.04576683,
      "balance_loss_mlp": 1.01916468,
      "epoch": 0.9850297601154332,
      "flos": 18734058743040.0,
      "grad_norm": 1.6838236817025747,
      "language_loss": 0.7352736,
      "learning_rate": 2.3320736492382644e-09,
      "loss": 0.75730038,
      "num_input_tokens_seen": 176837475,
      "step": 8192,
      "time_per_iteration": 2.6466212272644043
    },
    {
      "auxiliary_loss_clip": 0.01164141,
      "auxiliary_loss_mlp": 0.01025248,
      "balance_loss_clip": 1.04730725,
      "balance_loss_mlp": 1.01836026,
      "epoch": 0.9851500030060723,
      "flos": 22308247514880.0,
      "grad_norm": 1.6104388877820732,
      "language_loss": 0.68168223,
      "learning_rate": 2.29461839299816e-09,
      "loss": 0.70357615,
      "num_input_tokens_seen": 176857190,
      "step": 8193,
      "time_per_iteration": 3.4917006492614746
    },
    {
      "auxiliary_loss_clip": 0.01177873,
      "auxiliary_loss_mlp": 0.01020473,
      "balance_loss_clip": 1.04380465,
      "balance_loss_mlp": 1.01338637,
      "epoch": 0.9852702458967113,
      "flos": 26353691746560.0,
      "grad_norm": 1.6946737968237227,
      "language_loss": 0.80065668,
      "learning_rate": 2.257466187076229e-09,
      "loss": 0.82264018,
      "num_input_tokens_seen": 176876395,
      "step": 8194,
      "time_per_iteration": 2.776317596435547
    },
    {
      "auxiliary_loss_clip": 0.01172736,
      "auxiliary_loss_mlp": 0.01207499,
      "balance_loss_clip": 1.0446099,
      "balance_loss_mlp": 1.00065923,
      "epoch": 0.9853904887873505,
      "flos": 20883599314560.0,
      "grad_norm": 1.6761509203386513,
      "language_loss": 0.71410334,
      "learning_rate": 2.2206170371081854e-09,
      "loss": 0.73790574,
      "num_input_tokens_seen": 176894980,
      "step": 8195,
      "time_per_iteration": 2.645747423171997
    },
    {
      "auxiliary_loss_clip": 0.01172442,
      "auxiliary_loss_mlp": 0.01029627,
      "balance_loss_clip": 1.04294562,
      "balance_loss_mlp": 1.02139568,
      "epoch": 0.9855107316779895,
      "flos": 25263444188160.0,
      "grad_norm": 1.7860484238462198,
      "language_loss": 0.8466543,
      "learning_rate": 2.1840709486842247e-09,
      "loss": 0.86867499,
      "num_input_tokens_seen": 176914600,
      "step": 8196,
      "time_per_iteration": 2.7157809734344482
    },
    {
      "auxiliary_loss_clip": 0.01162946,
      "auxiliary_loss_mlp": 0.01025577,
      "balance_loss_clip": 1.04237723,
      "balance_loss_mlp": 1.0179472,
      "epoch": 0.9856309745686286,
      "flos": 19062102677760.0,
      "grad_norm": 1.9737786058992972,
      "language_loss": 0.79260558,
      "learning_rate": 2.1478279273481335e-09,
      "loss": 0.8144908,
      "num_input_tokens_seen": 176933085,
      "step": 8197,
      "time_per_iteration": 2.687006711959839
    },
    {
      "auxiliary_loss_clip": 0.01169593,
      "auxiliary_loss_mlp": 0.0102545,
      "balance_loss_clip": 1.04802942,
      "balance_loss_mlp": 1.01817822,
      "epoch": 0.9857512174592677,
      "flos": 34130758060800.0,
      "grad_norm": 2.4464086455030185,
      "language_loss": 0.80261111,
      "learning_rate": 2.1118879785981815e-09,
      "loss": 0.8245616,
      "num_input_tokens_seen": 176953225,
      "step": 8198,
      "time_per_iteration": 2.7830965518951416
    },
    {
      "auxiliary_loss_clip": 0.0116699,
      "auxiliary_loss_mlp": 0.01023296,
      "balance_loss_clip": 1.0434711,
      "balance_loss_mlp": 1.01639366,
      "epoch": 0.9858714603499068,
      "flos": 25994693266560.0,
      "grad_norm": 1.8690150410041677,
      "language_loss": 0.79302788,
      "learning_rate": 2.0762511078862288e-09,
      "loss": 0.81493074,
      "num_input_tokens_seen": 176973570,
      "step": 8199,
      "time_per_iteration": 2.747054100036621
    },
    {
      "auxiliary_loss_clip": 0.01177054,
      "auxiliary_loss_mlp": 0.0102149,
      "balance_loss_clip": 1.04325259,
      "balance_loss_mlp": 1.01434374,
      "epoch": 0.9859917032405459,
      "flos": 23696230907520.0,
      "grad_norm": 2.542640108874143,
      "language_loss": 0.64756447,
      "learning_rate": 2.0409173206186183e-09,
      "loss": 0.66954994,
      "num_input_tokens_seen": 176992810,
      "step": 8200,
      "time_per_iteration": 2.6690011024475098
    },
    {
      "auxiliary_loss_clip": 0.01171738,
      "auxiliary_loss_mlp": 0.01023645,
      "balance_loss_clip": 1.04495716,
      "balance_loss_mlp": 1.01666856,
      "epoch": 0.986111946131185,
      "flos": 19938287134080.0,
      "grad_norm": 2.113613273770564,
      "language_loss": 0.87208933,
      "learning_rate": 2.0058866221550617e-09,
      "loss": 0.89404315,
      "num_input_tokens_seen": 177011050,
      "step": 8201,
      "time_per_iteration": 4.140935897827148
    },
    {
      "auxiliary_loss_clip": 0.01165014,
      "auxiliary_loss_mlp": 0.0102281,
      "balance_loss_clip": 1.04514265,
      "balance_loss_mlp": 1.0156213,
      "epoch": 0.9862321890218241,
      "flos": 19828831415040.0,
      "grad_norm": 2.272367072372103,
      "language_loss": 0.7500757,
      "learning_rate": 1.971159017809976e-09,
      "loss": 0.77195394,
      "num_input_tokens_seen": 177029340,
      "step": 8202,
      "time_per_iteration": 2.658785343170166
    },
    {
      "auxiliary_loss_clip": 0.01168503,
      "auxiliary_loss_mlp": 0.01024487,
      "balance_loss_clip": 1.04676867,
      "balance_loss_mlp": 1.01652932,
      "epoch": 0.9863524319124631,
      "flos": 21652051904640.0,
      "grad_norm": 2.5789931612326193,
      "language_loss": 0.77484661,
      "learning_rate": 1.93673451285159e-09,
      "loss": 0.79677653,
      "num_input_tokens_seen": 177048390,
      "step": 8203,
      "time_per_iteration": 3.630979061126709
    },
    {
      "auxiliary_loss_clip": 0.01078512,
      "auxiliary_loss_mlp": 0.01002976,
      "balance_loss_clip": 1.01000834,
      "balance_loss_mlp": 1.00218296,
      "epoch": 0.9864726748031023,
      "flos": 52769977920000.0,
      "grad_norm": 0.7307171761283463,
      "language_loss": 0.56459153,
      "learning_rate": 1.9026131125019495e-09,
      "loss": 0.58540642,
      "num_input_tokens_seen": 177105760,
      "step": 8204,
      "time_per_iteration": 3.23531436920166
    },
    {
      "auxiliary_loss_clip": 0.01162498,
      "auxiliary_loss_mlp": 0.01020401,
      "balance_loss_clip": 1.0454421,
      "balance_loss_mlp": 1.01319134,
      "epoch": 0.9865929176937414,
      "flos": 23364631526400.0,
      "grad_norm": 1.8410933555124425,
      "language_loss": 0.86850077,
      "learning_rate": 1.8687948219371363e-09,
      "loss": 0.89032972,
      "num_input_tokens_seen": 177124985,
      "step": 8205,
      "time_per_iteration": 2.8037285804748535
    },
    {
      "auxiliary_loss_clip": 0.01172485,
      "auxiliary_loss_mlp": 0.0102512,
      "balance_loss_clip": 1.04790044,
      "balance_loss_mlp": 1.01731133,
      "epoch": 0.9867131605843804,
      "flos": 21616679986560.0,
      "grad_norm": 2.091648000746111,
      "language_loss": 0.88906384,
      "learning_rate": 1.835279646287491e-09,
      "loss": 0.91103995,
      "num_input_tokens_seen": 177142995,
      "step": 8206,
      "time_per_iteration": 3.5029566287994385
    },
    {
      "auxiliary_loss_clip": 0.01176507,
      "auxiliary_loss_mlp": 0.01029831,
      "balance_loss_clip": 1.04751348,
      "balance_loss_mlp": 1.02226138,
      "epoch": 0.9868334034750196,
      "flos": 22271403139200.0,
      "grad_norm": 1.8224452734785284,
      "language_loss": 0.76550531,
      "learning_rate": 1.8020675906371685e-09,
      "loss": 0.78756869,
      "num_input_tokens_seen": 177162390,
      "step": 8207,
      "time_per_iteration": 2.7139620780944824
    },
    {
      "auxiliary_loss_clip": 0.01173009,
      "auxiliary_loss_mlp": 0.01026652,
      "balance_loss_clip": 1.03983128,
      "balance_loss_mlp": 1.01945162,
      "epoch": 0.9869536463656586,
      "flos": 25809573548160.0,
      "grad_norm": 2.2373596082280827,
      "language_loss": 0.75627065,
      "learning_rate": 1.7691586600243612e-09,
      "loss": 0.7782672,
      "num_input_tokens_seen": 177181290,
      "step": 8208,
      "time_per_iteration": 2.798431634902954
    },
    {
      "auxiliary_loss_clip": 0.01167899,
      "auxiliary_loss_mlp": 0.0102314,
      "balance_loss_clip": 1.0460155,
      "balance_loss_mlp": 1.01560342,
      "epoch": 0.9870738892562977,
      "flos": 16398500613120.0,
      "grad_norm": 4.358595538999349,
      "language_loss": 0.87542701,
      "learning_rate": 1.7365528594415202e-09,
      "loss": 0.89733744,
      "num_input_tokens_seen": 177195360,
      "step": 8209,
      "time_per_iteration": 2.6652421951293945
    },
    {
      "auxiliary_loss_clip": 0.01173354,
      "auxiliary_loss_mlp": 0.01207844,
      "balance_loss_clip": 1.04564571,
      "balance_loss_mlp": 1.00066996,
      "epoch": 0.9871941321469369,
      "flos": 35481358373760.0,
      "grad_norm": 1.8416654156133097,
      "language_loss": 0.67923832,
      "learning_rate": 1.7042501938346888e-09,
      "loss": 0.70305026,
      "num_input_tokens_seen": 177218090,
      "step": 8210,
      "time_per_iteration": 2.769592761993408
    },
    {
      "auxiliary_loss_clip": 0.01158927,
      "auxiliary_loss_mlp": 0.01022166,
      "balance_loss_clip": 1.03959048,
      "balance_loss_mlp": 1.01463771,
      "epoch": 0.9873143750375759,
      "flos": 21434217874560.0,
      "grad_norm": 1.8136826416721445,
      "language_loss": 0.76287949,
      "learning_rate": 1.6722506681043913e-09,
      "loss": 0.78469038,
      "num_input_tokens_seen": 177237050,
      "step": 8211,
      "time_per_iteration": 2.7816617488861084
    },
    {
      "auxiliary_loss_clip": 0.01174727,
      "auxiliary_loss_mlp": 0.01026078,
      "balance_loss_clip": 1.04475546,
      "balance_loss_mlp": 1.01892507,
      "epoch": 0.987434617928215,
      "flos": 16326499800960.0,
      "grad_norm": 2.8234128757119508,
      "language_loss": 0.69014162,
      "learning_rate": 1.640554287104745e-09,
      "loss": 0.71214962,
      "num_input_tokens_seen": 177255325,
      "step": 8212,
      "time_per_iteration": 2.868584156036377
    },
    {
      "auxiliary_loss_clip": 0.01175539,
      "auxiliary_loss_mlp": 0.01022462,
      "balance_loss_clip": 1.03933144,
      "balance_loss_mlp": 1.0148747,
      "epoch": 0.9875548608188541,
      "flos": 17851984456320.0,
      "grad_norm": 6.002174860951654,
      "language_loss": 0.80467451,
      "learning_rate": 1.609161055644348e-09,
      "loss": 0.82665455,
      "num_input_tokens_seen": 177271250,
      "step": 8213,
      "time_per_iteration": 2.62898850440979
    },
    {
      "auxiliary_loss_clip": 0.0117663,
      "auxiliary_loss_mlp": 0.01025011,
      "balance_loss_clip": 1.04622459,
      "balance_loss_mlp": 1.01768565,
      "epoch": 0.9876751037094932,
      "flos": 26132876887680.0,
      "grad_norm": 4.012409232381826,
      "language_loss": 0.68369186,
      "learning_rate": 1.5780709784849467e-09,
      "loss": 0.70570832,
      "num_input_tokens_seen": 177288270,
      "step": 8214,
      "time_per_iteration": 2.709951162338257
    },
    {
      "auxiliary_loss_clip": 0.01185968,
      "auxiliary_loss_mlp": 0.01025872,
      "balance_loss_clip": 1.04503465,
      "balance_loss_mlp": 1.01832283,
      "epoch": 0.9877953466001322,
      "flos": 15991344973440.0,
      "grad_norm": 2.135666873441361,
      "language_loss": 0.82594037,
      "learning_rate": 1.5472840603436565e-09,
      "loss": 0.8480587,
      "num_input_tokens_seen": 177305500,
      "step": 8215,
      "time_per_iteration": 2.890859365463257
    },
    {
      "auxiliary_loss_clip": 0.01177555,
      "auxiliary_loss_mlp": 0.01024447,
      "balance_loss_clip": 1.04619229,
      "balance_loss_mlp": 1.01710951,
      "epoch": 0.9879155894907714,
      "flos": 18806777827200.0,
      "grad_norm": 2.1178616460775244,
      "language_loss": 0.78333342,
      "learning_rate": 1.5168003058900757e-09,
      "loss": 0.80535346,
      "num_input_tokens_seen": 177323500,
      "step": 8216,
      "time_per_iteration": 2.6744534969329834
    },
    {
      "auxiliary_loss_clip": 0.01173641,
      "auxiliary_loss_mlp": 0.0102633,
      "balance_loss_clip": 1.04149818,
      "balance_loss_mlp": 1.01905203,
      "epoch": 0.9880358323814105,
      "flos": 22382044007040.0,
      "grad_norm": 1.8989589047108235,
      "language_loss": 0.92316669,
      "learning_rate": 1.4866197197491715e-09,
      "loss": 0.94516635,
      "num_input_tokens_seen": 177342860,
      "step": 8217,
      "time_per_iteration": 2.7799899578094482
    },
    {
      "auxiliary_loss_clip": 0.01174347,
      "auxiliary_loss_mlp": 0.01208422,
      "balance_loss_clip": 1.04639649,
      "balance_loss_mlp": 1.0007081,
      "epoch": 0.9881560752720495,
      "flos": 15668831733120.0,
      "grad_norm": 3.897898820325104,
      "language_loss": 0.78929383,
      "learning_rate": 1.4567423064988371e-09,
      "loss": 0.8131215,
      "num_input_tokens_seen": 177360210,
      "step": 8218,
      "time_per_iteration": 2.7865140438079834
    },
    {
      "auxiliary_loss_clip": 0.01168051,
      "auxiliary_loss_mlp": 0.01024702,
      "balance_loss_clip": 1.04697776,
      "balance_loss_mlp": 1.01693511,
      "epoch": 0.9882763181626887,
      "flos": 21500113374720.0,
      "grad_norm": 2.046526477679493,
      "language_loss": 0.78246528,
      "learning_rate": 1.4271680706718913e-09,
      "loss": 0.80439281,
      "num_input_tokens_seen": 177377885,
      "step": 8219,
      "time_per_iteration": 3.6059868335723877
    },
    {
      "auxiliary_loss_clip": 0.01174274,
      "auxiliary_loss_mlp": 0.01026847,
      "balance_loss_clip": 1.04886794,
      "balance_loss_mlp": 1.01878262,
      "epoch": 0.9883965610533277,
      "flos": 28034598551040.0,
      "grad_norm": 1.6693533743376552,
      "language_loss": 0.82697326,
      "learning_rate": 1.3978970167543013e-09,
      "loss": 0.84898448,
      "num_input_tokens_seen": 177398065,
      "step": 8220,
      "time_per_iteration": 2.6757194995880127
    },
    {
      "auxiliary_loss_clip": 0.0116294,
      "auxiliary_loss_mlp": 0.01027917,
      "balance_loss_clip": 1.04248285,
      "balance_loss_mlp": 1.02019477,
      "epoch": 0.9885168039439668,
      "flos": 14098601710080.0,
      "grad_norm": 2.2498499907656297,
      "language_loss": 0.7762531,
      "learning_rate": 1.3689291491867372e-09,
      "loss": 0.79816163,
      "num_input_tokens_seen": 177416380,
      "step": 8221,
      "time_per_iteration": 2.7735955715179443
    },
    {
      "auxiliary_loss_clip": 0.01167647,
      "auxiliary_loss_mlp": 0.01028722,
      "balance_loss_clip": 1.04666388,
      "balance_loss_mlp": 1.02069342,
      "epoch": 0.988637046834606,
      "flos": 26432013352320.0,
      "grad_norm": 5.417608029060404,
      "language_loss": 0.73885351,
      "learning_rate": 1.3402644723636836e-09,
      "loss": 0.76081717,
      "num_input_tokens_seen": 177438410,
      "step": 8222,
      "time_per_iteration": 2.750457525253296
    },
    {
      "auxiliary_loss_clip": 0.01166522,
      "auxiliary_loss_mlp": 0.01024956,
      "balance_loss_clip": 1.04610658,
      "balance_loss_mlp": 1.01791048,
      "epoch": 0.988757289725245,
      "flos": 25229113764480.0,
      "grad_norm": 2.388348384573656,
      "language_loss": 0.83508366,
      "learning_rate": 1.311902990633218e-09,
      "loss": 0.85699844,
      "num_input_tokens_seen": 177457375,
      "step": 8223,
      "time_per_iteration": 2.6485581398010254
    },
    {
      "auxiliary_loss_clip": 0.01165387,
      "auxiliary_loss_mlp": 0.0102147,
      "balance_loss_clip": 1.0410856,
      "balance_loss_mlp": 1.01476407,
      "epoch": 0.9888775326158841,
      "flos": 26359042872960.0,
      "grad_norm": 1.9273579771838514,
      "language_loss": 0.71329248,
      "learning_rate": 1.2838447082978987e-09,
      "loss": 0.73516107,
      "num_input_tokens_seen": 177478530,
      "step": 8224,
      "time_per_iteration": 2.6347644329071045
    },
    {
      "auxiliary_loss_clip": 0.0116631,
      "auxiliary_loss_mlp": 0.01025428,
      "balance_loss_clip": 1.04483044,
      "balance_loss_mlp": 1.01737547,
      "epoch": 0.9889977755065231,
      "flos": 24316120846080.0,
      "grad_norm": 2.8345256560378425,
      "language_loss": 0.8355763,
      "learning_rate": 1.2560896296143208e-09,
      "loss": 0.8574937,
      "num_input_tokens_seen": 177496995,
      "step": 8225,
      "time_per_iteration": 2.7316031455993652
    },
    {
      "auxiliary_loss_clip": 0.01166575,
      "auxiliary_loss_mlp": 0.01023449,
      "balance_loss_clip": 1.04690075,
      "balance_loss_mlp": 1.01603472,
      "epoch": 0.9891180183971623,
      "flos": 18951066760320.0,
      "grad_norm": 2.1530025435070006,
      "language_loss": 0.81884331,
      "learning_rate": 1.2286377587926722e-09,
      "loss": 0.84074354,
      "num_input_tokens_seen": 177513785,
      "step": 8226,
      "time_per_iteration": 2.611820697784424
    },
    {
      "auxiliary_loss_clip": 0.01164776,
      "auxiliary_loss_mlp": 0.01023285,
      "balance_loss_clip": 1.04569674,
      "balance_loss_mlp": 1.01545286,
      "epoch": 0.9892382612878013,
      "flos": 26176580760960.0,
      "grad_norm": 2.5257489136186497,
      "language_loss": 0.74526298,
      "learning_rate": 1.2014890999973992e-09,
      "loss": 0.76714355,
      "num_input_tokens_seen": 177530705,
      "step": 8227,
      "time_per_iteration": 4.15140700340271
    },
    {
      "auxiliary_loss_clip": 0.01164157,
      "auxiliary_loss_mlp": 0.0102381,
      "balance_loss_clip": 1.0460012,
      "balance_loss_mlp": 1.0167141,
      "epoch": 0.9893585041784404,
      "flos": 25449605400960.0,
      "grad_norm": 2.824098068364295,
      "language_loss": 0.78402603,
      "learning_rate": 1.1746436573472073e-09,
      "loss": 0.8059057,
      "num_input_tokens_seen": 177552440,
      "step": 8228,
      "time_per_iteration": 2.6982271671295166
    },
    {
      "auxiliary_loss_clip": 0.01180307,
      "auxiliary_loss_mlp": 0.01023318,
      "balance_loss_clip": 1.04525042,
      "balance_loss_mlp": 1.01609969,
      "epoch": 0.9894787470690796,
      "flos": 20189302352640.0,
      "grad_norm": 1.9460864599235317,
      "language_loss": 0.69297516,
      "learning_rate": 1.1481014349141726e-09,
      "loss": 0.71501148,
      "num_input_tokens_seen": 177569660,
      "step": 8229,
      "time_per_iteration": 2.64520525932312
    },
    {
      "auxiliary_loss_clip": 0.01171969,
      "auxiliary_loss_mlp": 0.01025392,
      "balance_loss_clip": 1.04429412,
      "balance_loss_mlp": 1.01724386,
      "epoch": 0.9895989899597186,
      "flos": 24644308435200.0,
      "grad_norm": 2.8766492764893505,
      "language_loss": 0.84422469,
      "learning_rate": 1.121862436724852e-09,
      "loss": 0.8661983,
      "num_input_tokens_seen": 177588500,
      "step": 8230,
      "time_per_iteration": 3.6336159706115723
    },
    {
      "auxiliary_loss_clip": 0.01171712,
      "auxiliary_loss_mlp": 0.01025633,
      "balance_loss_clip": 1.04902315,
      "balance_loss_mlp": 1.01815295,
      "epoch": 0.9897192328503577,
      "flos": 21799034357760.0,
      "grad_norm": 1.5795340811995129,
      "language_loss": 0.70798057,
      "learning_rate": 1.0959266667598388e-09,
      "loss": 0.729954,
      "num_input_tokens_seen": 177607315,
      "step": 8231,
      "time_per_iteration": 2.7279715538024902
    },
    {
      "auxiliary_loss_clip": 0.01178809,
      "auxiliary_loss_mlp": 0.01028601,
      "balance_loss_clip": 1.04558659,
      "balance_loss_mlp": 1.02061439,
      "epoch": 0.9898394757409968,
      "flos": 21325229032320.0,
      "grad_norm": 1.9376660716567438,
      "language_loss": 0.74789196,
      "learning_rate": 1.0702941289533196e-09,
      "loss": 0.76996613,
      "num_input_tokens_seen": 177625990,
      "step": 8232,
      "time_per_iteration": 2.7803194522857666
    },
    {
      "auxiliary_loss_clip": 0.01172501,
      "auxiliary_loss_mlp": 0.010228,
      "balance_loss_clip": 1.04354322,
      "balance_loss_mlp": 1.01633334,
      "epoch": 0.9899597186316359,
      "flos": 18545024442240.0,
      "grad_norm": 2.1535455400151045,
      "language_loss": 0.88965034,
      "learning_rate": 1.0449648271939615e-09,
      "loss": 0.91160333,
      "num_input_tokens_seen": 177642335,
      "step": 8233,
      "time_per_iteration": 3.652824878692627
    },
    {
      "auxiliary_loss_clip": 0.01181088,
      "auxiliary_loss_mlp": 0.01207816,
      "balance_loss_clip": 1.0439868,
      "balance_loss_mlp": 1.00073814,
      "epoch": 0.990079961522275,
      "flos": 23766723348480.0,
      "grad_norm": 1.647467383813174,
      "language_loss": 0.72686684,
      "learning_rate": 1.0199387653240243e-09,
      "loss": 0.75075585,
      "num_input_tokens_seen": 177662025,
      "step": 8234,
      "time_per_iteration": 2.803091049194336
    },
    {
      "auxiliary_loss_clip": 0.01166364,
      "auxiliary_loss_mlp": 0.01020297,
      "balance_loss_clip": 1.04541421,
      "balance_loss_mlp": 1.01322842,
      "epoch": 0.9902002044129141,
      "flos": 16399182971520.0,
      "grad_norm": 1.5840347339430165,
      "language_loss": 0.70624858,
      "learning_rate": 9.952159471400267e-10,
      "loss": 0.7281152,
      "num_input_tokens_seen": 177679065,
      "step": 8235,
      "time_per_iteration": 2.7535266876220703
    },
    {
      "auxiliary_loss_clip": 0.0117052,
      "auxiliary_loss_mlp": 0.01207911,
      "balance_loss_clip": 1.04505575,
      "balance_loss_mlp": 1.00079155,
      "epoch": 0.9903204473035532,
      "flos": 22559657783040.0,
      "grad_norm": 1.8670425112046785,
      "language_loss": 0.84184086,
      "learning_rate": 9.707963763923022e-10,
      "loss": 0.86562514,
      "num_input_tokens_seen": 177698115,
      "step": 8236,
      "time_per_iteration": 2.6859524250030518
    },
    {
      "auxiliary_loss_clip": 0.01167909,
      "auxiliary_loss_mlp": 0.01023461,
      "balance_loss_clip": 1.04169655,
      "balance_loss_mlp": 1.01600134,
      "epoch": 0.9904406901941922,
      "flos": 16144001775360.0,
      "grad_norm": 2.2657223334489944,
      "language_loss": 0.7914508,
      "learning_rate": 9.466800567854427e-10,
      "loss": 0.81336451,
      "num_input_tokens_seen": 177716715,
      "step": 8237,
      "time_per_iteration": 2.755129337310791
    },
    {
      "auxiliary_loss_clip": 0.01176203,
      "auxiliary_loss_mlp": 0.01023435,
      "balance_loss_clip": 1.04022324,
      "balance_loss_mlp": 1.01526356,
      "epoch": 0.9905609330848314,
      "flos": 26651499408000.0,
      "grad_norm": 1.962486151653852,
      "language_loss": 0.68230474,
      "learning_rate": 9.228669919778553e-10,
      "loss": 0.70430112,
      "num_input_tokens_seen": 177735640,
      "step": 8238,
      "time_per_iteration": 2.775313138961792
    },
    {
      "auxiliary_loss_clip": 0.01166764,
      "auxiliary_loss_mlp": 0.01025484,
      "balance_loss_clip": 1.04204929,
      "balance_loss_mlp": 1.01731849,
      "epoch": 0.9906811759754705,
      "flos": 23111820627840.0,
      "grad_norm": 2.297497522507782,
      "language_loss": 0.79539824,
      "learning_rate": 8.993571855817617e-10,
      "loss": 0.8173207,
      "num_input_tokens_seen": 177754470,
      "step": 8239,
      "time_per_iteration": 2.6953752040863037
    },
    {
      "auxiliary_loss_clip": 0.01169507,
      "auxiliary_loss_mlp": 0.01022187,
      "balance_loss_clip": 1.0454278,
      "balance_loss_mlp": 1.01482296,
      "epoch": 0.9908014188661095,
      "flos": 22090593052800.0,
      "grad_norm": 2.4293907670045196,
      "language_loss": 0.74761093,
      "learning_rate": 8.761506411638642e-10,
      "loss": 0.76952785,
      "num_input_tokens_seen": 177773935,
      "step": 8240,
      "time_per_iteration": 2.66082763671875
    },
    {
      "auxiliary_loss_clip": 0.01168609,
      "auxiliary_loss_mlp": 0.01028173,
      "balance_loss_clip": 1.04428172,
      "balance_loss_mlp": 1.02085972,
      "epoch": 0.9909216617567487,
      "flos": 19242948677760.0,
      "grad_norm": 1.7813701688879207,
      "language_loss": 0.74195039,
      "learning_rate": 8.53247362244236e-10,
      "loss": 0.76391822,
      "num_input_tokens_seen": 177792745,
      "step": 8241,
      "time_per_iteration": 2.7180283069610596
    },
    {
      "auxiliary_loss_clip": 0.01172565,
      "auxiliary_loss_mlp": 0.01022745,
      "balance_loss_clip": 1.0450362,
      "balance_loss_mlp": 1.01550949,
      "epoch": 0.9910419046473877,
      "flos": 23621213352960.0,
      "grad_norm": 1.8561489699297569,
      "language_loss": 0.68362767,
      "learning_rate": 8.306473522976532e-10,
      "loss": 0.70558071,
      "num_input_tokens_seen": 177812150,
      "step": 8242,
      "time_per_iteration": 2.7667510509490967
    },
    {
      "auxiliary_loss_clip": 0.01166386,
      "auxiliary_loss_mlp": 0.01020185,
      "balance_loss_clip": 1.04804063,
      "balance_loss_mlp": 1.01240063,
      "epoch": 0.9911621475380268,
      "flos": 22711380831360.0,
      "grad_norm": 2.845960982602493,
      "language_loss": 0.71924114,
      "learning_rate": 8.083506147522623e-10,
      "loss": 0.74110687,
      "num_input_tokens_seen": 177831545,
      "step": 8243,
      "time_per_iteration": 2.628544569015503
    },
    {
      "auxiliary_loss_clip": 0.01162668,
      "auxiliary_loss_mlp": 0.01028756,
      "balance_loss_clip": 1.04358113,
      "balance_loss_mlp": 1.02122211,
      "epoch": 0.991282390428666,
      "flos": 13516956777600.0,
      "grad_norm": 1.9388635665311502,
      "language_loss": 0.85259891,
      "learning_rate": 7.863571529906909e-10,
      "loss": 0.87451315,
      "num_input_tokens_seen": 177847130,
      "step": 8244,
      "time_per_iteration": 2.74871826171875
    },
    {
      "auxiliary_loss_clip": 0.01070847,
      "auxiliary_loss_mlp": 0.0100103,
      "balance_loss_clip": 1.01047778,
      "balance_loss_mlp": 1.00028479,
      "epoch": 0.991402633319305,
      "flos": 61830492071040.0,
      "grad_norm": 0.7218507034648154,
      "language_loss": 0.59641826,
      "learning_rate": 7.646669703489372e-10,
      "loss": 0.61713701,
      "num_input_tokens_seen": 177911440,
      "step": 8245,
      "time_per_iteration": 3.289583444595337
    },
    {
      "auxiliary_loss_clip": 0.01190093,
      "auxiliary_loss_mlp": 0.01022643,
      "balance_loss_clip": 1.03399825,
      "balance_loss_mlp": 1.01555324,
      "epoch": 0.9915228762099441,
      "flos": 18770148933120.0,
      "grad_norm": 2.0935962868987557,
      "language_loss": 0.57685882,
      "learning_rate": 7.432800701177023e-10,
      "loss": 0.59898615,
      "num_input_tokens_seen": 177929440,
      "step": 8246,
      "time_per_iteration": 3.820720672607422
    },
    {
      "auxiliary_loss_clip": 0.010777,
      "auxiliary_loss_mlp": 0.01001915,
      "balance_loss_clip": 1.01088047,
      "balance_loss_mlp": 1.00109291,
      "epoch": 0.9916431191005832,
      "flos": 65936660244480.0,
      "grad_norm": 0.9702936870330691,
      "language_loss": 0.5778085,
      "learning_rate": 7.221964555415017e-10,
      "loss": 0.59860468,
      "num_input_tokens_seen": 177989100,
      "step": 8247,
      "time_per_iteration": 3.4267568588256836
    },
    {
      "auxiliary_loss_clip": 0.01167371,
      "auxiliary_loss_mlp": 0.0102051,
      "balance_loss_clip": 1.04380214,
      "balance_loss_mlp": 1.01397133,
      "epoch": 0.9917633619912223,
      "flos": 16581573256320.0,
      "grad_norm": 2.328313802312335,
      "language_loss": 0.75060987,
      "learning_rate": 7.01416129818222e-10,
      "loss": 0.77248871,
      "num_input_tokens_seen": 178006720,
      "step": 8248,
      "time_per_iteration": 2.6801116466522217
    },
    {
      "auxiliary_loss_clip": 0.01180385,
      "auxiliary_loss_mlp": 0.01023002,
      "balance_loss_clip": 1.04373288,
      "balance_loss_mlp": 1.01589739,
      "epoch": 0.9918836048818613,
      "flos": 25411108999680.0,
      "grad_norm": 2.0985566240102127,
      "language_loss": 0.5865494,
      "learning_rate": 6.809390961006745e-10,
      "loss": 0.60858327,
      "num_input_tokens_seen": 178026850,
      "step": 8249,
      "time_per_iteration": 2.7277591228485107
    },
    {
      "auxiliary_loss_clip": 0.01168914,
      "auxiliary_loss_mlp": 0.0102613,
      "balance_loss_clip": 1.04412878,
      "balance_loss_mlp": 1.01897168,
      "epoch": 0.9920038477725005,
      "flos": 25046867134080.0,
      "grad_norm": 2.289219266660533,
      "language_loss": 0.68430066,
      "learning_rate": 6.607653574948191e-10,
      "loss": 0.70625108,
      "num_input_tokens_seen": 178047630,
      "step": 8250,
      "time_per_iteration": 2.7079575061798096
    },
    {
      "auxiliary_loss_clip": 0.01159857,
      "auxiliary_loss_mlp": 0.01022966,
      "balance_loss_clip": 1.04216337,
      "balance_loss_mlp": 1.01615667,
      "epoch": 0.9921240906631396,
      "flos": 21829773421440.0,
      "grad_norm": 1.9102781835688127,
      "language_loss": 0.81573623,
      "learning_rate": 6.408949170613187e-10,
      "loss": 0.83756447,
      "num_input_tokens_seen": 178066895,
      "step": 8251,
      "time_per_iteration": 2.6713168621063232
    },
    {
      "auxiliary_loss_clip": 0.01170421,
      "auxiliary_loss_mlp": 0.01024853,
      "balance_loss_clip": 1.04391658,
      "balance_loss_mlp": 1.01665092,
      "epoch": 0.9922443335537786,
      "flos": 24864225454080.0,
      "grad_norm": 4.602516481426818,
      "language_loss": 0.81919634,
      "learning_rate": 6.213277778144288e-10,
      "loss": 0.84114909,
      "num_input_tokens_seen": 178088540,
      "step": 8252,
      "time_per_iteration": 2.7065014839172363
    },
    {
      "auxiliary_loss_clip": 0.01181787,
      "auxiliary_loss_mlp": 0.01026126,
      "balance_loss_clip": 1.03863883,
      "balance_loss_mlp": 1.01823473,
      "epoch": 0.9923645764444178,
      "flos": 21613088626560.0,
      "grad_norm": 2.351523840867709,
      "language_loss": 0.67043513,
      "learning_rate": 6.020639427224416e-10,
      "loss": 0.6925143,
      "num_input_tokens_seen": 178106185,
      "step": 8253,
      "time_per_iteration": 3.6776821613311768
    },
    {
      "auxiliary_loss_clip": 0.01170879,
      "auxiliary_loss_mlp": 0.01021716,
      "balance_loss_clip": 1.04539728,
      "balance_loss_mlp": 1.0143311,
      "epoch": 0.9924848193350568,
      "flos": 25001798544000.0,
      "grad_norm": 2.314313642744238,
      "language_loss": 0.72627914,
      "learning_rate": 5.831034147076864e-10,
      "loss": 0.74820513,
      "num_input_tokens_seen": 178123435,
      "step": 8254,
      "time_per_iteration": 2.7676291465759277
    },
    {
      "auxiliary_loss_clip": 0.0106767,
      "auxiliary_loss_mlp": 0.01000464,
      "balance_loss_clip": 1.01100492,
      "balance_loss_mlp": 0.99970382,
      "epoch": 0.9926050622256959,
      "flos": 68912543151360.0,
      "grad_norm": 0.6852228030032983,
      "language_loss": 0.55630374,
      "learning_rate": 5.644461966463065e-10,
      "loss": 0.57698512,
      "num_input_tokens_seen": 178191045,
      "step": 8255,
      "time_per_iteration": 3.2881295680999756
    },
    {
      "auxiliary_loss_clip": 0.01171905,
      "auxiliary_loss_mlp": 0.01021859,
      "balance_loss_clip": 1.0457412,
      "balance_loss_mlp": 1.01511168,
      "epoch": 0.9927253051163349,
      "flos": 20923675914240.0,
      "grad_norm": 1.855629513178068,
      "language_loss": 0.76032591,
      "learning_rate": 5.460922913687049e-10,
      "loss": 0.78226352,
      "num_input_tokens_seen": 178210135,
      "step": 8256,
      "time_per_iteration": 2.634409189224243
    },
    {
      "auxiliary_loss_clip": 0.01175267,
      "auxiliary_loss_mlp": 0.01208278,
      "balance_loss_clip": 1.03916669,
      "balance_loss_mlp": 1.00077748,
      "epoch": 0.9928455480069741,
      "flos": 22308211601280.0,
      "grad_norm": 1.9630036661807417,
      "language_loss": 0.7539314,
      "learning_rate": 5.280417016593208e-10,
      "loss": 0.77776682,
      "num_input_tokens_seen": 178229925,
      "step": 8257,
      "time_per_iteration": 3.6991047859191895
    },
    {
      "auxiliary_loss_clip": 0.01167069,
      "auxiliary_loss_mlp": 0.01206879,
      "balance_loss_clip": 1.04783797,
      "balance_loss_mlp": 1.0006808,
      "epoch": 0.9929657908976132,
      "flos": 17383889393280.0,
      "grad_norm": 1.6703327997158635,
      "language_loss": 0.7483449,
      "learning_rate": 5.102944302559642e-10,
      "loss": 0.77208436,
      "num_input_tokens_seen": 178247420,
      "step": 8258,
      "time_per_iteration": 2.561074733734131
    },
    {
      "auxiliary_loss_clip": 0.01188437,
      "auxiliary_loss_mlp": 0.01023573,
      "balance_loss_clip": 1.04268551,
      "balance_loss_mlp": 1.015854,
      "epoch": 0.9930860337882522,
      "flos": 22674680110080.0,
      "grad_norm": 1.9411474411781098,
      "language_loss": 0.79745418,
      "learning_rate": 4.9285047985137e-10,
      "loss": 0.81957424,
      "num_input_tokens_seen": 178266840,
      "step": 8259,
      "time_per_iteration": 3.6038856506347656
    },
    {
      "auxiliary_loss_clip": 0.01172107,
      "auxiliary_loss_mlp": 0.01029111,
      "balance_loss_clip": 1.04595447,
      "balance_loss_mlp": 1.02180386,
      "epoch": 0.9932062766788914,
      "flos": 28147789284480.0,
      "grad_norm": 1.7097637926609746,
      "language_loss": 0.74522942,
      "learning_rate": 4.757098530916436e-10,
      "loss": 0.7672416,
      "num_input_tokens_seen": 178287285,
      "step": 8260,
      "time_per_iteration": 2.680682897567749
    },
    {
      "auxiliary_loss_clip": 0.01175204,
      "auxiliary_loss_mlp": 0.01026827,
      "balance_loss_clip": 1.0491066,
      "balance_loss_mlp": 1.01941848,
      "epoch": 0.9933265195695304,
      "flos": 20156659868160.0,
      "grad_norm": 2.589609147587294,
      "language_loss": 0.77700168,
      "learning_rate": 4.5887255257670563e-10,
      "loss": 0.79902202,
      "num_input_tokens_seen": 178304325,
      "step": 8261,
      "time_per_iteration": 2.585681915283203
    },
    {
      "auxiliary_loss_clip": 0.01167024,
      "auxiliary_loss_mlp": 0.01027858,
      "balance_loss_clip": 1.04687905,
      "balance_loss_mlp": 1.02021027,
      "epoch": 0.9934467624601695,
      "flos": 21362037494400.0,
      "grad_norm": 2.223903190688392,
      "language_loss": 0.77436519,
      "learning_rate": 4.4233858086117906e-10,
      "loss": 0.796314,
      "num_input_tokens_seen": 178322850,
      "step": 8262,
      "time_per_iteration": 2.5896031856536865
    },
    {
      "auxiliary_loss_clip": 0.01177796,
      "auxiliary_loss_mlp": 0.01027954,
      "balance_loss_clip": 1.04593182,
      "balance_loss_mlp": 1.02024114,
      "epoch": 0.9935670053508087,
      "flos": 19756040503680.0,
      "grad_norm": 2.179020620883396,
      "language_loss": 0.67663187,
      "learning_rate": 4.261079404528356e-10,
      "loss": 0.69868934,
      "num_input_tokens_seen": 178342330,
      "step": 8263,
      "time_per_iteration": 2.625257968902588
    },
    {
      "auxiliary_loss_clip": 0.01167642,
      "auxiliary_loss_mlp": 0.01022433,
      "balance_loss_clip": 1.04540038,
      "balance_loss_mlp": 1.01485443,
      "epoch": 0.9936872482414477,
      "flos": 21978838863360.0,
      "grad_norm": 5.292749736858594,
      "language_loss": 0.69054455,
      "learning_rate": 4.1018063381437205e-10,
      "loss": 0.71244532,
      "num_input_tokens_seen": 178362715,
      "step": 8264,
      "time_per_iteration": 2.71109938621521
    },
    {
      "auxiliary_loss_clip": 0.01064009,
      "auxiliary_loss_mlp": 0.01000988,
      "balance_loss_clip": 1.01272583,
      "balance_loss_mlp": 1.00033784,
      "epoch": 0.9938074911320868,
      "flos": 69810667839360.0,
      "grad_norm": 0.8777943795548492,
      "language_loss": 0.61057544,
      "learning_rate": 3.9455666336141167e-10,
      "loss": 0.63122541,
      "num_input_tokens_seen": 178426495,
      "step": 8265,
      "time_per_iteration": 3.2272274494171143
    },
    {
      "auxiliary_loss_clip": 0.01164925,
      "auxiliary_loss_mlp": 0.01022825,
      "balance_loss_clip": 1.04649186,
      "balance_loss_mlp": 1.01574993,
      "epoch": 0.9939277340227259,
      "flos": 15084170058240.0,
      "grad_norm": 2.65993834233241,
      "language_loss": 0.83763099,
      "learning_rate": 3.7923603146450267e-10,
      "loss": 0.85950851,
      "num_input_tokens_seen": 178442555,
      "step": 8266,
      "time_per_iteration": 2.5826268196105957
    },
    {
      "auxiliary_loss_clip": 0.0117459,
      "auxiliary_loss_mlp": 0.01026773,
      "balance_loss_clip": 1.04066801,
      "balance_loss_mlp": 1.0193696,
      "epoch": 0.994047976913365,
      "flos": 17712364291200.0,
      "grad_norm": 1.8332841526174104,
      "language_loss": 0.80966628,
      "learning_rate": 3.642187404473418e-10,
      "loss": 0.83167994,
      "num_input_tokens_seen": 178460715,
      "step": 8267,
      "time_per_iteration": 2.634608268737793
    },
    {
      "auxiliary_loss_clip": 0.01167902,
      "auxiliary_loss_mlp": 0.01020158,
      "balance_loss_clip": 1.04372704,
      "balance_loss_mlp": 1.01287425,
      "epoch": 0.994168219804004,
      "flos": 19171558396800.0,
      "grad_norm": 2.0737275706124834,
      "language_loss": 0.86099911,
      "learning_rate": 3.495047925885508e-10,
      "loss": 0.88287967,
      "num_input_tokens_seen": 178479050,
      "step": 8268,
      "time_per_iteration": 2.613227605819702
    },
    {
      "auxiliary_loss_clip": 0.01168756,
      "auxiliary_loss_mlp": 0.01025795,
      "balance_loss_clip": 1.04192531,
      "balance_loss_mlp": 1.01834142,
      "epoch": 0.9942884626946432,
      "flos": 17851589406720.0,
      "grad_norm": 2.0325934008168445,
      "language_loss": 0.83031189,
      "learning_rate": 3.350941901199e-10,
      "loss": 0.85225743,
      "num_input_tokens_seen": 178495970,
      "step": 8269,
      "time_per_iteration": 2.601621150970459
    },
    {
      "auxiliary_loss_clip": 0.01176053,
      "auxiliary_loss_mlp": 0.01023143,
      "balance_loss_clip": 1.04531574,
      "balance_loss_mlp": 1.01590991,
      "epoch": 0.9944087055852823,
      "flos": 18796578364800.0,
      "grad_norm": 2.9991293202059888,
      "language_loss": 0.83645916,
      "learning_rate": 3.2098693522764066e-10,
      "loss": 0.85845113,
      "num_input_tokens_seen": 178509170,
      "step": 8270,
      "time_per_iteration": 2.6297519207000732
    },
    {
      "auxiliary_loss_clip": 0.011773,
      "auxiliary_loss_mlp": 0.01207574,
      "balance_loss_clip": 1.04524708,
      "balance_loss_mlp": 1.00067222,
      "epoch": 0.9945289484759213,
      "flos": 20996969616000.0,
      "grad_norm": 2.432676102908766,
      "language_loss": 0.81147522,
      "learning_rate": 3.071830300516165e-10,
      "loss": 0.83532405,
      "num_input_tokens_seen": 178527000,
      "step": 8271,
      "time_per_iteration": 2.5711557865142822
    },
    {
      "auxiliary_loss_clip": 0.01175976,
      "auxiliary_loss_mlp": 0.01027943,
      "balance_loss_clip": 1.04722345,
      "balance_loss_mlp": 1.02017009,
      "epoch": 0.9946491913665605,
      "flos": 14756952136320.0,
      "grad_norm": 2.0513126477490706,
      "language_loss": 0.70940983,
      "learning_rate": 2.9368247668615234e-10,
      "loss": 0.73144907,
      "num_input_tokens_seen": 178545590,
      "step": 8272,
      "time_per_iteration": 3.5090246200561523
    },
    {
      "auxiliary_loss_clip": 0.01171806,
      "auxiliary_loss_mlp": 0.01028,
      "balance_loss_clip": 1.04900157,
      "balance_loss_mlp": 1.02018571,
      "epoch": 0.9947694342571995,
      "flos": 12669931186560.0,
      "grad_norm": 3.841358305118559,
      "language_loss": 0.61490053,
      "learning_rate": 2.804852771789434e-10,
      "loss": 0.63689864,
      "num_input_tokens_seen": 178558890,
      "step": 8273,
      "time_per_iteration": 2.6025550365448
    },
    {
      "auxiliary_loss_clip": 0.01163475,
      "auxiliary_loss_mlp": 0.01024326,
      "balance_loss_clip": 1.04578424,
      "balance_loss_mlp": 1.0171144,
      "epoch": 0.9948896771478386,
      "flos": 18843442634880.0,
      "grad_norm": 1.8889844343200228,
      "language_loss": 0.5598824,
      "learning_rate": 2.675914335321661e-10,
      "loss": 0.58176035,
      "num_input_tokens_seen": 178577645,
      "step": 8274,
      "time_per_iteration": 2.506793737411499
    },
    {
      "auxiliary_loss_clip": 0.01176052,
      "auxiliary_loss_mlp": 0.01024664,
      "balance_loss_clip": 1.0471046,
      "balance_loss_mlp": 1.0164566,
      "epoch": 0.9950099200384778,
      "flos": 24900207903360.0,
      "grad_norm": 2.3160447176733876,
      "language_loss": 0.79576814,
      "learning_rate": 2.550009477018111e-10,
      "loss": 0.81777531,
      "num_input_tokens_seen": 178596415,
      "step": 8275,
      "time_per_iteration": 2.6002633571624756
    },
    {
      "auxiliary_loss_clip": 0.01172775,
      "auxiliary_loss_mlp": 0.0120777,
      "balance_loss_clip": 1.0463537,
      "balance_loss_mlp": 1.00067043,
      "epoch": 0.9951301629291168,
      "flos": 23733613987200.0,
      "grad_norm": 2.0781292689504696,
      "language_loss": 0.63400453,
      "learning_rate": 2.4271382159790634e-10,
      "loss": 0.65781003,
      "num_input_tokens_seen": 178613845,
      "step": 8276,
      "time_per_iteration": 2.612318992614746
    },
    {
      "auxiliary_loss_clip": 0.01192915,
      "auxiliary_loss_mlp": 0.01029903,
      "balance_loss_clip": 1.04242158,
      "balance_loss_mlp": 1.02264953,
      "epoch": 0.9952504058197559,
      "flos": 22236893147520.0,
      "grad_norm": 2.318096696524705,
      "language_loss": 0.86020178,
      "learning_rate": 2.3073005708429406e-10,
      "loss": 0.88242996,
      "num_input_tokens_seen": 178633490,
      "step": 8277,
      "time_per_iteration": 2.768944025039673
    },
    {
      "auxiliary_loss_clip": 0.01173039,
      "auxiliary_loss_mlp": 0.01019852,
      "balance_loss_clip": 1.04606116,
      "balance_loss_mlp": 1.01316452,
      "epoch": 0.995370648710395,
      "flos": 21211032718080.0,
      "grad_norm": 2.6758771605816465,
      "language_loss": 0.72459674,
      "learning_rate": 2.190496559788535e-10,
      "loss": 0.74652565,
      "num_input_tokens_seen": 178651775,
      "step": 8278,
      "time_per_iteration": 2.6190192699432373
    },
    {
      "auxiliary_loss_clip": 0.01169474,
      "auxiliary_loss_mlp": 0.01025235,
      "balance_loss_clip": 1.04346192,
      "balance_loss_mlp": 1.0180583,
      "epoch": 0.9954908916010341,
      "flos": 14866731077760.0,
      "grad_norm": 2.525156686341422,
      "language_loss": 0.76730144,
      "learning_rate": 2.0767262005372265e-10,
      "loss": 0.78924859,
      "num_input_tokens_seen": 178669290,
      "step": 8279,
      "time_per_iteration": 2.635275363922119
    },
    {
      "auxiliary_loss_clip": 0.01180967,
      "auxiliary_loss_mlp": 0.01022802,
      "balance_loss_clip": 1.04324102,
      "balance_loss_mlp": 1.01611495,
      "epoch": 0.9956111344916732,
      "flos": 19208259118080.0,
      "grad_norm": 2.047815560623572,
      "language_loss": 0.75343692,
      "learning_rate": 1.965989510346322e-10,
      "loss": 0.77547467,
      "num_input_tokens_seen": 178688410,
      "step": 8280,
      "time_per_iteration": 3.675912380218506
    },
    {
      "auxiliary_loss_clip": 0.01171379,
      "auxiliary_loss_mlp": 0.0102618,
      "balance_loss_clip": 1.04097903,
      "balance_loss_mlp": 1.01791871,
      "epoch": 0.9957313773823123,
      "flos": 20047060494720.0,
      "grad_norm": 2.2023637417142985,
      "language_loss": 0.71168447,
      "learning_rate": 1.8582865060134955e-10,
      "loss": 0.73366004,
      "num_input_tokens_seen": 178706600,
      "step": 8281,
      "time_per_iteration": 2.614935874938965
    },
    {
      "auxiliary_loss_clip": 0.01062696,
      "auxiliary_loss_mlp": 0.01001307,
      "balance_loss_clip": 1.00998449,
      "balance_loss_mlp": 1.00057065,
      "epoch": 0.9958516202729514,
      "flos": 57483253768320.0,
      "grad_norm": 0.7783957971663075,
      "language_loss": 0.55731142,
      "learning_rate": 1.7536172038790098e-10,
      "loss": 0.57795149,
      "num_input_tokens_seen": 178766910,
      "step": 8282,
      "time_per_iteration": 3.1782066822052
    },
    {
      "auxiliary_loss_clip": 0.01171439,
      "auxiliary_loss_mlp": 0.01020643,
      "balance_loss_clip": 1.04599929,
      "balance_loss_mlp": 1.01347852,
      "epoch": 0.9959718631635904,
      "flos": 27782900974080.0,
      "grad_norm": 2.533634474158853,
      "language_loss": 0.69473624,
      "learning_rate": 1.651981619819054e-10,
      "loss": 0.71665704,
      "num_input_tokens_seen": 178784060,
      "step": 8283,
      "time_per_iteration": 2.636356830596924
    },
    {
      "auxiliary_loss_clip": 0.01181887,
      "auxiliary_loss_mlp": 0.01022353,
      "balance_loss_clip": 1.042961,
      "balance_loss_mlp": 1.01497364,
      "epoch": 0.9960921060542296,
      "flos": 24024095274240.0,
      "grad_norm": 2.6577734625574636,
      "language_loss": 0.71062618,
      "learning_rate": 1.5533797692546257e-10,
      "loss": 0.73266864,
      "num_input_tokens_seen": 178802795,
      "step": 8284,
      "time_per_iteration": 3.612740993499756
    },
    {
      "auxiliary_loss_clip": 0.01165169,
      "auxiliary_loss_mlp": 0.01022241,
      "balance_loss_clip": 1.04385257,
      "balance_loss_mlp": 1.01473713,
      "epoch": 0.9962123489448687,
      "flos": 18697393935360.0,
      "grad_norm": 3.056133121651801,
      "language_loss": 0.84108996,
      "learning_rate": 1.4578116671404296e-10,
      "loss": 0.86296409,
      "num_input_tokens_seen": 178821075,
      "step": 8285,
      "time_per_iteration": 2.5677740573883057
    },
    {
      "auxiliary_loss_clip": 0.01163307,
      "auxiliary_loss_mlp": 0.01027646,
      "balance_loss_clip": 1.04645276,
      "balance_loss_mlp": 1.02020788,
      "epoch": 0.9963325918355077,
      "flos": 20010754823040.0,
      "grad_norm": 2.401547940550743,
      "language_loss": 0.71601468,
      "learning_rate": 1.3652773279759777e-10,
      "loss": 0.73792422,
      "num_input_tokens_seen": 178837725,
      "step": 8286,
      "time_per_iteration": 3.5184051990509033
    },
    {
      "auxiliary_loss_clip": 0.01168897,
      "auxiliary_loss_mlp": 0.01028187,
      "balance_loss_clip": 1.04550338,
      "balance_loss_mlp": 1.02070069,
      "epoch": 0.9964528347261468,
      "flos": 33108488991360.0,
      "grad_norm": 1.679384498130479,
      "language_loss": 0.63266182,
      "learning_rate": 1.2757767657989305e-10,
      "loss": 0.65463257,
      "num_input_tokens_seen": 178861515,
      "step": 8287,
      "time_per_iteration": 2.735142946243286
    },
    {
      "auxiliary_loss_clip": 0.01167861,
      "auxiliary_loss_mlp": 0.01021923,
      "balance_loss_clip": 1.04579222,
      "balance_loss_mlp": 1.014925,
      "epoch": 0.9965730776167859,
      "flos": 23109342589440.0,
      "grad_norm": 1.8774333546726374,
      "language_loss": 0.87191975,
      "learning_rate": 1.1893099941850948e-10,
      "loss": 0.89381754,
      "num_input_tokens_seen": 178880410,
      "step": 8288,
      "time_per_iteration": 2.666285753250122
    },
    {
      "auxiliary_loss_clip": 0.01174054,
      "auxiliary_loss_mlp": 0.01025386,
      "balance_loss_clip": 1.04165578,
      "balance_loss_mlp": 1.01781058,
      "epoch": 0.996693320507425,
      "flos": 22965843755520.0,
      "grad_norm": 2.766850778542462,
      "language_loss": 0.77639365,
      "learning_rate": 1.105877026252866e-10,
      "loss": 0.79838812,
      "num_input_tokens_seen": 178898740,
      "step": 8289,
      "time_per_iteration": 2.72993540763855
    },
    {
      "auxiliary_loss_clip": 0.01167476,
      "auxiliary_loss_mlp": 0.01026412,
      "balance_loss_clip": 1.04608512,
      "balance_loss_mlp": 1.01797175,
      "epoch": 0.996813563398064,
      "flos": 13222740476160.0,
      "grad_norm": 2.155538727769033,
      "language_loss": 0.72080863,
      "learning_rate": 1.0254778746565663e-10,
      "loss": 0.74274755,
      "num_input_tokens_seen": 178914015,
      "step": 8290,
      "time_per_iteration": 2.530308961868286
    },
    {
      "auxiliary_loss_clip": 0.01172631,
      "auxiliary_loss_mlp": 0.01026907,
      "balance_loss_clip": 1.04435015,
      "balance_loss_mlp": 1.01922131,
      "epoch": 0.9969338062887032,
      "flos": 14647855553280.0,
      "grad_norm": 2.344674597777894,
      "language_loss": 0.73021978,
      "learning_rate": 9.481125515953259e-11,
      "loss": 0.75221521,
      "num_input_tokens_seen": 178932075,
      "step": 8291,
      "time_per_iteration": 2.773843288421631
    },
    {
      "auxiliary_loss_clip": 0.01176731,
      "auxiliary_loss_mlp": 0.01024198,
      "balance_loss_clip": 1.03839576,
      "balance_loss_mlp": 1.0168457,
      "epoch": 0.9970540491793423,
      "flos": 25735741142400.0,
      "grad_norm": 1.639291316217942,
      "language_loss": 0.80193526,
      "learning_rate": 8.737810688064228e-11,
      "loss": 0.82394457,
      "num_input_tokens_seen": 178951910,
      "step": 8292,
      "time_per_iteration": 2.7175192832946777
    },
    {
      "auxiliary_loss_clip": 0.01169442,
      "auxiliary_loss_mlp": 0.01032556,
      "balance_loss_clip": 1.04216635,
      "balance_loss_mlp": 1.02303684,
      "epoch": 0.9971742920699813,
      "flos": 21470236237440.0,
      "grad_norm": 3.7025645482377705,
      "language_loss": 0.79290563,
      "learning_rate": 8.024834375608414e-11,
      "loss": 0.81492561,
      "num_input_tokens_seen": 178970500,
      "step": 8293,
      "time_per_iteration": 2.6907994747161865
    },
    {
      "auxiliary_loss_clip": 0.01062947,
      "auxiliary_loss_mlp": 0.01000434,
      "balance_loss_clip": 1.01031053,
      "balance_loss_mlp": 0.99969769,
      "epoch": 0.9972945349606205,
      "flos": 72211223629440.0,
      "grad_norm": 0.82206251325131,
      "language_loss": 0.62776434,
      "learning_rate": 7.342196686788149e-11,
      "loss": 0.64839816,
      "num_input_tokens_seen": 179023665,
      "step": 8294,
      "time_per_iteration": 3.0442616939544678
    },
    {
      "auxiliary_loss_clip": 0.01167381,
      "auxiliary_loss_mlp": 0.01028422,
      "balance_loss_clip": 1.04617763,
      "balance_loss_mlp": 1.02107894,
      "epoch": 0.9974147778512595,
      "flos": 19678293515520.0,
      "grad_norm": 2.0852363715419204,
      "language_loss": 0.69076467,
      "learning_rate": 6.689897725142834e-11,
      "loss": 0.71272266,
      "num_input_tokens_seen": 179043140,
      "step": 8295,
      "time_per_iteration": 2.578097343444824
    },
    {
      "auxiliary_loss_clip": 0.01173227,
      "auxiliary_loss_mlp": 0.01026558,
      "balance_loss_clip": 1.04354167,
      "balance_loss_mlp": 1.01920867,
      "epoch": 0.9975350207418986,
      "flos": 15960821391360.0,
      "grad_norm": 3.5828715418696624,
      "language_loss": 0.88594043,
      "learning_rate": 6.067937589615545e-11,
      "loss": 0.90793836,
      "num_input_tokens_seen": 179061215,
      "step": 8296,
      "time_per_iteration": 2.646425247192383
    },
    {
      "auxiliary_loss_clip": 0.01079898,
      "auxiliary_loss_mlp": 0.01003301,
      "balance_loss_clip": 1.01163447,
      "balance_loss_mlp": 1.00261259,
      "epoch": 0.9976552636325378,
      "flos": 59961879768960.0,
      "grad_norm": 0.7475189080144082,
      "language_loss": 0.57623476,
      "learning_rate": 5.476316374575241e-11,
      "loss": 0.59706682,
      "num_input_tokens_seen": 179124700,
      "step": 8297,
      "time_per_iteration": 3.151867389678955
    },
    {
      "auxiliary_loss_clip": 0.01169102,
      "auxiliary_loss_mlp": 0.01023071,
      "balance_loss_clip": 1.04760277,
      "balance_loss_mlp": 1.01480937,
      "epoch": 0.9977755065231768,
      "flos": 22487872452480.0,
      "grad_norm": 1.9791050130867747,
      "language_loss": 0.72611064,
      "learning_rate": 4.9150341697723476e-11,
      "loss": 0.74803233,
      "num_input_tokens_seen": 179144590,
      "step": 8298,
      "time_per_iteration": 2.665508270263672
    },
    {
      "auxiliary_loss_clip": 0.01170174,
      "auxiliary_loss_mlp": 0.01030752,
      "balance_loss_clip": 1.04688036,
      "balance_loss_mlp": 1.02297354,
      "epoch": 0.9978957494138159,
      "flos": 26030280666240.0,
      "grad_norm": 1.970809942157869,
      "language_loss": 0.66622865,
      "learning_rate": 4.384091060338768e-11,
      "loss": 0.68823791,
      "num_input_tokens_seen": 179165060,
      "step": 8299,
      "time_per_iteration": 3.585906505584717
    },
    {
      "auxiliary_loss_clip": 0.01166366,
      "auxiliary_loss_mlp": 0.01025573,
      "balance_loss_clip": 1.04549229,
      "balance_loss_mlp": 1.01835513,
      "epoch": 0.998015992304455,
      "flos": 22637835734400.0,
      "grad_norm": 4.310947140251132,
      "language_loss": 0.73566049,
      "learning_rate": 3.883487126810081e-11,
      "loss": 0.7575798,
      "num_input_tokens_seen": 179184320,
      "step": 8300,
      "time_per_iteration": 2.636403799057007
    },
    {
      "auxiliary_loss_clip": 0.01159267,
      "auxiliary_loss_mlp": 0.01024096,
      "balance_loss_clip": 1.04307652,
      "balance_loss_mlp": 1.01640129,
      "epoch": 0.9981362351950941,
      "flos": 18223444955520.0,
      "grad_norm": 3.4472334807633263,
      "language_loss": 0.79425609,
      "learning_rate": 3.41322244516995e-11,
      "loss": 0.81608975,
      "num_input_tokens_seen": 179202265,
      "step": 8301,
      "time_per_iteration": 2.603149652481079
    },
    {
      "auxiliary_loss_clip": 0.01161119,
      "auxiliary_loss_mlp": 0.0102377,
      "balance_loss_clip": 1.03883934,
      "balance_loss_mlp": 1.01647735,
      "epoch": 0.9982564780857331,
      "flos": 33474095573760.0,
      "grad_norm": 4.78318856309416,
      "language_loss": 0.63121045,
      "learning_rate": 2.9732970866946925e-11,
      "loss": 0.6530593,
      "num_input_tokens_seen": 179222145,
      "step": 8302,
      "time_per_iteration": 2.742234706878662
    },
    {
      "auxiliary_loss_clip": 0.0116102,
      "auxiliary_loss_mlp": 0.01024601,
      "balance_loss_clip": 1.03875232,
      "balance_loss_mlp": 1.01675117,
      "epoch": 0.9983767209763723,
      "flos": 15523465392000.0,
      "grad_norm": 2.8496794089573054,
      "language_loss": 0.78045082,
      "learning_rate": 2.563711118175327e-11,
      "loss": 0.80230701,
      "num_input_tokens_seen": 179239030,
      "step": 8303,
      "time_per_iteration": 2.6656994819641113
    },
    {
      "auxiliary_loss_clip": 0.01170659,
      "auxiliary_loss_mlp": 0.01025089,
      "balance_loss_clip": 1.04458714,
      "balance_loss_mlp": 1.01765037,
      "epoch": 0.9984969638670114,
      "flos": 19974377324160.0,
      "grad_norm": 1.9252067272975841,
      "language_loss": 0.83798319,
      "learning_rate": 2.184464601717728e-11,
      "loss": 0.85994065,
      "num_input_tokens_seen": 179257345,
      "step": 8304,
      "time_per_iteration": 2.6169426441192627
    },
    {
      "auxiliary_loss_clip": 0.0117449,
      "auxiliary_loss_mlp": 0.01023688,
      "balance_loss_clip": 1.04888844,
      "balance_loss_mlp": 1.01584959,
      "epoch": 0.9986172067576504,
      "flos": 20375750874240.0,
      "grad_norm": 2.8908702231397223,
      "language_loss": 0.78245682,
      "learning_rate": 1.8355575948758585e-11,
      "loss": 0.80443859,
      "num_input_tokens_seen": 179275330,
      "step": 8305,
      "time_per_iteration": 2.569108247756958
    },
    {
      "auxiliary_loss_clip": 0.0117038,
      "auxiliary_loss_mlp": 0.01026611,
      "balance_loss_clip": 1.04225254,
      "balance_loss_mlp": 1.0187968,
      "epoch": 0.9987374496482896,
      "flos": 23727903724800.0,
      "grad_norm": 2.1355993943839855,
      "language_loss": 0.73846138,
      "learning_rate": 1.5169901505407424e-11,
      "loss": 0.76043123,
      "num_input_tokens_seen": 179292395,
      "step": 8306,
      "time_per_iteration": 3.5614917278289795
    },
    {
      "auxiliary_loss_clip": 0.01169479,
      "auxiliary_loss_mlp": 0.01021086,
      "balance_loss_clip": 1.04442942,
      "balance_loss_mlp": 1.01398993,
      "epoch": 0.9988576925389286,
      "flos": 25044029959680.0,
      "grad_norm": 2.0442583849342086,
      "language_loss": 0.74358922,
      "learning_rate": 1.228762317073695e-11,
      "loss": 0.76549488,
      "num_input_tokens_seen": 179311225,
      "step": 8307,
      "time_per_iteration": 2.695068597793579
    },
    {
      "auxiliary_loss_clip": 0.01169202,
      "auxiliary_loss_mlp": 0.01023248,
      "balance_loss_clip": 1.04374957,
      "balance_loss_mlp": 1.01627684,
      "epoch": 0.9989779354295677,
      "flos": 31285627637760.0,
      "grad_norm": 2.033233682418159,
      "language_loss": 0.79313672,
      "learning_rate": 9.70874138195299e-12,
      "loss": 0.81506121,
      "num_input_tokens_seen": 179333135,
      "step": 8308,
      "time_per_iteration": 2.6745433807373047
    },
    {
      "auxiliary_loss_clip": 0.01167241,
      "auxiliary_loss_mlp": 0.01025102,
      "balance_loss_clip": 1.04621243,
      "balance_loss_mlp": 1.01796722,
      "epoch": 0.9990981783202069,
      "flos": 19573398823680.0,
      "grad_norm": 2.6903965551227436,
      "language_loss": 0.74611521,
      "learning_rate": 7.433256530076093e-12,
      "loss": 0.76803863,
      "num_input_tokens_seen": 179353090,
      "step": 8309,
      "time_per_iteration": 2.575496196746826
    },
    {
      "auxiliary_loss_clip": 0.01180594,
      "auxiliary_loss_mlp": 0.01022449,
      "balance_loss_clip": 1.04062605,
      "balance_loss_mlp": 1.01560354,
      "epoch": 0.9992184212108459,
      "flos": 17199667514880.0,
      "grad_norm": 4.136335633051518,
      "language_loss": 0.75756896,
      "learning_rate": 5.46116896038562e-12,
      "loss": 0.77959937,
      "num_input_tokens_seen": 179367500,
      "step": 8310,
      "time_per_iteration": 2.661226749420166
    },
    {
      "auxiliary_loss_clip": 0.0116543,
      "auxiliary_loss_mlp": 0.01025074,
      "balance_loss_clip": 1.04285181,
      "balance_loss_mlp": 1.01772809,
      "epoch": 0.999338664101485,
      "flos": 46497853681920.0,
      "grad_norm": 2.1111301014200845,
      "language_loss": 0.61859298,
      "learning_rate": 3.792478972197699e-12,
      "loss": 0.64049804,
      "num_input_tokens_seen": 179388085,
      "step": 8311,
      "time_per_iteration": 3.687147378921509
    },
    {
      "auxiliary_loss_clip": 0.01163571,
      "auxiliary_loss_mlp": 0.0101904,
      "balance_loss_clip": 1.04490709,
      "balance_loss_mlp": 1.01170909,
      "epoch": 0.9994589069921241,
      "flos": 15158253859200.0,
      "grad_norm": 4.088849348066388,
      "language_loss": 0.69947159,
      "learning_rate": 2.4271868181990895e-12,
      "loss": 0.72129774,
      "num_input_tokens_seen": 179405250,
      "step": 8312,
      "time_per_iteration": 2.59051513671875
    },
    {
      "auxiliary_loss_clip": 0.01171006,
      "auxiliary_loss_mlp": 0.01023723,
      "balance_loss_clip": 1.0457592,
      "balance_loss_mlp": 1.01597464,
      "epoch": 0.9995791498827632,
      "flos": 12531460256640.0,
      "grad_norm": 2.2353081598985067,
      "language_loss": 0.81403834,
      "learning_rate": 1.3652927060014973e-12,
      "loss": 0.83598566,
      "num_input_tokens_seen": 179420845,
      "step": 8313,
      "time_per_iteration": 3.4556660652160645
    },
    {
      "auxiliary_loss_clip": 0.01176546,
      "auxiliary_loss_mlp": 0.01026434,
      "balance_loss_clip": 1.04296255,
      "balance_loss_mlp": 1.01849461,
      "epoch": 0.9996993927734023,
      "flos": 19245175320960.0,
      "grad_norm": 1.8690901775669018,
      "language_loss": 0.64070362,
      "learning_rate": 6.067967965872612e-13,
      "loss": 0.66273344,
      "num_input_tokens_seen": 179440455,
      "step": 8314,
      "time_per_iteration": 2.641357898712158
    },
    {
      "auxiliary_loss_clip": 0.01171174,
      "auxiliary_loss_mlp": 0.01024225,
      "balance_loss_clip": 1.0428679,
      "balance_loss_mlp": 1.01722097,
      "epoch": 0.9998196356640414,
      "flos": 62952804518400.0,
      "grad_norm": 3.015594374245497,
      "language_loss": 0.76993454,
      "learning_rate": 1.5169920497548615e-13,
      "loss": 0.79188854,
      "num_input_tokens_seen": 179465075,
      "step": 8315,
      "time_per_iteration": 3.0093014240264893
    },
    {
      "auxiliary_loss_clip": 0.01122572,
      "auxiliary_loss_mlp": 0.01015196,
      "balance_loss_clip": 1.0271163,
      "balance_loss_mlp": 1.01085806,
      "epoch": 0.9999398785546805,
      "flos": 50922375073920.0,
      "grad_norm": 1.176932636868241,
      "language_loss": 0.54956806,
      "learning_rate": 0.0,
      "loss": 0.57094574,
      "num_input_tokens_seen": 179513955,
      "step": 8316,
      "time_per_iteration": 3.1308085918426514
    },
    {
      "epoch": 0.9999398785546805,
      "num_input_tokens_seen": 179513955,
      "step": 8316,
      "total_flos": 6.996749092776837e+17,
      "train_loss": 0.7906862020621568,
      "train_runtime": 24693.8029,
      "train_samples_per_second": 13.471,
      "train_steps_per_second": 0.337
    }
  ],
  "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
}